Graham posted some great tips for those attending the conference virtually over at his blog yesterday. To help keep you connected with the conference without actually being there, we’ve added the ScienceOnline2010 Twitter hashtag (#scio10) feed to the NGS sidebar. This way, everytime you visit NGS over the weekend, you can easily see what’s being talked about.

It’s simple to post automatic updates from the ScienceOnline2010 hashtag feed on your WordPress blog. The code below will import and display items using the built-in WordPress function, wp_rss(), which provides WordPress with feed-fetching and feed-parsing functionality. All you need to do is place the following code below (download here) where you want the feed displayed within your theme template file (e.g., sidebar.php for the homepage or sidebarsingle.php for single article pages). You can customize the number of items posted on line 11.

For FriendFeed users, we’ve also added real-time discussion from the ScienceOnline2010 FriendFeed group further down the sidebar.

Are you a twitter user? Tweet this!

Thank you for subscribing by RSS or email. We work hard to make the articles on Next Generation Science engaging and we truly appreciate your interest and readership!

This article was published on Next Generation Science.

Hope had an opportunity to talk with Judge Schonfeld, CEO and Chief Scientist of CureHunter. In part II of their interview below, Hope focuses on the users and uses of CureHunter; Judge discusses the differences between CureHunter and Wolfram|Alpha, and compares search results from CureHunter to novo|seek, GoPubMed and PubMed.

The Interview, part II

One of the things I have noticed in the field of medical search is that some of the companies refer to their products in several different ways. For example, novo|seek calls itself both an information extraction system and a search engine for biomedical literature. How does CureHunter differ from novo|seek? On your site you refer to CureHunter as a discovery engine. How does that differ from an information extraction system and how does CureHunter differ from the much hyped Wolfram|Alpha, which calls itself a computational knowledge engine? And GoPubMed uses the term, “knowledge-based search” and talks of “using a domain ontology as structured background knowledge.” Please help us understand all of these arcane matters.

In general, the newer semantic and Web 2.0 search technologies are moving toward seek, read, and extract models — as opposed to just listing supposedly relevant articles using classical KWIC indexing methods to find the “relevant ones.”

The more powerful ones are moving toward the “answer system goal” of the instrument makers under the assumption that there is really something specific you want to know when you carry out a search.

Obviously now that was one of CureHunter’s original design goals: machine tell me which drug has the highest probability of curing me — based only on the body of the peer-reviewed evidence.

Whether or not an individual engine reaches its goal depends on many design factors and the fundamental capabilities of the engineers and scientists who create the system and the data to which they have access.

Everyone may have access to a major public library, that does not mean everyone has read all the books in that library with intelligence and learned from their readings. The same is true of many medical search engines claiming to read PubMed.

How do they define the goals of the readings — write the engineering specification — in the first place and the problem(s) involved in reaching the particular goal. What are they going to measure to prove system design success: accuracy, precision, types of error, global margin of error, false positive or negative return rates, reference instruments to compare.

All these QA modules were built for CureHunter because we designed it as a virtual instrument that had to be testable. The machine had to generate hypotheses that humans could validate or invalidate. And finally it had to make numerous predictions based on its internal models that could be readily checked both by human experts and canonical texts.

Generally, speaking when you run a Google or medical domain search, you have no idea if the information is all good, all bad, or a giant helping of both — but routine usage usually proves the last case. So, Google, while an incredible tool is not a precision instrument producing replicable, testable results and neither are so called Medical Information engines, even though they have domain advantage over Google.

Right now, I think novo|seek, NextBio, and GoPubMed are all doing a much, much better engineering job than Google or Microsoft or Yahoo on improving functional health search by beginning to relate sub-component ideas that help define the cluster of information that is any particular idea.

I personally find them very similar in their models and methods to each other and less so to ourselves. They are really trying to enrich key bio med concepts by cross-linking in the manner of Wiki a lot of related topical information. For example, if you look up a gene, you might want to know what diseases it is associated with and what does the gene express; so you cross link the expression databases from Entrez. Same with chemical pathways, JMOLS, etc.

Two weeks ago I had a discussion with Professor Dr. Phil Bourne at the UC San Diego Supercomputer Center and Skaggs School of Pharmacy on the possibility of cross linking the protein data bank directly to CureHunter to fine focus on protein-protein interactions leading directly to clinical cures. [Dr. Bourne is the Associate Director of the Protein Data Bank] That’s a natural problem for our technology and on our development glide path, same for gene expression data. Almost three years ago Dr. Anthony Williams, founder of ChemSpider asked us if we would directly tie to his engine for all the right and similar reasons. I hope we get to work with his team shortly as well.

When you get enough of those points selected by the machine correctly — like a fingerprint match — it knows it has found its goal idea to a predictable level of certainty. All the points of the graphs are guiding you there.

I think, generally speaking, CureHunter is at the very bleeding edge of graph theoretical ontology models because we built it based on computational linguistic models founded in a lot of knowledge about many languages and how they work to construct new ideas from their own lexical networks.

I talked to Stephen Wolfram early on, two years ago about Wolfram|Alpha, and was an early tester on it for him. I was a big fan of NKS and still am. I really like the way Stephen thinks about problems and you can see we are in sync on this idea of not just finding valuable information but in making it directly computable.

Mathematica is such a rich product, some days I wish I had nothing to do but play with it. The Wolfram Alpha demo site is a great toy for any scientist … toys that will lead you to powerful new analytic applications and visualizations of complex data.

Many of the CEOs at major software houses at first saw the web as their enemy because boxware and shelfware had to move to online formats and that played havoc with many traditional software engineering considerations: architecture, modeling, delivery, data integrity and licensing revenues. All of which are still a big problem for the high quality information publishers who are losing control of their IP/data assets to the Internet free information and open source forces: Elsevier, BMJ, Wiley, Wolters, Thomson, McGraw Hill.

We don’t ever want to lose high quality, human edited, carefully documented, source controlled scientific publications. The web already has way to much noise to signal ratio on it and it is getting harder to tell the science from the chaff all the time.

Wolfram|Alpha is the real deal. It will be harder to use for most people than other types of search and answer tech because it is essentially mathematics-based. Its NLP is not as good as it needs to be, but again that’s sort of like saying those first PCs were junk. No. They were fantastic because they gave us a new way to think about the world and making that world computable … which, of course, is what we are all doing more of every day as we convert more and more analog and verbose source material to digital form.

So short answers to feature definitions:

Extraction = function were a set of measurable results with question-answering data is taken from or filtered out of the source articles other than simple citation facts or counts of total articles.

Knowledge Engine = algorithms can generalize over results to some degree even if stated somewhat differently and group related notions other than those that are simply keyword or key phrase matches. Knowledge relationships may be defined by canon, ontology or custom algorithms of the provider.

Semantic Web = some ability to resolve synonymous ideas when phrased or spelled differently; may include thesaurus-like filtering; some ability to understand the value of the returned content in terms of closeness of fit to the query intention.

Data Mining = some extraction functions and implies extraction functions enabling precise quantities of desirable substance to be retrieved from the mass of the search (the raw earth text). In one graphic we show CureHunter finding the GPS location of gold ore in random hills, measuring the density of the deposit, extracting the ore, assaying its purity, and turning it into a ring on your finger: The Cure. So how far does the system take you is a key question for data miners. And what is the purity of the mined element of value. How does it assay out in terms of answering real questions or solving real problems.

Discovery = also some data mining capability; implies ability to find data that no one knows is actually there or to connect extracted material that has increased value when related to another subject in the domain but which the user did not query per se. For example, a lot of health info programs list “Experts.” CureHunter lists experts, but only those whose papers were cited because their experiments had achieved a significant clinical outcome; and thus the data in their papers could be cross computed with the data from other papers of a similar expert nature — and they could be life changing when facing a deadly illness. The graphs in CureHunter cross link diseases that share effective drugs to provide greater understanding of utility and mechanisms of biological action. CureHunter discovers unsuspected connections and suspected connections and links them with known good connections that reinforce our knowledge of the whole cluster point.

Let’s try the search term, “warfarin” as an example — please tell us how to search in CureHunter, novo|seek, GoPubMed, PubMed and Wolfram|Alpha for that term and what we could best find using each.

Well, let’s try it:

CureHunter = Canonical definition, and list of numerous chemical, brand and trade names for the drug identifying it in the literature source, MeSh Breakout hierarchy for all chemical forms; list of 590 diseases where it has clinical utility broken out by specific statements of clinical outcome and statements from formal trials where warfarin was used. Related drugs indicated that also work on the target disease and the history and sources for all statements. The interactive network graph showing the nearest neighbor link by 5 degrees of separation with visual weighting for the most tightly clustered functional biological factors.

novo|seek = List of articles to read, not specific extractions; list of related categories of information with some bar graphs showing relative volume of information available , e.g. Genes and Proteins, Signs and Symptoms, Organisms, etc.; related information, however, is not linked to specific source article, disease, treatment, cure or outcome … so utility compared to CureHunter’s is not really comparable, although its good information to have. Links are casual, not computable.

