Article is paywalled, but this story has evolved over the last few weeks: - 2022-01-05: "Scoop: IBM tries to sell Watson Health again" https://www.axios.com/ib

Article is paywalled, but this story has evolved over the last few weeks: - 2022-01-05: "Scoop: IBM tries to sell Watson Health again" https://www.axios.com/ibm-tries-to-sell-watson-health-again-82f691a4-ab81-4b2b-a5bb-13a7556c8ef1.html https://www.axios.com/ibm-tries-to-sell-watson-health-again-... - 2022-01-07: "IBM reportedly shopping Watson Health just as healthcare gets hot" https://techcrunch.com/2022/01/07/ibm-reportedly-shopping-watson-health-just-as-healthcare-gets-hot/ https://techcrunch.com/2022/01/07/ibm-reportedly-shopping-wa... A lot of the hope seemed to be in document summarization from the latest medical literature, plus integrating patient data from electronic medical records. The autopsy of this could be interesting. Some of the critiques against using electronic health records previously was that many of them were designed for medical billing (I don't have a good link, but Eric Topol's "Deep Medicine" has some notes on this problem https://www.basicbooks.com/titles/eric-topol-md/deep-medicine/9781541644649/ https://www.basicbooks.com/titles/eric-topol-md/deep-medicin...).