Anthony Bonifonte, instructor in the Data Analytics program, gave two research talks and was a co-author on a third at the Institute for Operations Research and Management Science (INFORMS) annual conference in Houston, Texas, October 21 – 25.
His first talk, “Analytics in Blood Pressure Management and Control: From Data to Decisions,” develops a mathematical framework for making optimal blood pressure treatment decisions to reduce the risk of cardiovascular disease such as heart attacks. The talk shows how the results can be implemented in a decision support tool that clinicians and patients can use to consider treatment alternatives. The results also make suggestions for future randomized control trial design.
His second talk, “Mobile Health for Chronic Disease Management: A Changepoint Detection Approach,” develops statistical and analytical methods for using measurements from mobile health technologies, such as Fitbits, smart watches, and other wearables, to make disease screening decisions. Historically, patients went to their doctor once a year, health measurements such as blood pressure were recorded, and decisions about lifestyle and treatment were made. However, wearable devices have tremendously changed the field: now, users can have their health measurements recorded many times a day, albeit inaccurately. This talk demonstrates how algorithms can make use of these frequent measurements to make recommendations for when users should follow up with their physician.
Finally, he was a co-author on the talk “Prioritizing Hepatitis C Treatment in U.S. Prisons.” Nearly one-third of Hepatitis C infected individuals are incarcerated. Highly effective treatments for Hepatitis C exist, however they are prohibitively expensive, so prisons must make some decisions about who receives treatment on the basis of features such as length of sentence and age of the prisoner. This talk develops a tool to assist prison administrators in making these decisions to maximize both prison and societal benefit, while being ethically fair and unbiased.