GoPubMed = Articles list fairly good, but not comprehensive. Order by date sort, not clear why selected other than they contain target drug name. Some key statement extractions — not consistent from one search to another: not at all clear as to why the particular extracted statements were chosen or conclusion to draw from them? Term frequency distributions, knowledge base breakout of related information by categories: e.g. Organisms, Psychiatry and Psychology, Proteins, et al. Seems a little bit of a laundry list of where you can also find references to the parent term although some are very valuable depending on what context you are trying to understand vis-à-vis warfarin. Lot of good citation and additional breakout information. Good clean design. I think GoPubMed is a useful resource but it is not clear to me that you can compute anything very valuable from it directly or discover anything in it that you can’t also find somewhere else.

PubMed itself = the mother ship. NIH, NCBI, NLM have worked for years to maintain carefully edited and source controlled data. Their work is critical for CureHunter and all the rest of the neo search technologies. If you sample from this source you know you are drawing from a good well. By constantly upgrading from ASCII to HTML and XML and beyond and driving data and taxonomy standardization, the library directly and indirectly enables new levels of computability. What they aren’t doing is data mining and discovery; that’s still the job of end user specialist scientists and teams like CureHunter.

One of the things you said during our phone conversation that I found quite intriguing came when I asked you who your competitors are and what products are comparable to CureHunter. You replied stoutly that there really aren’t any comparable products. Could you elaborate? What makes CureHunter unique and who in the medical field should take a serious look at it? Chief medical information officers? Registered health information administrators? Would it be of interest to those in the field of nursing informatics?

CureHunter is different from all the others: its fundamental design is as a scientific instrument seeking to make medical knowledge directly computable and consistently quantifiable and testable and predictive.

That isn’t to say we are smarter than all the other really good companies engaged in medical information. There are many informative, useful, richly advanced search and CDSS products in Medical Informatics and Bioinformatics that are enhancing health care treatments and delivery.

There are 10,000+ health information sites on the web, many of which meet the honor code standards for quality and accuracy of the information they dispense.

But if you have been taking the standard drugs for years and not getting well, or someone has told you there’s nothing better for your chronic illness, or God forbid you have been given a diagnosis of terminal disease — you ought to look at CureHunter right now — because it will have gathered and measured all the best drug data for you.

If you are a doctor and want to practice evidence-based medicine and just can’t keep up with data overload, you owe it to yourself and patients to have a subscription to CureHunter for the cost of less than one professional journal per year.

If you are an administrator, payer, provider or insurer or on a formulary committee, there is no more objective or complete body of instant meta-analytic data available to you anywhere than CureHunter. You should have a subscription to our evidence monograph library and a site license. You will save thousands of dollars a year by knowing when a generic is just as good as a branded drug or which branded meds are best or which ones have the best chance of improving patient outcomes. You will add safety, efficacy, and cost control to everything you do.

CureHunter is the only scientific system on the web that can actually compute directly from the evidence the drugs with the highest possibility of making a person well and it does so with one mouse click — not endless user hours of browsing article, after article, after article.

CureHunter is the only information system that can write a source documented report on demand, just in time, and up to date to the hour of all the drugs that cure an illness or show hope and promise of curing it in the future. The only one that computes the relativistic efficacy of all such drugs.

That’s the content in the Patient Physician Summary Report downloadable from the site in the form of a disease-specific monograph on the patient’s illness. Graduate researchers could spend years compiling such a report and analysis by hand — that they can now get in 10 seconds for $24.00. (Ok, that’s with fries and a coke, med students).

CureHunter is certainly and last but not least, the only machine of its kind in the world that can autonomously compute new cures for human disease and has demonstrated that capability in advanced scientific settings such as US National Science Foundation Conferences and Canadian conferences and proprietary laboratories at major pharma. Any medical teaching center or pure research organization within a pharmaceutical organization should have CureHunter available to all their researchers because it can dramatically speed up their achieving new results and extended uses for the molecules and compounds they are studying. Universities that seek to develop and license new meds should definitely review their portfolios now as they may discover with CureHunter significant new clinical applications and revenues for the substances they have already researched. It’s as simple as entering the name of the key agent in the professional pharmaceutical search system.

Thus CureHunter is uniquely functional, up to date, and technologically advanced way beyond other tools that seem similar on the surface because they index articles from Medline. CureHunter reads the articles. Understands them to a significant degree and makes discoveries based on its readings, just like a bright human would. Many doctors have said quite astounded on first seeing the engine in action: “How does it do that?” or as Arthur C. Clarke wrote … any technology significantly advanced will appear like magic.

As we know, the Obama administration is pouring money into the development of healthcare IT but is also stressing the concepts of “meaningful use” and also of “comparative effectiveness.” Can you discuss those concepts and explain why healthcare administrators (who, after all, hire the IT guys, the medical librarians, the informaticians, etc.) would want to look into CureHunter?

If you use CureHunter’s on demand evidence functions to reduce the use of ineffective, over marketed, unsafe and adverse medications in a routine way at the point of patient treatment, IT managers can take their cost center and turn it into a profit center for better care. Nationally we have estimated we could save the health care system $20 billion per year if CureHunter Evidence Checks were standard operating procedure.

A very small amount of engineering pipes CureHunter directly into any major EMR. A great first step for President Obama to take would be to ask Dr. Zeke Emanuel to help us build CureHunter into VA Vista/CPRS [the Department of Veteran’s Affairs electronic health record, Computerized Patient Record System (CPRS)].

VISTA is an excellent software engineering platform and I would love to be able to do that. We could also use VISTA as a national test platform for AHECS and major medical teaching centers supporting health care in poor urban and rural environments — zones where specialists with expert drug knowledge are few and far between and out of cost reach.

During our phone conversation, I was quite intrigued by the way you argued that CureHunter is a sort of uber tool — useful for such varied audiences as frontline physicians (say, oncologists struggling from the many strains of information overload, heavy workloads and tricky, crucial treatment decision-making dilemmas), pharmaceutical scientists working in the drug discovery realm and clinical researchers working on meta-analyses and drug safety and efficacy issues. Could you give us real-world examples of in what sectors CureHunter is being used and what specific kinds of medical people and scientists are using it? For example, your say on your site, “Evidence-Based Medicine is now possible in Real Clinical Time.” Can you elaborate on the phrase “real clinical time?” Is CureHunter any more up to the minute than UpToDate, for example?

Dr. Lou Degennaro former Director of drug discovery Research at Wyeth and now Chief Scientific Officer at the Leukemia and Lymphoma Society has opened a dialogue with us about working in concert with his team on their target agents. We have done in depth research for friends and associated scientists on Alzheimer, Arthritis, Parkinson, Colon Cancer, Ovarian Cancer and MS.

Ten of the world’s major pharma have scientific use licenses for drug discovery. NIH and FDA are both heavy users of the system as are seven National Cancer Institutes around the world. BMJ, Lilly and Pfizer hit the site a lot as do millions of patients from all over the world. About 60% of licensing is now coming from pharma scientists and 40 from doctors and others.

The major contract drug research houses have on going discussions with us specifically for taking target agents they or their clients own and computationally re-targeting them. Automatic new drug discovery is, after all, a holy grail for patients and pharma researchers both.

A broad range of doctors in general practice have bought the Patient-Physician reports for all the diseases they treat in their specific clinics and patients, of course, purchase the ones for the diseases they or their families are facing head on. Wherever possible we do pro bono analyses, but we really are out of time without a grant partner to do much more.

Right now Dr. Rick Deyo, Kaiser Permanente Professor of Evidence-Based Medicine at Oregon Health Sciences University, and Dr. Eric Orwoll, Director of OCTRI: Oregon Clinical and Translational Research Institute, at OHSU have written NIH asking them to fund joint research between themselves and CureHunter. Getting CureHunter itself into a major clinical use trial is a major goal for us next year.

So if you are a patient, doctor, or advanced Ph.D. researcher at a supercomputer center, CureHunter has probably “blown your mind” more than once.

I addressed the just how UpToDate is UpToDate question above … historically, it was updated quarterly and then limited to the findings of its particular experts. CureHunter reads everything and updates its entire database every night to include even pre-press as much as four months prior to publication. And nothing in UpToDate is directly computable.

Evidence-Based Medicine has always meant historically empowering a panel of human experts to review the state of knowledge in a field, make formulary analyses, carry out meta-analyses of large samples of literature and compare clinical trial results. Obviously, before CureHunter, getting the evidence and analyzing it was a tremendously expensive, difficult and complex task not doable in real time: your patient isn’t going to stand in a gown in your clinic office while the committees go and find out if there is a better drug for her?

