Abstract #3142

Deep Learning approach for Automatic Segmentation and characterization of Traumatic Brain Injury using Multi-parametric MRI

Krishna Kanth Chitta¹, Abdalla Z Mohamed², Fatima Nasrallah², and Bhanu Prakash KN¹

¹Laboratory of Molecular Imaging, Singapore Bioimaging Consortium, Singapore, Singapore, ²Queensland Brain Institute, The University of Queensland, Brisbane, Australia

Automatic and accurate segmentation of Traumatic brain injury is vital to improve assessment of pathophysiology, plan treatment methods and enable large cohort studies. In this work we propose a framework based on 3D CNN and FCM to perform automatic segmentation of whole TBI volume and its sub-regions. The proposed framework utilizes multiple MRI contrasts and has shown high accuracy in delineating injury and sub-regions

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

Join Here