PocketNet: A Smaller Neural Network for Medical Image Analysis

Abstract

Medical imaging deep learning models are often large and complex, requiring specialized hardware to train and evaluate these models. To address such issues, we propose the PocketNet paradigm to reduce the size of deep learning models by throttling the growth of the number of channels in convolutional neural networks. We demonstrate that, for a range of segmentation and classification tasks, PocketNet architectures produce results comparable to that of conventional neural networks while reducing the number of parameters by multiple orders of magnitude, using up to 90% less GPU memory, and speeding up training times by up to 40%, thereby allowing such models to be trained and deployed in resource-constrained settings.

References in corpus (10)

• Adam: A Method for Stochastic Optimization
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Automated Design of Deep Learning Methods for Biomedical Image Segmentation
• Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentation
• Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations
• U-Net and its variants for medical image segmentation: theory and applications
• The Liver Tumor Segmentation Benchmark (LiTS)
• A Convergence Theory for Deep Learning via Over-Parameterization
• X-Net: Brain Stroke Lesion Segmentation Based on Depthwise Separable Convolution and Long-range Dependencies
• MgNet: A Unified Framework of Multigrid and Convolutional Neural Network

Cited by in corpus (1)

• Evolving Horizons in Radiotherapy Auto-Contouring: Distilling Insights, Embracing Data-Centric Frameworks, and Moving Beyond Geometric Quantification