Computing a Second Opinion: Automated Reasoning and Statistical Inference applied to Medical Data

The total volume of electronic medical data around the world is increasing rapidly, allowing for new and exciting applications to change the way we think about care. We set out to find answers to the questions of which aspects of care that involve knowledge sharing can be automated, and how this automation can be performed. Our work focused on automating two aspects of care that traditionally require human supervision. These aspects are generating personalized advice for schizophrenia patients and finding the best vector autoregression model for electronic patient diary data.

We have taken a number of important steps toward the automated processing of electronic medical data. On the patient side, we have shown that providing advice for schizophrenia patients, a delicate task that requires accuracy and precision, can be automated effectively. On the side of the clinicians, we showed that even complex processes, such as finding the optimal vector autoregression model for electronic patient diary data, can be automated.

In order to keep healthcare accessible and affordable in the coming decades, we believe that routine aspects of care that are derivative of electronic medical data will need to be fully automated. Before coming to rely solely on automated reports and recommendations, we envision a transitional period wherein such systems are developed, trained, and used as a second opinion.