Preoperative assessment of the risk of adverse complications during surgical anesthesia is the most critical step in ensuring patient safety. Accurate measurement of this risk prior to surgery can reduce the probability of postoperative complications and help identify high-risk individuals, enabling the modification of the surgical anesthesia plan. However, surgical anesthesia requires consideration of a significant amount of clinical information in order to accurately predict the likelihood of postoperative complications. This optimization is crucial for allocating medical resources efficiently and ensuring quality care. Therefore, Chi Mei Hospital utilizes big data in healthcare, incorporating comprehensive preoperative physiological and laboratory values of patients to establish an excellent predictive model and enhance ASA assessment capabilities. This groundbreaking surgical anesthesia risk prediction system achieves an overall accuracy of over 0.8. Currently, we have developed assessments for three types of surgeries: hip, laparotomy, and thoracic procedures. The system provides two application formats. Firstly, it can be integrated into the anesthesia preoperative visit system, providing risk predictions and interactive simulation for individual patients. Secondly, it can be presented as a dashboard on a large screen in the operating room, displaying the risk predictions for all patients scheduled for the three types of surgeries on that day. This AI prediction system outperforms the current ASA scoring method, significantly improving the accuracy of anesthesia risk assessment.

anesthesia risk, personalization, artificial intelligence, machine learning, precision medicine, prediction system, interactive simulation, dashboard