So how does CureHunter make evidence-based medicine available at the point of care without interruption in the clinical workflow? Click CureHunter in Patient EMR: in 10 seconds a meta-analytic graph computing relativistic drug evidence from the oldest to the newest data (1932 forward to the previous night) appears on the doctor’s screen.

“So Mrs. Jones, you aren’t doing well on the Avandia, let’s try the older and less expensive Metformin. The docs at Johns Hopkins found it just as good and often better. Here’s the data for the decision. Here’s a print out of the CureHunter meta-analysis comparing the efficacy of these drugs.”

This scenario is acted out in clinics all over the U.S. thousands of times a day: a patient has had an adverse reaction to first line default med, a patient can’t afford that med, a patient has an ER event as a result of the selected med, a patient has burned out an organ on one med because their condition is chronic, your first choice would interfere with a drug for another condition, your patient is allergic to your preferred med. Your patient is dying and you just don’t know what can help … until you check CureHunter. Some new drug maybe showing promise but you haven’t had time to research it. Those are just the basic ways CureHunter is functional right now in daily medical practice.

Every time a change of med decision is made, it will be more safe, effective and well tolerated if evidence-based. And that is what CureHunter functional utility is all about for the health care system. If a patient doesn’t come back because the drug you gave him or her the first time cured them … you just saved the system 2x the cost of treating them or more if the drug they did take causes serious adversity.

Some proof: two years before the law suits began the CureHunter engine highlighted MI, BP destabilization, and other major cardiac events as a good reason not to prescribe VIOXX. A ten second check would have saved millions of dollars in law suits not to mention, pain, suffering and death.

I could write a book on how we are approaching chemo therapy protocol optimization and most of that work is currently proprietary at the moment; but we do have several cases of patients reporting that the CureHunter recommendations for their doctors (all pro bono) improved their outcomes.

In the future, Justin and I hope to devote a great deal of our time to algorithms specifically designed for that task and we take it very seriously indeed.

Let’s talk healthcare information management. On your site we read, “CureHunter evidence data can be directly integrated into your EMR systems such as Epic, GE Centricity, and VA CPRS.” Let us say I am neurologist treating a 46-year-old male epileptic who has had such a terrible seizure that he ended up in the emergency room for the treatment of injuries from a bad fall and comes to me for help a few days later. I have never treated him before. I work in a small five-hospital network that lacks the sophistication of Kaiser Permanente when it comes to electronic medical records. What kind of EMR system am I likely to be using and what would CureHunter look like within it as I sit down to try to figure what the patient’s current drug regime is and as I decide what to do for the patient?

CureHunter EV-STAT is a prototype application showing the integration of the CureHunter evidence engine directly into the General Electric Centricity Electronic Medical Record. Physically the software interface can look on screen like an employee database record with various tabs for documenting the individual’s history with the organization. Most EMRs have treatment windows that show which drugs the patient has used before and is using currently along with various diagnostic and text windows.

Under a TAB called Drug Evidence, CureHunter can be hot linked into that record.

Nationwide, about 50% of EMRs are provided for smaller hospitals and independent practitioners by local medical system software integrators, and there are many simpler record systems available than those used by the major national providers. About 17 major health care payers control about 90% of the payment traffic for health care in the US today, so it is practical and efficient for small hospitals and practices to use a system that is a good fit for both their clinical uses and administrative and billing functions.

Now to your specific question on epilepsy: your neurologist probably knows the major drugs to treat epilepsy already. But he can with one mouse click in his EMR pop up the CureHunter blue graph of clinical efficacy where ten major high performing meds are displayed and rated. When the incoming patient describes the meds, if any, he is currently on, your neurologist can ask him how he is doing on them. And if not so well, with one look at the Chi graph the neurologist can find ten others with specific statements of evidence backing up their usage:

E.G. 05/01/2009:
“Epilepsy was well controlled in 65 out of 81 (81%), mainly with valproate and phenobarbital, and improved with age in all. ”

Checking the gold standard drug, finding an alternate, switching to a new med not contraindicated for the specific patient can all be done in seconds.

You mentioned on the phone that Britain’s National Health Service is one of your clients. As you know, in the current healthcare reform debate, the NHS is held up as a model of compassionate, socially equitable care by some and as a whipping boy by those who contend that its ambitious implementation of a nationwide EMR has been a hugely expensive fiasco. Can you discuss how CureHunter is used by the NHS and what lessons that use holds for policymakers and informaticians in the U.S.?

Over 65% of the heaviest users of CureHunter are from foreign countries that have very effective national health care systems. One of my oldest executive friends predicted this market phenomenon for CureHunter by pointing out that nations that can adopt technical standards rapidly, are in much better shape for improving care and lowering costs.

The Internet is so incredibly powerful and universal because everyone in the world speaks and writes TCP/IP, html and XML and C and Java. The PC industry only thrived when Windows became a dominant OS providing large markets through common specifications for all application developers. Science only prospers when we agree on what a kilometer equals.

So how do standards and EMRs and CureHunter and all that help the NHS and everyone else? It would be very simple to see how the country could save many billions of dollars per year, simply by using CureHunter to compute the optimal uses of generic drugs and getting more prescriptions right the first time.

I am quite interested in amyotrophic lateral sclerosis (ALS). And unfortunately, Rilutek is about it for drug treatments for ALS. I was intrigued by the wording on your homepage, “Discover new potential off-label applications … ” There have been studies (sadly, rather disappointing) of the potential use of lithium, minocycline and thalidomide in ALS. Can you give an example of how researchers could use CureHunter to discover potential off-label applications of existing drugs and how CureHunter can expedite such research?

Here’s the key underlying fact about off label drugs.

Doctors prescribe a lot of them, especially for their patients where all the known “good meds” are not working. Essentially they are making guesses that a drug related to a similar condition might help. Or a specialist friend down the hall says, “I’ve had some good luck with XXX.”

CureHunter takes the guess work out of off label prescription mathematically, algorithmically. We find all cases in the evidentiary literature where a drug not originally labeled for the target condition, none the less was used for it and a patient improved. You can see these patterns by clicking the related drugs and disease TABs in the interface after searching on your primary target drug or disease.

A second much more powerful function in CureHunter is implemented by Justin’s graph theory algorithms. In those analyses we are looking at graphs showing the maximal number of drug and biological mechanism connections shared by any two or more diseases. When a disease participates in a lot of shared biological connections, you can predict with some accuracy that a drug that worked for disease A, might also work for its friend, disease B.

People who belong to common groups are more likely to communicate with people who join their or similar groups. The cell phone company family and fave calling plans were all based on these types of data models. Who gets called by whom and how often?

So if you think that tumor necrosis factor alpha (TNF alpha) is very friendly with psoriasis, it might also know arthritis and be sending both those “pals” messages like, hey turn on, tune in and inflame yourself big time. Thus, if Network Graph Theory is used to define disease-drug communication clusters, a drug that makes a good call on one disease might do so on another.

The Center for Disease Control and Prevention (CDC) also uses Network Graph Theory to predict primary epidemic pathways and the growth of the number of individuals that potentially become vectors by “calling on friends and family” and school mates and work partners who don’t yet have the disease.

And again, using ALS as an example, can you discuss how working physicians and other healthcare providers could use CureHunter to research non-drug treatments for various conditions? For example, I just tried the “Relationship Network” feature — that is pretty cool and an interesting use of visual search. Could you discuss what users will see on the Research Interface and the concept underlying it?

The important thing for users of CureHunter to understand is that our implementation of the visual network is a show and tell function that brings together massive amounts of distributed information into an observable focus. 10,000 + data points might be supporting any single screen page full of connections. But you don’t really understand just how powerful those collections of data are until you export the underlying data sets to standard database and array formats. For example, with one mouse click you can ask to see every drug that ever successfully treated any cancer and how many different ones each drug treated successfully. If you are doing cancer research this lets you quantify in seconds the work of hundreds of thousands of investigators over 70 years.

In terms of my and Justin’s work on chemo therapy protocol optimizations we can find patterns among the most successful agents that lead to cocktails that can improve the total effectiveness of the regimen.

In CureHunter, you can just go in an enter the term “biomarkers” and all diseases for which we have found them will be linked with a click. Same with genes, hormones, specific proteins etc. So the system goes way beyond “drugs” as defined as prescription meds.

Recently we did an analysis of inflammation — a centric dysfunction — to see how many diseases it triggers and why. In seconds CureHunter can show you every disease where inflammation plays a major role from cancer to heart attacks and headaches. Thus pointing out why aspirin can be helpful in all of them, and even more interesting: statins.

You just can’t do that with any other tool in the world: find the center of evidence across the world’s largest medical library of peer-reviewed findings all cross-indexed automatically to curative properties and mechanisms of action associated with specific illnesses.

