In his post on the matter, Upbin points out that Watson has been training to be a health care diagnostician. While he may not be a candidate for Fox’s Dr. House’s elite fellowship yet (the producers should definitely invite Watson to be on the show), Watson is on the road to changing healthcare diagnosis in the next five to 10 years.
Cramming for Boards
The DeepQA software developers at IBM have been working for the past year to refine how Watson collects and analyzes health data. For instance, the human team (as is required in data analytics) headed by Dr. Eliot Siegal, a senior radiologist and vice chair of informatics at the University of Maryland, provided IBM with a list of the most important texts and journals for Watson to read. Much like a college student (but at a much accelerated pace), Watson hit the books (dozens of them) plus he absorbed the knowledge of the Medline and PubMed databases.
Also like a student, Watson is taking board exams by asking and answering the questions. He’s shown the same over-time improvement that he did in his rise to Jeopardy fame, reports Upbin.
Diagnostician Residency in Watson’s Future
Watson’s team will soon put the scary smart computer through a residency of “anonymized patient records so it can marry what it knows about diagnostics with the procedures, treatment and outcomes that follow.” The end goal is to aid doctors in treating real patients, especially in rural or underserved areas.
Analytics Behind What Watson Can Read
Another report we found says that Watson will soon be working past “natural language queries and advanced text analytics” to learn from more visual and auditory forms of communication – recordings and video. According to Michael Vizard’s blog at ITBusinessEdge, Watson’s development team “relies heavily” on the open source Unstructured Information Management Architecture (UIMA). The software will soon be supporting video and audio searches, which will allow Watson to search through “troves of audio and video files.”
Mastering the Bedside Manner?
Lastly, Watson is working on his conversational skills – asking questions – a key skill in data analytics. Right now, researchers are “fine-tuning the original query by comparing questions based on previous similar interactions.” This practice will allow Watson to pose more follow-up questions. The longer-term goal is to give Watson the ability to ask more spontaneous questions as he collects data, writes Vizard.
Watson is a key example of how important the human element and team collaboration really is in the field of data analytics. To learn more about the human element and collaboration as it relates to data analytics, check out our recent webcast on TIBCO Silver Spotfire.
Spotfire Blogging Team
photo courtesy of singularityhub.com