IBM Watson correctly diagnosis rare form of leukemia

9 August 2016
Recently IBM launched a pilot in which Watson is used to learn how to make a diagnosis based upon images such as scans. Now the AI platform  correctly diagnosed a 60-year-old woman’s rare form of leukemia within 10 minutes. The womans condition was a medical mystery that doctors had missed for months at the University of Tokyo. The treatment based upon the original leukemia diagnosis didn’t work, leading them to plug the patient’s genetic information into Watson’s program for answers.
New York Daily News writes that Watson sifted through 20 million cancer research papers, and came up with the proper diagnosis within 10 minutes, suggesting a new treatment that has since been more effective.

Watsons first steps on the medical front date from 2011, when the platform was put to work at the Columbia University and the University of Maryland School of Medicine. There, Watson was loaded up with medical cases and millions of pages of research papers. The program also went through several sessions with doctors, being corrected whenever it gave a wrong answer. Two years later the platform was made available for doctors and health insurance companies. IBM claims Watson had improved its performance by 240% from 2011 to 2013.

IBM Watson is a technology platform that uses natural language processing and machine learning to reveal insights from large amounts of unstructured data. One of the primary verticals IBM aims at with Watson is the health sector.