An interpretable deep learning model predicted intraoperative hypotension 10 minutes in advance with high accuracy, achieving an AUROC of 0.9145 in internal and 0.9035 in external validation.
Does an interpretable deep learning model accurately predict hypotension occurrence 10 minutes in advance in anesthetized patients?
An interpretable deep learning model can accurately predict intraoperative hypotension 10 minutes in advance based on 90-second arterial blood pressure records, providing physiological interpretability.
Effect estimate: AUROC 0.9145 (internal) and 0.9035 (external)
The monitoring of arterial blood pressure (ABP) in anesthetized patients is crucial for preventing hypotension, which can lead to adverse clinical outcomes. Several efforts have been devoted to develop artificial intelligence-based hypotension prediction indices. However, the use of such indices is limited because they may not provide a compelling interpretation of the association between the predictors and hypotension. Herein, an interpretable deep learning model is developed that forecasts hypotension occurrence 10 min before a given 90-s ABP record. Internal and external validations of the model performance show the area under the receiver operating characteristic curves of 0.9145 and 0.9035, respectively. Furthermore, the hypotension prediction mechanism can be physiologically interpreted using the predictors automatically generated from the proposed model for representing ABP trends. Finally, the applicability of a deep learning model with high accuracy is demonstrated, thus providing an interpretation of the association between ABP trends and hypotension in clinical practice.
Hwang et al. (Tue,) conducted a other in Intraoperative hypotension. Interpretable deep learning model was evaluated on Hypotension occurrence 10 min before a given 90-s ABP record (AUROC 0.9145 (internal) and 0.9035 (external)). An interpretable deep learning model predicted intraoperative hypotension 10 minutes in advance with high accuracy, achieving an AUROC of 0.9145 in internal and 0.9035 in external validation.