Share Email Print
cover

Proceedings Paper

Deep learning for medical image segmentation – using the IBM TrueNorth neurosynaptic system
Author(s): Steven Moran; Bilwaj Gaonkar; William Whitehead; Aidan Wolk; Luke Macyszyn; Subramanian S. Iyer
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Deep convolutional neural networks have found success in semantic image segmentation tasks in computer vision and medical imaging. These algorithms are executed on conventional von Neumann processor architectures or GPUs. This is suboptimal. Neuromorphic processors that replicate the structure of the brain are better-suited to train and execute deep learning models for image segmentation by relying on massively-parallel processing. However, given that they closely emulate the human brain, on-chip hardware and digital memory limitations also constrain them. Adapting deep learning models to execute image segmentation tasks on such chips, requires specialized training and validation.

In this work, we demonstrate for the first-time, spinal image segmentation performed using a deep learning network implemented on neuromorphic hardware of the IBM TrueNorth Neurosynaptic System and validate the performance of our network by comparing it to human-generated segmentations of spinal vertebrae and disks. To achieve this on neuromorphic hardware, the training model constrains the coefficients of individual neurons to {-1,0,1} using the Energy Efficient Deep Neuromorphic (EEDN)1 networks training algorithm. Given the ∼1 million neurons and 256 million synapses, the scale and size of the neural network implemented by the IBM TrueNorth allows us to execute the requisite mapping between segmented images and non-uniform intensity MR images >20 times faster than on a GPU-accelerated network and using <0.1 W. This speed and efficiency implies that a trained neuromorphic chip can be deployed in intra-operative environments where real-time medical image segmentation is necessary.

Paper Details

Date Published: 6 March 2018
PDF: 8 pages
Proc. SPIE 10579, Medical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications, 1057915 (6 March 2018); doi: 10.1117/12.2286419
Show Author Affiliations
Steven Moran, Ctr. for Heterogeneous Integration and Performance Scaling (United States)
Bilwaj Gaonkar, Univ. of California, Los Angeles (United States)
William Whitehead, Ctr. for Heterogeneous Integration and Performance Scaling (United States)
Aidan Wolk, Ctr. for Heterogeneous Integration and Performance Scaling (United States)
Luke Macyszyn, Univ. of California, Los Angeles (United States)
Subramanian S. Iyer, Ctr. for Heterogeneous Integration and Performance Scaling (United States)


Published in SPIE Proceedings Vol. 10579:
Medical Imaging 2018: Imaging Informatics for Healthcare, Research, and Applications
Jianguo Zhang; Po-Hao Chen, Editor(s)

© SPIE. Terms of Use
Back to Top