Evaluating Drug Efficacy and Patient Outcomes in Personalized Medicine: The Role of AI-Enhanced Neuroimaging and Digital Transformation in Biopharmaceutical Services
Sambasiva Rao Suura, Karthik Chava, Mahesh Recharla, Chaitran Chakilam
Abstract
This article concerns the development of statistical software, in Python and R, that is aimed at evaluating the effectiveness of personalized treatment strategies. This software is a framework for estimating, with quantified uncertainty, how much better a proposed personalized treatment strategy is expected to perform compared to a current strategy for allocating treatments. The framework and software are aimed at researchers or practitioners who have clinical trial data where electronic health records or baseline patient characteristics are sufficient to personalize treatment. Importantly, the software incorporates the last data points of a study and stops the trial when adequate conclusions can be drawn. It serves as a complement to three papers that have explored the theoretical properties of such decision rules.