Irrigated cardiac endocardial lesion assessment using single fiber-multispectral spectroscopy and neural networks

Patients with cardiac arrythmia most commonly require radiofrequency ablation to destroy arrhythmogenic electrical pathways and restore normal heart rhythm. However, arrhythmia resurgence exists from limited means to directly confirm the extent of lesion progression during RFA procedure. Optical spectroscopy is sensitive to tissue optical properties and changes in biomolecular composition. We propose a simplified optical spectroscopy through a single fiber integrated catheter to predict irrigated endocardial lesion progression using deep neural network model on ex-vivo model.