You have quite a variety of products and several different client bases. For example, consumers can use free the search tools on your home page. They can also order summary reports. Would you discuss how you see the e-patient/empowered patient movement developing given that you use such wording as “clarify diagnosis with your doctor?”

We want to empower patients with knowledge and our reports give them the pure scientific data on all the drugs that might help them. We know our reports are also great evidence summaries for our doctors because they order them and tell us so.

In general I see patient empowerment as a good thing when the patient can help his or her physician better understand the health problem or bring real high quality information to the table that their doctors have not yet had a chance to review. But I urge all patients to be empowered by science not hokum or sales literature or network TV ads. Do your homework, if you want to get well safely.

Could you discuss how some of the professions you discuss on your site (e.g. independent physicians, international users, private individuals and other certified prescribing personnel such as NPs and PA) could use CureHunter?

Well CureHunter Mobile, now available on your web-connected cell phone or PDA is a great tool for Nurses, PAs and med students to quickly check the definitions of major conditions or the drugs used for them. Our data is far more comprehensive than that from Epocrates or Stedman’s.

With CureHunter on your PDA you could be quite literally stuck in an airport waiting for your Christmas flight home and discover a cure for cancer by thinking about the results brought back to your screen.

In many parts of the U.S. — not to mention the rest of the world — specialists and pharmacists with deep credentials are very few and far between. There is nobody in the “next office” to ask what do I use now, when the default med fails. Well, with a mouse click you can ask CureHunter on your office PC — no fancy EMR system needed.

Pharma research scientists, however, are the ones that are totally in love with the system. They see its power quite clearly and directly as having major impact on their work and the bottom line of their companies.

Obviously, if you can directly compute new cures (drugs) for human disease you can dramatically speed up drug development and lower the financial costs to get new products into the market. Re-targeting existing meds for good new clinical purposes is a very powerful function we can apply over any company’s existing portfolio of agents and immediately convert old research costs to profitable assets.

By working with meds that have already passed human clinical trials for the original target, the most expensive dollar and time components of fresh development can be minimized.

Are any medical libraries using CureHunter? Is that a potential market?

Medical Libraries all over the world are using the product — we gave out about 5,000 free licenses for testing ALPHA-BETA versions — but I am afraid to continue our research we are going to have to start converting all users — except charitable foundations — to commercial licenses.

You mentioned that you have some rather interesting projects going on with various academic partners. Are you at liberty to discuss those?

Well I have good friends at Stanford and the University of California and the Pasteur Institute in France and at several universities in Great Britain working on cancer research, but because we are Portland Oregon-based and we have such a great teaching center here in the Oregon Health Science University and a terrific Bioinformatics Program running under Dr. Bill Hersh, I am hoping NIH or a major EMR provider will step up to the grant plate and let us do all kinds of experimental tests with the University.

Doctors Rick Deyo and Eric Orwoll Director of OCTRI are very interested in the translational medicine applications of CureHunter and that, of course, has been the theme of all NIH Bench to Bedside programs and a core message of President Obama. Get applied real world health care improvement payback now from advanced research.

People who read this interview may be puzzled (as I was) at the relative dearth of information on your Web site about the management and scientific team at CureHunter. Could you explain your philosophy of letting the science speak for itself and pouring money into the development of CureHunter itself as opposed to providing extensive professional profiles of your executive team?

Thank you for asking the question. It’s one dear to my heart. The Internet has proven itself to be one of the great human inventions of all time. It is also a den of iniquity of all kinds and a blind for much bad behavior.

One kind of bad behavior that I particularly dislike is where the start up company goes out and gets a laundry list of famous entrepreneurs or “advisors” and then lists them on a beautifully designed Flash page as some kind of BOD or Management Committee to indicate that their company is just doing the best work on the planet — when time and again those people have nothing to do really with the company or the quality of its products or even its staying power in the market, or even good management.

It’s sometimes the case that they advertise the Fortune 500 case winning lawyer and sell you the paralegal.

Does that mean I don’t like and respect venture capitalists, or brilliant young MBA Advisors, other successful entrepreneurs or good scientific counselors, no … not at all. We could use help from all of them ourselves.

But I don’t want people to validate us because of borrowed interest in a somewhat famous individual. Our whole story is the science and nothing but the science — the evidence — and that’s what I want people to know us for, test us for and buy us for.

It’s CureHunter’s ability to compute the evidence in real clinical time that makes it a powerful visionary product, a unique product that passes the Alan Turing test over 250,000x a month around the world.

Everything we do, every day is open to public scientific challenge and scrutiny because we do what no other health info system on the planet does: we predict outcomes that can be tested, compared, checked, validated by experts and novices, med students and senior physicians, Ph.D.-M.D.s from the Hopkins and Harvard, Oxford and Cambridge and National Science Foundation and NIH and FDA and NHS and NASA (another one of our heavy users). We have the real deal, no smoke, no mirrors.

So, what’s in a name? Not anywhere near as much “truth” as is in the data. Especially data subject to peer-review and scientific test by third parties with zero vested interest in our game.

How do you see CureHunter a year from now? In five years? In ten?

Well, we have our work cut out for us. We would like to find the cures for at least 10 major killer diseases and become the premier source in the world for evidence based medicine. Integration of our technology into one of the major EMRs is a clear next step in the 5-year window.

It would be a great service to science and health care if we established a central anonymized electronic health record depository — so powerful algorithms could process the learning from millions of real world case treatments and compare findings with those in the peer-review publications.

I would also like to have 5 — 10 world class pharmaceutical scientists working on drug discovery with us.

I think with our particular skills sets integrated into our scientific “swat team” of just really forward thinking guys and gals that respect and like each other a whole lot, we can get there much more quickly than anyone else.

I would like to add a few young clinical chemists, smart docs and mathematicians to the team as we go forward, plus just some great creative infrastructure people that want the most exciting job in the universe.

In the short term I am hoping some major foundation will step up to the plate and give us a charter to compute a cure for major illnesses. I think we deserve the real X prize for developing the first machine on earth to autonomously discover new cures for human disease. If we get a major outside funding round, we won’t just cure one disease, we will cure many.

But one of my major dreams is really simple in structure and doable immediately: because OHSU, Oregon Health Sciences University, wants to work with us on both our objectives for clinical research and drug discovery, a major pharmaceutical manufacturer could really speed their new product pipeline fill by endowing the OHSU-CureHunter Center for the Computation of the Cures for Human Disease.

Be practical, Judge, you say. Well that is extremely practical, for a grant of $3 — 5 million the sponsoring pharma could get first IP rights on many new agents and clinical applications: we could essentially complete the Molecule to Medicine Pathway at one center where researchers, doctors, and patients can all work together in synchronicity. The savings in new drug development costs would completely dwarf their out of pocket. Royalties could also feed back to OHSU researchers on drugs ultimately moved to market much faster.

Finally, who are your personal heroes in technology and in any other area?

I see you saved the toughest question for last, Hope.

In contemporary technology, I really do admire the people who turned the world upside down and helped us see it in a new way. I don’t think “hero” is the right word at all for our PC tech entrepreneurs like Misters Gates and Jobs or the Lion of DEC, Ken Olsen who first taught us all there was computational life after the IBM mainframe, or Mr. Turner that shook up the media network game or the Adobe boys that changed all we know about art and graphics by converting the hand drawn line to a set of a curvaceous numbers along with the likes of George Lucas and the Industrial Light and Magic brigade of compuartists that turned polygons into pure fantasyscapes. Dr. Wolfram’s book, NKS or a New Kind of Science … I’d have to say is the coolest single work that appeals to me personally. It’s just so wonderfully egotistical (just like me), driven, powerful, visionary and whacked enough to force serious new thinking outside all kinds of boxes.

Doctors without Borders. President Obama and his wife shattering all the old stereotypes ceilings. Oprah, I love that gal. You know what, Oprah has done more for Skype and all the rest of us humans in many ways than many of our best scientists by just showing everyone how cool it is to read. Reading is the real game changer, not Pong.

Thank you for your time.

You’re welcome, Hope … and be well.

Are you a Twitter user? Tweet this!

In part I of Hope’s interview with Judge Schonfeld, Judge talks about the development of CureHunter, the definition of “autonomous search” and the difference between CureHunter and other authoritative online reference services.

Thank you for subscribing by RSS or email. We work hard to make the articles on Next Generation Science engaging and we truly appreciate your interest and readership!

This article was published on Next Generation Science.

Hope had an opportunity to talk with Judge Schonfeld, CEO and Chief Scientist of CureHunter. In part I of their interview below, Judge talks about the development of CureHunter, the definition of “autonomous search” and the difference between CureHunter and other authoritative online reference services.

