Accelerating MRI Uncertainty Estimation with Mask-based Bayesian Neural Network

TL;DR

This paper introduces a co-optimized algorithm-hardware approach to enhance MRI analysis by transforming IVIM-NET into a Bayesian neural network for reliable uncertainty estimation, achieving significant speedups and power efficiency.

Contribution

It presents a novel transformation of IVIM-NET into a mask-based Bayesian neural network and FPGA-based hardware optimizations for efficient MRI uncertainty estimation.

Findings

Achieved 7.5x speedup over GPU

Achieved 32.5x speedup over CPU

Reduced power consumption significantly

Abstract

Accurate and reliable Magnetic Resonance Imaging (MRI) analysis is particularly important for adaptive radiotherapy, a recent medical advance capable of improving cancer diagnosis and treatment. Recent studies have shown that IVIM-NET, a deep neural network (DNN), can achieve high accuracy in MRI analysis, indicating the potential of deep learning to enhance diagnostic capabilities in healthcare. However, IVIM-NET does not provide calibrated uncertainty information needed for reliable and trustworthy predictions in healthcare. Moreover, the expensive computation and memory demands of IVIM-NET reduce hardware performance, hindering widespread adoption in realistic scenarios. To address these challenges, this paper proposes an algorithm-hardware co-optimization flow for high-performance and reliable MRI analysis. At the algorithm level, a transformation design flow is introduced to convert IVIM-NET to a mask-based Bayesian Neural Network (BayesNN), facilitating reliable and efficient uncertainty estimation. At the hardware level, we propose an FPGA-based accelerator with several hardware optimizations, such as mask-zero skipping and operation reordering. Experimental results demonstrate that our co-design approach can satisfy the uncertainty requirements of MRI analysis, while achieving 7.5 times and 32.5 times speedup on an Xilinx VU13P FPGA compared to GPU and CPU implementations with reduced power consumption.