30 August 2013

Search Engines is applied to health-care data

The power of search engines is applied to health-care data


Through the window of his office in Seattle, Thomas Payne can see two hospitals and he use same EMR Software System. And yet, according to Payne, medical Administrator of University of Washington, yet continue to exchange information by fax or paper.



This situation is the current standard in the fragmented and distributed healthcare universe where electronic medical records (EMRs) increasingly must extract more information from the various health institutions, clinics, hospitals and specialists, where people receive care. However, providing physicians with access to the right information at the right time dramatically simplify healthcare. It is estimated about U.S. $ 77.8 billion, or 5 percent of health care costs could be saved each year in the U.S. if it was installed a system of completely inter operable exchange of records, according to a 2005 study published in Health Affairs . Most of the savings would come from avoiding duplication of tests.


Michael Zalis, a radiologist at Massachusetts General Hospital, has experienced first-hand the problems of chipping system. It states that, in particular, patient data collection, she stole a long time doctors, occupying up to 20 percent of their day. Even with an EMR, medical records span multiple databases that do not necessarily communicate with each other. Test results older and less relevant may well end up as often as the new, says Zalis, making the doctor has to analyze all visually. The problem will likely enlarge as many of the national hospitals and medical consultations using the new stimulus funding to implement EMR systems.

Zalis wants to relieve this problem by applying the lessons learned from the search engines in medical databases. By compare the current circumstances within the EMR to the early days of internet: an amount of data becoming more and larger with a growing need for indexing and search capabilities. Before they appeared good search engines like Google, the search for a proper web page could become a challenge. Many links and pages had little relevance, or were so out of date that it was almost useless.


In the past five years, a team led by Zalis and fellow Massachusetts General Hospital Mitchell Harris has developed a program called Patient Inference Dossier Queriable. The program combines a search engine with a programming system to automatically extract data from multiple EMRs and several databases and processing information. Although this may seem simple, actually represents a major improvement for physicians, since most EMRs have little or no built-in search capability. They described their system in the August 2010 Journal of the American College of Radiology.

Although Google's Page Rank system works by giving more weight to those pages that are linked more often, EMRs have no ties and, therefore, cannot use this approach. Instead, the system of records has the ability to "learn" certain types of queries from their users, understanding that a search for "squamous cell carcinoma" and another search for "lung cancer" actually looking for the same information.

The program, which is used in a network of hospitals in the Boston area, currently has over 800 registered users and publishes more than 10,000 pages of medical records information daily. Zalis says other major health organizations are starting to use it, but expand their use probably end up bringing difficulties. While the program Dossier is able to integrate with other EMR systems to provide advanced search capabilities, this requires the permission of the various hospitals and medical centers involved and export the appropriate capabilities. Some hospitals report that the EMR system vendors have hindered access to the information contained in the databases from outside their own programs.

And despite improved access to medical information is likely to decrease costs, is something that could actually reduce the income of hospitals and physicians performing diagnostic tests. (Insurers, on the other hand, would benefit because they probably cover less reps). Payne notes that "when it comes to exchange information between organizations, the biggest problem is the alignment of incentives."

References
technologyreview.com

wikipedia.org

Labels: , ,