Before we begin, Judge, I’d like to give our readers a little background. I had looked at CureHunter some months ago but didn’t really get it. You and I then chatted for quite some time on the phone and you walked me through the many features of CureHunter and I was very impressed both by CureHunter and by the depth and breadth of your knowledge about the complex fields of medical data mining, artificial intelligence software and other topics. I think our readers need to know about CureHunter and would benefit from your comments on the state of search and the place of medical data mining in the worlds of healthcare information management, healthcare IT, the electronic medical record and the whole field of semantic search, specifically in medicine.

The Interview, part I

First of all, please tell us about the origins of CureHunter.

I and my sons have built a super computational expert biomedical system that at the push of a button — no more user interaction required — reads the National Library of Medicine archive from 1932 to 2009, thinks about what it has learned and completely autonomously discovers [directly computes] new cures for human disease.

Tonight, as I have done every night for the last 4 years, I will push a button and the baby CureHunter Machine will re-read the entire National Library of Medicine Medline archive, update all its knowledge modules, and e-mail me in the morning with any new cures it has found for cancer or any of another of 11,600 diseases.

And by the way, Hope, CureHunter can email any of your readers directly , too, on illnesses they treat, research or suffer from. Just go to www.curehunter.com, search on a disease and click the RSS orange news feed button at the top of the next page to which you are taken. There’s no charge to get the feeds.

And best of all CureHunter is not Medical Google or WebMD. It doesn’t think like they do, because they don’t think, and it does. Don’t be fooled by the presence of a simple search box. CureHunter doesn’t make lists of articles for you to read: it has read everything, evaluated the evidence itself and gives you an answer. Its “Machine 2nd Opinion” of the best drugs to treat any known disease.

From my point of view, it was really the integration of those three professional impulse engines of scholarship, business and science that trained me in how to build our machine.

I think the common denominator in our scientific swat team of a family is a love of symbolic languages — natural and computational and understanding how they function at a very deep level.

Every nursing and medical student and doctor and scientist who wants to understand CureHunter should start by playing around with the Visual Medical Dictionary function Alex built especially for students visiting our web site: http://www.curehunter.com/public/dictionary.do

When the Network Graph appears start double clicking each of the nodes that appear … until you can see how CureHunter’s brain connects the dots of discovery using its built in graph engine. What ideas link? Why?

The CureHunter Visual Dictionary will show you how clusters of medical and biomedical notions extend themselves from formal taxonomies into functional clinical data and leading hypotheses. It shows you that the machine thinks like you do with a neural network: it keeps putting two and two together … until you go, ah ha.

Dig a little deeper and you will discover that whole beautiful dedicated storehouses of alpha wisdom like the National Library of Medicine and Entrez databases holding trillions of words aimed at cataloguing close to a century of empirical observations from chemistry, biology and clinical medicine — are beginning to come together in one integrated data space.

Unified by a shared taxonomy and ontology.

So much data, so little time.

Scientists are born to search, but more importantly to measure. Often finding the key variable in phenomena to measure is at the center of new scientific understandings. CureHunter was born not to search for medical ideas, but to measure them. Born to instrument and quantify medical knowledge in a consistent, replicable way that would allow machines to think about specific findings, form hypotheses and predict new cures based on “quanta of biomedical knowledge.” How much do we know? What is the volume of our knowledge? What is the density of our knowledge? How strong is the force of what we know?

Like data clouds in modern wireless networks, information packets could be modeled as dispersing along vectors and instead of Internet TCP (transmission control protocol), CureHunter TCP would be based on transmissive curative packet networking.

How many messages does a drug send to its disease target?

You have a checksum, message receipt confirmation, when the disease changes its behavior and goes into decline. “Die evil tumor die” is the targeted message sent by most chemotherapy drugs. Tumor regression, tumor shrinkage, normal blood counts, no metastases, returning patient strength and a myriad of other measurable signs prove to you that the message got through. Incremented change was empirically knowable.

All of this thinking about CureHunter as a hard science instrument, grounded in information theory itself, is fundamental to why and how it can actually self-discover new cures for human disease.

Consider these three expressions:

Imagine these are linguistic equations — instructions in a high level programming language — that a machine can understand.

What comes after the = sign?

Change one key variable term in the NLP and you have a game changer technology. You have reached The CureHunter Center for the Direct Computation of the Cures for Human Disease.

You are no longer talking about search engines with people writing random queries into text boxes that direct them to lists of millions of articles they are supposed to go off and read one by one to determine if there was any meaningful data in the returns.

You are instead asking a computer for its answer. Its end point calculation of this thing called variable name cure.

Functionally, it’s very similar to the question our doctors answer 100x every day: what drug = X should (not can) I take for my Y = disease?

Many health information sites will tell you what drugs you CAN take … only CureHunter actually calculates from the evidence those you SHOULD take: the ones with the highest probability of making you well.

Every web site in vast homage to Google has today a search box where the users compose their own queries.

CureHunter, however, doesn’t work that way. We have a “search” box, but you don’t compose a query of any kind — because you are only asking one question.

At CureHunter you are always asking: “Machine, may I have your 2nd opinion, consult please on the best meds to treat my disease.”

So all you enter is the name of the disease you want to cure.

It is important for your readers to understand that in CureHunter all of the data inputs are “relevant” to start with, all of the time. They are precise measurements of quanta of key information about clinical efficacy. There is no such thing as relevance ranking algorithms — the basis of most web search engines — in the system. There is no laundry list of vaguely related articles you might want to read some day if you live long enough.

By rigorously defining the properties of trillions of variables existing in the medical knowledge universe, the CureHunter machine can systematically, algorithmically measure the expansion of what we know about what heals and predict the evolution of clinical efficacy.

Operationalized, the CureHunter Machine is a massive clinical outcomes relational database holding about a trillion variables that are fine focused and networked as functionally related agents and factors that achieve efficacy in human systems biology.

Ultra high precision context sensitive search feeds its natural language parser whose only goal is to extract with very high accuracy, 95% +, key clinical findings from raw text in the Medline archive.

These measured “facts” about what really worked to heal someone are stored in a single numeric array over which Network Graph Theory models and predictive analytics run transparently to the user and autonomously predict the best medications to treat any human disease based only on the evidence.

The user of CureHunter whether doctor, patient, or biomedical research scientist never writes a traditional search engine query. Never enters Boolean operators or filters, stop or non-stop words or alternate spellings and phrases — or dates, or times, or authors’ names.

He or she only enters one word: the name of a target disease or the name of a target drug or biological factor (protein, gene, vitamin, ligand, kinase, mineral et al). CureHunter’s brain automatically connects all the dots (theoretically possible queries) that contribute to successful clinical outcomes: individual disease-drug sets of relations are compared automatically to all possible relationships based on our fundamental concept of clinical efficacy indexes.

What data point can we measure that tells us this drug works against that disease? What did the instrument see?

Example of a typical Medline raw text sentence auto extracted by CureHunter from the peer-reviewed literature for analysis: “Remicade, a trade name for infliximab, a monoclonal antibody, achieved significant remission in PASI scores and reported pain in distal phalanges for 32% of the double-blind trial patients with both psoriasis and arthritis by suppressing Tumor Necrosis Factor-alpha induction of cytokine cascade and autoimmune inflammatory response.”

You can see that this single sentence extraction (one of many millions in CureHunter) has data points on multiple diseases, several mechanisms of action, key bio factors and pathology data along with specifics of clinical outcome response on established severity scales related to the illness.

Now imagine that every sentence in Medline 1932 — 2009 updated daily has been read by CureHunter and parsed to a similar depth and far deeper (cross linked to chemical abstracts, protein databank and molecular visualization systems) to store a network model of this idea about autoimmunity as structured data in its monolithic array of all causes, and cures, and outcomes.

Now imagine a doctor or scientist with a large enough brain to have read 20 million research articles, remember every data point in every article, and cross connect every similar finding to see patterns in that data leading to the cure of a human disease. That’s CureHunter the AI Machine. It’s not a search engine like any on this planet.

It self-authors the 11,600 Patient-Physician Summary Report studies available at our site as disease-specific monographs, untouched by a human hand or editor.

In those reports every drug that ever showed significant clinical efficacy against the target illness is meta-analyzed for its relativistic clinical utility; and CureHunter’s machine-written meta-analyses rival those produced by human expert teams in precision and accuracy. And often they are more comprehensive than human meta-analyses by an order of magnitude because the machine has no limits on how much it will read until it gets to the totality of all that is known, i.e. has been published in peer-review 70 years ago or last night.

Could you go to Google, or WebMD or Microsoft Health, or GoPubMed, or NextBio, or Elsevier, or Wolters or Thomson or UpToDate and ask with a mouse click that a fully documented peer reviewed evidence-based meta-analysis of over 200 drugs for your particular disease of choice be sent to you in 10 seconds?

