Prediction of postoperative cardiac events in multiple surgical cohorts using a multimodal and integrative decision support system

Abstract

Postoperative patients are at risk of life-threatening complications such as hemodynamic decompensation or arrhythmia. Automated detection of patients with such risks via a real-time clinical decision support system may provide opportunities for early and timely interventions that can significantly improve patient outcomes. We utilize multimodal features derived from digital signal processing techniques and tensor formation, as well as the electronic health record (EHR), to create machine learning models that predict the occurrence of several life-threatening complications up to 4 hours prior to the event. In order to ensure that our models are generalizable across different surgical cohorts, we trained the models on a cardiac surgery cohort and tested them on vascular and non-cardiac acute surgery cohorts. The best performing models achieved an area under the receiver operating characteristic curve (AUROC) of 0.94 on training and 0.94 and 0.82, respectively, on testing for the 0.5-hour interval. The AUROCs only slightly dropped to 0.93, 0.92, and 0.77, respectively, for the 4-hour interval. This study serves as a proof-of-concept that EHR data and physiologic waveform data can be combined to enable the early detection of postoperative deterioration events.

Artificial Intelligence in Cardiac Surgery: A Systematic Review

“…AI applications also affect clinician outcomes, specifically, clinician decision making, clinician workflow and efficiency, and clinician evaluations and acceptance of AI applications. In this review, thirty-two studies reported clinician outcomes of AI in cardiac surgery (4,5,7,8,10,11,13–38).…”

“…Clinicians could potentially be guided by AI applications in making better medical decisions. Eighteen studies reported that AI applications can support clinician decision making (4,7,10,13,16,17,19,21,23,24,26,28,(30)(31)(32)(33)35,38). Machine learning models improve clinician's medical decisions by providing better preoperative risk assessment, stratification and prognostication (10,17,21,24,(30)(31)(32)35,38).…”

Accelerating the integration of ChatGPT and other large‐scale AI models into biomedical research and healthcare

“…• Personalized decision-making: VLMs can be used to create multimodal learning models that predict postoperative deterioration events in surgical intensive care unit patients for precise early intervention by utilizing multimodal features from physiological signals and EHR data. 67,68 • Monitoring patients: VLMs can be used to monitor patients in a remote-monitoring care setting. For example, the integration of data from noninvasive devices such as smartwatches or bands with data from EHRs and other sensors, can be used to improve the reliability of fall detection systems 69 and gait analysis performance.…”

“…VLMs have the potential to be a powerful tool for a wide range of clinical applications. Some examples of the opportunities for these models in this field include: Personalized decision‐making : VLMs can be used to create multimodal learning models that predict postoperative deterioration events in surgical intensive care unit patients for precise early intervention by utilizing multimodal features from physiological signals and EHR data 67,68 Monitoring patients : VLMs can be used to monitor patients in a remote‐monitoring care setting.…”

“…Personalized decision‐making : VLMs can be used to create multimodal learning models that predict postoperative deterioration events in surgical intensive care unit patients for precise early intervention by utilizing multimodal features from physiological signals and EHR data 67,68 …”