Funded by: Afeyan Family Foundation
Co PIs: Aram Butavyan, Varduhi Yeghiazaryan
Project description: A number of recent studies, have indicated the great potential of traditional 3D hyperspectral imaging (HSI) and spectrally enhanced 4D HSI in revealing otherwise invisible surgical targets. However, the high dimensionality and resolution of HSI data still make automated processing and analysis of such data difficult or intractable. Furthermore, while there is an extensive body of literature on the automated segmentation/classification of HSI data in the field of remote sensing, the automated processing of HSI data for biomedical applications remains underexplored. The project aimed to design and implement deep-learning-based procedures for fully automated parallel segmentation and pixel-level classification of 3D and 4D hyperspectral images of atrial tissue undergoing a radiofrequency ablation procedure, incorporating spatio-spectral information from the original image and the segmentation output. The project also aimed to determine the optimal 3D/4D data configuration for achieving the best tissue classification results. The team utilized eight datasets produced at the L.A. Orbeli Institute of Physiology and established an initial experimental setup. Six deep learning models had been chosen for evaluation.