No you cannot.

But, you can at the CureHunter web site, just enter the name of your target disease in the Patient text box at the upper left of the home page: www.curehunter.com

The rest is automatic.

CureHunter updates itself every night and changes its on line “machine 2nd opinion” consult drug recommendations if new data has just been published. Its self-written meta-analysis of effective meds for neoplasms is 770 pages long and hot links the reader of its PDF version to the source data for every drug ever found clinically useful against a cancer in peer review.

CureHunter is not a black box — if you want to audit the data trail from source to conclusion, with one mouse click you can from hot links in the report. And you can buy this book with all the best cancer research of the last 70 years meta-analyzed for $24.00, in 2 minutes on line at our site. Just enter “neoplasms” in the text box upper left of the home page. Read it, and you will pretty much know what all our best oncologists think.

Not only does the CureHunter Engine write its own reports with one mouse click, it also learns by cross connecting the clinical data from every study in its network graph head. And that’s how it can discover new cures and off-label applications for all existing medications. In a sense it knows how to think about everything we know and solves the burial by data problem by thinking across massive clinical and scientific silos of data to come up with a calculated answer to the question: what cures?

Please tell us about your two sons Alexander and Justin and their role in the development of CureHunter.

If you think AI, Medical Lexicography, Computational Linguistics, Artificial Life, Gene Sequencing and Code Pattern Recognition, Computational Biology, Machine Translation and very deep understanding of symbolic and natural languages — you begin to get the picture of the family business and scientific swat team behind CureHunter.

If you add in the availability of vast amounts of low cost computing power, on line access to the world’s best bioinformatics databases via the emergent Net, and years of best in class programming experience to solve these complex problems, you can see how our small team was able to produce a very powerful and innovative machine.

With zero advertising, PR, marketing or promotion within one year, the British National Health Service had become the largest single user of CureHunter in the world, with its Physicians asking for its Machine 2nd Opinion of the best medication to treat a target human disease over 250,000 times per month.

Doctors and Research Scientists at Stanford, Harvard, the Mayo Clinic, the Pasteur Institute, NIH, FDA, Pfizer and many other major pharma were also audited heavy web users of the server.

Independent Web auditors measured the CureHunter audience as most often visiting also: the British Medical Journal Group, Science, Nature and the New England Journal of Medicine. Out of the box and on to the Web, our young company was matching the best in class authoritative sources for the highest quality scientific medical information.

How did you come up with the name CureHunter? Why not, say, “TreatmentFinder?”

A CURE is the ultimate benefit of all medical treatments and the central focus of the CureHunter Engine: the provision of an evidence-based cure with one mouse click.

Sick people aren’t looking for treatments. They are looking to be well, to feel right, to function correctly — to be cured.

The ability to deliver a cure is why we have doctors, biomedical researchers, health care providers, and pharmaceutical scientists in the first place.

A cure is the gold in any medical data mining system. And contrary to popular opinion, the term appears extensively in the scientific literature as the desired end point of a clinical treatment.

What do you mean by “autonomous search” — what is the origin of that phrase and how does it relate to semantic search? What is the difference?

Those of us who have studied search, formal semantics, and linguistic technology for many years know that the way you ask a question will determine both the quantity and quality of the answer. A single an/or/of/the or other stop word can vary returns by many orders of magnitude and degrees of accuracy.

“Autonomous Search” the Expert System search in has a virtual expert medical librarian built into its cognitive model with the ability to resolve synonymous cross references, multiple naming conventions and multiple meanings with great accuracy. Notice if you pause your mouse over any technical term in CureHunter canonical definitions of all the various names and abbreviations for the drug or disease or bio agent you have selected pop up in a yellow window block.

Notice when using CureHunter, you write no queries. The machine knows all possible queries, and they are hard wired into it for speed, accuracy, precision and direct computation of scientifically verifiable results. You get no results that are sort of relevant or sort of more like this or sort of more like that. All of its results are specific data points suitable for use in computations that reduce error, rather than induce it.

In general, the phrase Semantic Web Search and Web 2.0 technologies refers to all text search and analysis capabilities that have some level of semantic intelligence embedded in their technical implementations with the goal always being to resolve ambiguity in the original query to the greatest extent possible.

Various engines are more or less successful at dealing with the problem. I will say for the record that Google and Microsoft just don’t understand the problem very well at all and don’t seem to know the correct way to approach solving it because they have been trying for over a decade with many brilliant people and a small fortune in funding — and still they don’t get it.

This is demonstrable on any given day just by putting the same exact query into multiple big market share search engines and studying what they bring back as “relevant” Imagine in analogy with machine translation that you ask someone to bring you an apple and they keep coming back with a giraffe. Some party to the bipartite communication isn’t getting the message accurately at all … and that’s why you the questioner aren’t getting the answer you seek.

You say on your site, “… our mission is to help people get well by offering both patients and physicians the opportunity to get a “machine 2nd opinion” on all complex drug decisions.” Could you explain how CureHunter differs from UpToDate, MD Consult and McGraw-Hill’s AccessMedicine?

The high quality consult products you list are very good — sound, ethical products with good science underlying them in general. They don’t bring back search noise in the manner of a Google, Yahoo or MSN general purpose engine as a general rule relative to the volume of valid signal they produce.

But fundamentally they are created by asking a theoretically expert group of human specialists — with the necessary tunnel vision and limited currency of directed expertise — to periodically up date the rest of us on what they know.

That assumes the experts — many of whom are in daily practice — have the time to read all the new research, compare it to all the old established research, take notes, structure data, and carry out an objective meta-analysis for every opinion they write.

Do you believe they are working that way? Honestly, I do not.

It is not humanly possible to carry out their tasks that way any more.

Consider again that burial by data is the problem and it is getting worse all the time for all of us and all our experts. UpToDate — while it is an excellent quality resource — is ironically nowhere nearly as “up to date” as CureHunter and every second it is less so.

Over 50% of the peer-reviewed literature reporting a successful clinical outcome is likely NOT to be published in the Journal of Record for the specialty experts nominally caring for that disease — or the specific expert contracted by UpToDate, Consumer Reports or WebMD to write for that specialty.

For example, a key finding in an oncology or neurology journal relates directly to an autoimmune disease but is published in a journal on cardiology or dermatology and is never read by the oncologist or neurologist.

This is the law of diminishing knowledge returns that is proportional to the increasing depth of specialization in many scientific fields: it is burial by data in action.

One of our early QA studies to see if CureHunter was missing important findings was a histogram showing the distribution of good journal data over sources not read by the primary treating physician. In short, how measurably ignorant — unread — is the treating doctor likely to be?

No good doctor is stupid, but many are ignorant of massive bodies of data that could change the way they treat their patients.

You can also see this result empirically by comparing meta-analysis papers retrieved from Medline to those created by CureHunter. Invariably, the human authors miss many important and substantive bodies of data because their original search query for source papers was totally inadequate, even though in many cases they collected a lot of data.

To be expert in the way that CureHunter is expert, the human would have to read all of the journals all of the time for all the specialties and general practice journals and cross connect significantly related findings from all of them automatically.

CureHunter does that every time you click your mouse and it is easy to see measurable results by clicking the Related Diseases TAB where diseases sharing key drugs and bio agents across their cure or pathology history are automatically connected.

Example: What do ADHD, Dementia, Parkinson, Multiple Sclerosis and Schizophrenia all have in common? What proteins, what treatment drugs, what genes? What do the interconnects tell you?

You can’t find those answers with the tools you mentioned, unless perhaps the medical librarian takes months off to go and look down every interconnect trail and document her every finding.

On the other hand, if you want to wrap a lot of medical canonical and often commodity context information around an established treatment, they are excellent tools; and I would hope that all my physicians have access to them.

Sometimes a doctor will want to check his diagnosis, a list of symptoms, possible complications, possible co-morbidity, or default med selection against another authority; the above are good tools for the task.

I have great respect for authoritative textbooks, canon, careful incremental science and complete documentation especially in the field of clinical medicine where the doctor is burdened with life and death responsibility every time he writes a prescription whether for aspirin or exotically toxic chemo.

In our phone conversation you stressed that CureHunter is more of clinical decision support system sort of tool than a search engine. For those of us with medical library rather than healthcare information management backgrounds, could you please discuss how you would classify such things as CureHunter, UpToDate, MD Consult and AccessMedicine versus what I would call search engines such as DeepDyve, Mednar and the gold standard of medical search, PubMed? To wit, who uses what and for what?

If you think about CureHunter, you realize that its instrumental job was to put its “machine eyeballs” on mountains of raw text and convert it consistently to digital format that can be used to directly compute the answer to the problem of the best drugs to treat human diseases. And, furthermore, to extrapolate from those known data stores to new cures.

