As we mentioned in a previous post, recently at the ASTMH conference, we held a kickoff meeting for a project titled, “Development of a Bayesian diagnostic for etiological management of pediatric diarrhea.” This project is funded by the Bill & Melinda Gates Foundation, spearheaded by their Innovative Technology Solutions team. Our overarching goal is to develop and pilot a Smartphone App that will help clinicians in LMICs better manage kids with diarrhea. The “Bayesian” part is that this App would incorporate clinical, epidemiological, and environmental “priors”, such as climate data and clinical features of previous patients, into a clinical prediction App. For example, the App would ask the clinician to enter certain symptoms that the patient came with, computationally combine this information with cloud-based “priors” data (e.g. recent weather patterns, aggregate data from previous patients), and output the probability that a certain patient’s diarrhea is caused by a virus vs. bacteria. The clinician can then make a more informed decision about whether to give the patient an antibiotic or not. After we develop and validate the App, we will pilot it at two international sites. We are thankful for the expertise of our collaborators at Brown University, University of Florida, and University of Maryland, who are all part of this grant, as well as input from those at Institute for Disease Modeling and University of Virginia. In the end, we are really excited to have the opportunity to translate some of our current analytical efforts deriving prediction rules into an application that clinicians can use!
