Deep Learning for Transesophageal Echocardiography View Classification

Kirsten Steffner, Matthew Christensen, George Gill, Michael Bowdish, Justin Rhee, Abirami Kumaresan, Bryan He, James Zou, David Ouyang

Abstract: h3>Abstract</h3> <p>Transesophageal echocardiography (TEE) imaging is a vital monitoring and diagnostic tool used during all major cardiac surgeries, guiding perioperative diagnoses, surgical decision-making, and hemodynamic evaluation in real-time. A key limitation to the automated evaluation of TEE data is the complexity and unstructured nature of the images, which demonstrate significant heterogeneity across varied views in the evaluation of different cardiac structures. In this study, we describe the first machine learning model for TEE view classification. We trained a convolutional neural network (CNN) to predict standardized TEE views using labeled intraoperative and intraprocedural TEE videos from Cedars-Sinai Medical Center (CSMC). We externally validated our model on intraoperative TEE videos from Stanford University Medical Center (SUMC). Accuracy of our model was high across all labeled views. The highest performance was achieved for the Trans-Gastric Left Ventricular Short Axis View (area under the receiver operating curve [AUC] = 0.971 at CSMC, 0.957 at SUMC), the Mid-Esophageal Long Axis View (AUC = 0.954 at CSMC, 0.905 at SUMC), the Mid-Esophageal Aortic Valve Short Axis View (AUC = 0.946 at CSMC, 0.898 at SUMC), and the Mid-Esophageal 4-Chamber View (AUC = 0.939 at CSMC, 0.902 at SUMC). Ultimately, we demonstrate that our unique deep learning model can accurately classify standardized TEE views, which will facilitate further downstream analyses for intraoperative TEE imaging.</p>

External IDs:doi:10.1101/2023.06.11.23290759