Thus the big difference in a lot of the products in the medical information space has to do with their original specified design purpose. What information are they trying to model: pharmaceutical research, treatment research, drug delivery, or other and combinations? How well have they digitized input sources?

What answers are the packages trying to compute if any? How much interaction with the interface do they require of their users? Do they provide results that are better, faster, deeper and cheaper than those that can be retrieved by human hands in the old fashioned way of graduate researchers everywhere. Go to the library, find the right articles, read a few, make notes, derive a conclusion from the data.

Some software tools are screwdrivers with unique tips, others are ratchets with multiple heads-functions-some are one-size fits all adjustable information tools that can tell you a little bit about almost everything, but nothing very original or with great precision about anything.

Each one of the tools you mention has value:

In the final analysis, some products are more useful for research — off line — than in daily practice and some solve immediate clinical problems. For example, most modern pharmacies run contraindication alert software that flags possible drug hazards for patients already on multiple drugs or about to be prescribed a new med. That’s a very specific narrow, but important function. No research or intelligence is in the tool. Tables are compared.

The CDSS class of software, the clinical decision support products, often yield immediately actionable diagnostic or prescription results: e.g. If patient shows fever with X, then Y is the highest probability of diagnosis.

Dr. Octo Barnett’s DxPlain software he developed with his Harvard Medical School students and in use for years at Mass General is a good example of a diagnostic answer system supporting Clinical Decisions.

If patient is contraindicated for drug X, then 2nd line default is Y. Typically the CDSS has significantly more intelligence than drug alerts, but they are designed to provide alerting answers to clinical questions in as close to real time as possible. They are logic, flow chart, and table driven to put a known answer, and not another question in the doctor’s head.

Mednar, while trying to add search value through focus on the medical domain, does not clearly (to me) offer any technical advantage over any other generic medical search engine whose primary output is a list of articles for the reader to go read him or herself on the generic topic they have entered in the query field. Their top 10 returns are usually from Medline Plus or PubMed. If WebMD, Mednar, Google, MS Health and Medline Plus itself all deliver the same top 10 articles for the human to go read personally from primarily the same free sources, what’s the added technical value of one over the other?

PubMed as the grandmother of all biomedical archival and search systems is comprehensive and authoritative, up to date, and properly maintained year over year; but it is not an immediate answer or action system.

It does not let you directly compute any treatment answers per se, automatically generate reports or automatically discover new cures for human disease. You can’t directly export key findings in its holdings to computable analytics systems and, of course, that’s exactly why you would use CureHunter.

PubMed is the best biomedical STUDY resource in the world, but it does not deliver STAT answers, generally speaking: it gives you a list of articles to read.

CureHunter combines both STAT and STUDY functionality: the blue meta-analytic graphs of relativistic drug efficacy appear in 10 seconds … if you then want to study the data sources supporting the graph — which has plotted the best drugs based on the evidence — click on through to the hot links to the sources.

I wouldn’t want to criticize the science of others further without doing a detailed analysis of their algorithms and methods and sources and design goals. That not always being possible, the best thing for medical librarians to do is what IT managers have done for years: meet with their users and understand the applications in great detail. Have product demos for your user stakeholders. Sign up for trial licenses, etc. Test and you shall find.

Seeking and searching alone are not enough.

Are you a Twitter user? Tweet this!

In part II of Hope’s interview with Judge Schonfeld, Hope focuses on the users and uses of CureHunter; Judge discusses the differences between CureHunter and Wolfram|Alpha, and compares search results from CureHunter to novo|seek, GoPubMed and PubMed.

Thank you for subscribing by RSS or email. We work hard to make the articles on Next Generation Science engaging and we truly appreciate your interest and readership!

This article was published on Next Generation Science.

In an age of federated and semantic search engines such as Mednar and DeepDyve, which provide links to a variety of pages on the Internet, Wolfram|Alpha instead responds to specific queries using its own detailed store of factual information and a family of algorithms found using the NKS system.

This is something altogether different than an index of the web.

Taken from Wolfram’s 2002 book, A New Kind of Science – NKS for short – the NKS system is based on the idea that each natural phenomenon can be regarded as a computation that has a very simple underlying rule. Stephen Wolfram sees Wolfram|Alpha as tool to systematize knowledge, to make the world computable.

Although many are skeptical of the project, I for one and excited about Wolfram|Alpha. My initial training in the sciences was in physics. I spent a great deal of time developing mathematical models of natural processes, so I’m very comfortable with the idea of using algorithms to take what we know and determine something new.

I’m confident the system will work (heck, even Google can do simple computation). However, I think more important questions are “what are its limits to extend beyond mathematical computation” and “how well will it interpret natural language queries?”

According to the Wolfram|Alpha Blog, they’ll start webcasting preparations at 7pm CDT on justin.tv and include behind-the-scenes views of what it’s taken to create Wolfram|Alpha.

For more on Wolfram|Alpha, check out Rudy Rucker’s in-depth interview with Stephen Wolfram.

Are you a Twitter user? Tweet this!

Thank you for subscribing by RSS or email. We work hard to make the articles on Next Generation Science engaging and we truly appreciate your interest and readership!

This article was published on Next Generation Science.

Okay, a search engine or business strategy has to be really intriguing or really powerful to get me out of bed in the middle of the night to write about it as I am doing right now with DeepDyve.

DeepDyve delivers.

I am also a sucker for catchy slogans, like DeepDyve’s, “search for research.” DeepDyve not only offers some pretty impressive technology, it is also developing just out and out brilliant marketing strategies that create win-wins for publishers, authors and those of us like medical librarians, other information professionals and academics who make a living from and live for the fun of the fastest possible acquisition and provision to patrons, clients and peers of the best data possible.

Please, please take notice scientific and technical publishers. There is a vast potential and in many cases deep-pocketed audience for your superb material and DeepDyve can help you get it to us. In the age of Twitter where viral marketing and microcontent rules, DeepDyve is the best of friend of multiple actors. I wish I were as smart as these boys and as someone who loves finding just the right abstract of just that key article to send to a patron and alert him or her to the existence of a journal neither of us had heard of before, I am fondly hoping that publishers will pounce on the marketing savvy of DeepDyve to spread the news of their treasure troves of scholarly material that is lamentably under-disseminated at this point.

Got me on all that? It is hard to write about all of this succinctly and pithily because there is so much to discuss here given the fact that I want to discuss DeepDyve both as a search engine and as an innovator in information discovery and marketing.

Let’s start with the search engine aspects. In a word, superb. The interface is attractive and sleek. Pretty even. (Okay, men — it is okay that a search engine is pretty — if I am going to spend hour after hour in a search environment, I want it to be beautiful. DeepDyve and I are going to get to know each other well and I like the looks of it both aesthetically and technologically.)

And there are gobs of powerful features. I have just been trying them out with my usual favored search term, amyotrophic lateral sclerosis. Loved the excellent results I could call up with “More Like This” and loved being able to highlight certain phrases and drill down incredibly quickly and with dream-like effortlessness.

Read Hope’s entire review here.

Thank you for subscribing by RSS or email. We work hard to make the articles on Next Generation Science engaging and we truly appreciate your interest and readership!

This article was published on Next Generation Science.

Mountain View, California based NetBase launched its Content Intelligence platform today, which is powering a new generation of consumer and B2B content-rich applications. I had the opportunity to attend a briefing last week where they showed off their technology and I have to tell you, as someone who searches Google and PubMed daily, and is well versed in constructing complex search queries to try and return the most relevant results using keyword search, I was extremely impressed with their technology.

How many times have you run a web search and paged through tens, if not hundreds of pages of results? Research firm IDC (Interactive Data Corporation) estimates that in 2006 the world produced 161 Exabytes (trillion Megabytes) of digital data encompassing 70 million blogs and 150 million Web sites. That’s almost 5 terabytes per second. By 2010, it’s estimated that number will grow tenfold. Yet traditional search by keyword fails to address this explosion of information because it returns too many documents with too many irrelevant results and lacks actionable insights and answers.

NetBase’s Content Intelligence platform uses state-of-the-art analytics technology to search, parse and summarize information from any number of industry sectors. Content Intelligence addresses search in a new way by reading every sentence inside documents on an ongoing basis, linguistically analyzing them and identifying relationships between entities and attributes. The information is then stored in structured semantic indexes. NetBase CEO and co-founder Jonathan Spier and Jens Tellefsen, VP of Marketing and Product Strategy, talk more about what makes NetBase and the Content Intelligence platform different from traditional search engines in the video below.

The evolution of search

Keyword search using statistics such as number of incoming links to calculate relevancy and popularity is context independent, and that’s a problem as the amount of information on the web continues to grow.

Linguistic search is different, relying on pre-defined lexicons to extract domain-specific entities from documents. While this is more useful than keyword statistics, lexicons require a lot of time and resources to construct and maintain. Further, linguistic search is often unable to deliver actionable insights and answers from large amounts of data across multiple domains.

