A clinically actionable and explainable real-time risk assessment framework for stroke-associated pneumonia

Abstract

<p>The current <a href="https://www-sciencedirect-com.eproxy.lib.hku.hk/topics/computer-science/medical-practice" title="Learn more about medical practice from ScienceDirect's AI-generated Topic Pages">medical practice</a> is more responsive rather than proactive, despite the widely recognized value of early disease detection, including improving the quality of care and reducing medical costs. One of the cornerstones of early disease detection is clinically actionable predictions, where predictions are expected to be accurate, stable, real-time and interpretable. As an example, we used stroke-associated pneumonia (SAP), setting up a transformer-encoder-based model that analyzes highly heterogeneous <a href="https://www-sciencedirect-com.eproxy.lib.hku.hk/topics/medicine-and-dentistry/electronic-health-record" title="Learn more about electronic health records from ScienceDirect's AI-generated Topic Pages">electronic health records</a> in real-time. The model was proven accurate and stable on an independent test set. In addition, it issued at least one warning for 98.6 % of SAP patients, and on average, its alerts were ahead of physician diagnoses by 2.71 days. We applied Integrated Gradient to glean the model's reasoning process. Supplementing the risk scores, the model highlighted critical historical events on patients' trajectories, which were shown to have high clinical relevance.</p><ul>​​​​​​​</ul>