Streamlining Clinical Evaluation with Explanatory Data Analytics for Adverse Events in Medical Devices

The clinical evaluation process is a continuous and iterative process that aims to monitor the clinical performance and effectiveness of a medical device. Collecting relevant clinical data from various sources is essential for this evaluation, as it helps identify difficulties that are connected to substances or technologies used in medical devices. Also, these reports may indicate potential new or previously unknown risks associated with medical devices. However, as the database of adverse event reports grows on a daily basis, collecting and processing data from multiple adverse event data sources can be challenging and time-consuming. Therefore, it is valuable to automate this process and integrate the data effectively into the clinical evaluation with proper explanations. To tackle this challenge, The study proposes a data analytics framework that is designed to process and categorize adverse event texts efficiently. This framework especially considers the explainability of the results. The evaluation results show that the proposed data analytic ecosystem can efficiently handle the substantial volume of adverse event data, and offer promising explanations for medical image analysis software in the clinical evaluation process.