In contrast, Content Intelligence focuses on understanding the meaning of sentences — independent of lexicons — by identifying the connection between entities. The relationship between keywords is crucial to the understanding of sentences and that’s where Content Intelligence shines.

NetBase believes that Content Intelligence will solve the problem of information overload, enriching existing content and surfacing high quality, meaningful, contextually aware insights. NetBase’s sematic index can be deployed to access any content and is highly scaleable. The Content Intellegence platform doesn’t rely on hard-wired or human-edited data to identify relevant relationships within content.

Elsevier’s illumin8

Last month, NetBase announced that it’s continuing its customer relationship with science and health information publisher Elsevier. Elsevier has been using NetBase’s Content Intelligence platform to power illumin8 since early last year. illumin8 is a web-based research tool that integrates natural language search technology with content from Elsevier’s full-text scientific articles, millions of scientific abstracts, patents and billions of web sources to give users actionable solutions for research initiatives. The agreeement extends its commitment to NetBase for another three years. To see illumin8 in action click here to view an illumin8 search on “chip cooling”. The interface is designed to return results organized in categories including organizations, products, people, approaches, benefits and related results.

NetBase for Healthcare

NetBase has also announced that it is expanding it focus and marketing to other industries, in particular healthcare. At a time when healthcare professionals and biomedical researchers are looking to collaboratively leverage research, intelligence and new technologies, cutting time to finding relevant information has become all the more important. NetBase for Healthcare enables researchers, doctors and patients to quickly and efficiently search a vast and rapidly expanding number of books, medical journals, databases, Web sites and patient records, creating a new way to discover and use medical content. For example doctors and nurses can look up a symptom like hypertension and instantly see an organized, up-to-the-minute summary of the causes, effects, symptoms, complications, side effects and treatments of the condition. Additionally, consumers and patients could do their own research for alternative treatments, suggested lifestyle changes, or less expensive medications or treatments to cut down on increasing healthcare costs.


As the amount of digital data on the web continues to increase, more sophisticated solutions are necessary to find the actionable insights and answers we need without sifting through thousands of documents. NetBase’s Content Intelligence platform is unique in its ability to understand every sentence of every document without lexicons or human editing and extends search by an order of magnitude over previous approaches. illumin8 was designed specifically for corporate R&D knowledge workers. With NetBase’s expansion into healthcare, we should soon begin to see their search platform become more widely available.

Now, if only I could use this search technology on PubMed directly (PubMed currently uses keyword search). One has to wonder why scientists are using antiquated technology based on keyword statistics to search through cutting-edge biomedical research literature. National Library of Medicine, are you listening?

Are you a Twitter user? Tweet this!

Thank you for subscribing by RSS or email. We work hard to make the articles on Next Generation Science engaging and we truly appreciate your interest and readership!

This article was published on Next Generation Science.

Hope attended the Web 2.0 Expo last week in San Francisco, California, covering it for AltSearchEngines. The Web 2.0 Expo features some of the most innovative and successful individuals in the industry from companies such as Google, Twitter, Microsoft, Amazon, Electronic Arts, PBwiki, Nokia, Six Apart and Adobe.

The conference’s theme this year was The Power of Less:

We do some of our best work when we’re constrained: by budgets, by headcount, by technology, by the economy. These are the times when bureaucracy and waste die by necessity. What’s left are ideas, and the muscle to make them real.

2009 will be a tough year in many ways, but now more than ever, the core concepts of Web 2.0 provide an advantage. Lightweight tools, user interfaces, and development models will help streamline productivity and focus resources; new business models will emerge out of the environment of change. Transparency and openness will help avoid disasters and extend our influence, as we learn to trust users as co-developers. Marketers can’t afford to ignore the value of social media, communities, and a new set of analytics. On an individual, team, company, and global level, this is the year we will choose to work on what matters.

And while the fundamentals of Web 2.0 become increasingly relevant and urgent, this is also the time to look ahead. The landscape today enables interactions that were not possible even two years ago, as the devices around us and the data they generate evolve rapidly. Tomorrow’s big ideas are quietly percolating now, in the aisles of major retailers, the back office of enterprises, and in the labs of passionate, independent-minded hackers. We’ll explore the future potential of Web 2.0, so when this storm passes, you’ll be ready.

Over the course of four days, the expo spotlights experts, leaders and under-the-radar innovations, and provides attendees the opportunity to attend lectures, workshops and exhibitions. You can read more about Hope’s thoughts and experiences at the conference in these articles posted over at AltSearchEngines:

• The Leman Report – Web 2.0 Expo Day One

  Hope’s first report — a late night recap of impressions from the first day — includes a narrative on a session focused on Web monitoring and Web analytics.

• The Leman Report – Christina Wodtke of LinkedIn

  An even later late night recap of day one: Hope describes a session she attended on social website design by Christina Wodtke of LinkedIn.

• The Leman Report – Observations from the Web 2.0 Expo

  On day two of the Expo, Hope attended sessions on Web design for the disabled, IT transformation with cloud computing and leveraging Twitter.

• The Leman Report – Larry Chiang and the happy accident

  Hope recounts a session by Larry Chiang titled Peek Into a Secret Society of Entrepreneurs.

• The Leman Report – An Inside Look at Web 2.0 Expo

  Hope provides a sampling of sessions held on day three of the Expo and highlights several technology exhibits.

• The Leman Report – the final day of the Web 2.0 Expo

  Hope reflects on her final day at the Expo and discusses whether Web 2.0 has separated into subgroups, precipitating less need for a large-scale conference such as the Web 2.0 Expo.

Last week, Hope also took the time to guest post over at Nextbio, discussing how FriendFeed and Twitter are part of a larger revolution in scientific communication. Widespread academic use of email began roughly fifteen years ago. Today, researchers can’t imagine not using email to communicate. It will be interesting to see if and when microblogging and feed aggregation become indispensable tools for researchers and scientists to communicate and share information.

DId you attend the Web 2.0 Expo? Leave us a comment below on your experience and impressions of the conference.

Are you a Twitter user? Tweet this!

Thank you for subscribing by RSS or email. We work hard to make the articles on Next Generation Science engaging and we truly appreciate your interest and readership!

This article was published on Next Generation Science.

Evan Williams is the co-founder of Twitter. He spoke recently at the annual TED conference in Long Beach, California. TED stands for Technology, Entertainment, Design, and provides some of the world’s most fascinating thinkers and doers the opportunity to talk about science, business, the arts and global issues facing our world.

One of the things I found interesting about Twitter is that it was launched as a side project. In his TED talk, Williams shares that he’s learned to follow his intuition:

Now, it’s hard to justify doing a side-project at a startup where focus is so critical, but I had actually launched Blogger as a side-project at my previous company thinking is was just a little thing we’d do on the side, and it ended up taking over not only the company but my life for the next five or six years, so I learned to follow hunches even though you can’t necessarily justify them or know where they’re going to go. And that’s what’s happened with Twitter time after time.

He goes on to explain how many of the ideas driving Twitter’s growth have come from the users themselves. Check out his short 8 minute talk below.

Last month, Nash over at The Daily Nash-on came up with a list of reasons Why Scientists Won’t Use Twitter. He points out that the reasons he came up with are also applicable to other “Web 2.0″ Science resources. Perhaps it’s no surprise then that someone like me who embraces Web 2.0 Science resources would be using Twitter. I’ve found Twitter to be a valuable networking and communication tool, especially for one-on-one conversation/messaging.

The problem with Twitter

The problem with Twitter is that it can be a “noisy” social media platform. The way to manage the noise is to only follow high-value Twitterers. That is, only follow people who closely share your interests and post relevant content. That way you maintain a high signal-to-noise ratio. You can determine high-value Twitterers by screening new followers before you start following them back. Here’s some tips:

• Check out recent tweets and take a look at who they follow — their network will reflect their interests and the conversations they may participate in.
• How many people are they following? The ratio of people followed to followers can give you an idea if they also screen new followers.
• Check out the link provided on their profile. If it links to a blog, read some of the recent posts. Look for an “About” page to learn more.
• Check and see if they have a FriendFeed or LinkedIn profile.

David Bradley at Sciencebase has been compiling a list of Twitter users that are connected to science in some way. Review his list of 252 Scientific Twitter Friends. In addition, WeFollow, a user powered Twitter Directory, currently has a list of 164 users that have tagged themselves with the term “science”.

Do you use Twitter? If so, let us know in the comments below!

… and don’t forget to follow us @NextGenScience

Are you a Twitter user? Tweet this!

Thank you for subscribing by RSS or email. We work hard to make the articles on Next Generation Science engaging and we truly appreciate your interest and readership!

This article was published on Next Generation Science.