Skip to Content
Uncategorized

Doctor Database

New search technology from IBM could help patients and doctors locate life-saving treatments.
April 7, 2006

Sometimes the best hope for a person with a serious illness is to become a subject in a clinical drug trial. Such trials are often hard to find, though, as they’re rarely well publicized. Additionally, doctors may not know about the best trial for a patient, because at any one time thousands of studies are being conducted around the world. As a result, finding a useful trial has usually required hours of intensive searching or having a doctor who’s conducting an appropriate trial or knows other doctors who are – or just plain luck.

Now an initiative is making information from more than 88,000 completed and ongoing clinical trials searchable through a single website. In late March, the International Federation of Pharmaceutical Manufacturers and Associations (IFPMA) and IBM announced the IFPMA Clinical Trials Portal that they hope will enable doctors and patients to find potentially useful trials and to make more informed medical decisions based on past trials. To facilitate this, the portal is designed to cut through medical jargon, correct misspelled search terms, and search for results in five different languages.

“Clinical trials information is scattered all over the place,” says Marc Andrews, director of strategy and business development for content discovery at IBM. “These trials have been conducted by multiple companies, and it’s been difficult to find information needed to participate in trials relevant to a life-threatening diseases,” he says. “There was no one place you could go.”

The new, free portal is powered by IBM search software called OmniFind, which pulls together disparate information to make it searchable, Andrews says. OmniFind is based on the Unstructured Information Management Architecture, a set of processing engines that sift through different types of data (PDF, text, and HTML files) from many different sources (for instance, databases and websites), to pick out the information buried within documents that best match the search terms.

“We’re not just processing through basic indexes like most Web search,” says Andrews. While Web search engines such as Google sort through indexed web pages, ranked by title, key words, and the number of hyperlinks connected to a page, OmniFind digs into the body of the text. It pulls in specific information and assembles it into concepts that are useful to someone who wants to search through technical documents. “Instead of indexing words,” Andrews says, “we’re indexing concepts that are referenced in the documents.” For example, when “lung” is searched, the software will also look for the word “pulmonary” in documents and files.

Andrew says that IBM added this layer of intelligent search by “training” the software with human experts. IBM’s Business Consulting Service (BCS) employs these experts in various technical areas. They apply their knowledge to tag words, phrases, and concepts that the software then uses to find relevant information. In the development of the clinical trials portal, it took health and medicine experts a few months to create these specialized tags, according to Andrews.

Another feature of the portal is language translation. “Quite a lot of the world speaks English, and much of the trials information is authored in English,” says Johnnie Summerfield of IBM BCS, who worked on the project. But not everyone speaks and reads English of course. So the portal is also searchable in German, French, Japanese, and Spanish.

According to Andrews, the language translation feature was built into OmniFind using IBM software called LanguageWare, which also corrects misspelled words. The language translation software is customizable for words in specific fields. Since medical terminology is used in clinical trials information, the software pulls from a dictionary that includes medical terms in five languages.

Summerfield suspects that, given all its usability features, the clinical trials portal could affect more than doctors and patients. Another area where the project could have an impact, he says, is in the transparency of clinical trials procedures and results. It’s a topic that’s been making headlines for a number of years, and recently in the clinical trial of PolyHeme, a blood substitute developed by Northfield Laboratories, in which ten patients suffered heart attacks and two died after receiving the treatment. The trial was stopped early and the results weren’t made public, the Wall Street Journal reported in February.

“By providing transparency, [IFPMA’s Clinical Trials Portal] will improve information because many of the organizations referenced through the portal will inevitably seek to make their information clearer and better represented through the site,” Summerfield says. “This will increase the quality and also probably help to increase and improve the sensibility of that information.”

In the meantime, however, the clinical trials portal has already elicited responses from cancer survivors, Andrews says. A day after announcing the portal, for instance, IBM received an e-mail from a lung cancer survivor who believes a clinical trial helped save her life. She described her challenge in locating a clinical trial. “She finally found one at a university website,” Andrews says, “but it took her a while.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

How scientists traced a mysterious covid case back to six toilets

When wastewater surveillance turns into a hunt for a single infected individual, the ethics get tricky.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.