Curated by Hank Shaw, a Xerox contributor

There’s a goldmine of information in medical documents.

Scientific publications. Clinical guidelines. Reports on clinical trials. Not to mention patient records, epidemiological studies and clinical notes.

Every day, this mountain of knowledge grows larger. So does its potential to advance the field of medicine and improve the lives of patients around the world.

But there’s a problem.

This article was first published in Real Business, a website from Xerox that provides ideas and information for decision-makers in business and government.

This article was first published in Real Business, a website from Xerox that provides ideas and information for decision-makers in business and government.

Much of this valuable intelligence is contained in countless unstructured text-based documents. So it’s virtually impossible for medical researchers and clinicians to keep up with the explosion of information by reading and annotating every page.

Now for the good news. New technologies for natural language processing can unlock the information in these documents. Going far beyond the limitations of keyword searches, they can analyze text, extract information, automatically find important connections and correlations, create structured databases, and even find subtle shifts in language associated with new discoveries and results.

Here are five ways these powerful tools can make a big impact on the field of medicine and clinical care in the months and years ahead.

1. Advance the cause of medical research. With help from advanced text analytics and smart information extraction tools, researchers can quickly analyze a huge volume of scientific documents and discover hidden correlations that can lead to new discoveries and treatments.

Take the fast-growing fields of genomics and phenomics, for example. Advanced text analytics can detect references to gene protein interactions contained in scientific papers, and even create a comprehensive map of interactions. These technologies can also help researchers search through large-scale epidemiological studies to find links between diseases and specific genes.

2. Keep critical databases up to date. Natural language processing tools are so advanced today that they can detect ‘paradigm shifts’ in syntax and semantics often associated with new findings and discoveries. Take this phrase for example: “…a new experiment has demonstrated that molecule X is no longer…”

By finding similar phrases in scientific papers and medical research documentation, these technologies help experts quickly identify important new results. It dramatically simplifies the task of keeping critical medical research databases up-to-date.

3. Help hospitals reduce errors and manage risk. Automated language processing tools can sift through an enormous variety of clinical documents to find information associated with errors or risks. In a pilot project in France, an automated system designed to detect hospital-acquired infections in discharged patients was accurate in 87 percent of the cases.

 4. Improve clinical guidelines. Automated language technologies can also be used to mine clinical literature and propose changes and improvements to the experts who manage clinical guidelines. As a result, they help bridge the gap between medical research and medical care.

5. Facilitate the enrollment of patients in clinical trials. Language processing tools can even analyze patient eligibility criteria and find suitable matches in patient records, expediting the enrollment process for clinical trials.

Bottom line? Text-based documents play a vital role in medicine and healthcare. But the proliferation of these documents makes it difficult for experts to maximize the use of the valuable information they contain.

With help from automated text-mining tools and other natural language processing te

chnologies, however, researchers and clinicians can deal with the explosion of information in their fields and find new ways to advance the cause of medical research and clinical care.

Now, isn’t that a healthy development?

Based upon the article “Medicine Leans On Natural Language Technology to Advance Science and Improve Patient Care,” authored by Xerox Research Centre Europe (XRCE) researcher Caroline Hagège, and Denys Proux, an XRCE project manager.

 

It is from a series of articles from XRCE researchers that celebrate the lab’s 20th anniversary.