Meeting Banner
Abstract #1662

Cine Image Based Cardiac Disease Classification Using Random Forest Classifier and Graph Based Deep Learning Approach

Akos Varga-Szemes1, Teodora Chitiboi2, U. Joseph Schoepf1, Athira J Jacob2, Sayan Ghosal3, Fei Xiong4, Puneet Sharma2, Jonathan Aldinger1, and Tilman Emrich1
1Medical University of South Carolina, Charleston, SC, United States, 2Siemens Healthcare, Princeton, NJ, United States, 3Johns Hopkins University, Baltimore, MD, United States, 4Siemens Healthcare, Charleston, SC, United States


Existing work demonstrates the value of image-based cardiovascular diagnosis with AI. We aimed to develop and test a machine learning algorithm for cardiovascular disease classification based on cine image datasets of 570 consecutive patients. Disease classification was performed using a random forest (RF) classifier and a disease classification network based on graph attention networks. The fully automated deep learning algorithm showed high accuracy for cardiac disease classification based on cine images only. Such algorithm has the potential to improve the efficiency of the reading process, especially by identifying and filtering out patients with normal cardiac anatomy and function.

This abstract and the presentation materials are available to members only; a login is required.

Join Here