Skip to main navigation Skip to search Skip to main content

Bayesian Deep-Learning Processor for Real-Time Bio-Applications with Structured Monte Carlo Dropout for High-Volume Sample Generation

  • Han Gyeol Mun
  • , Jeong Min Woo
  • , Seunghyun Moon
  • , Byungjun Kim
  • , Jongmin Lee
  • , Jae Yoon Sim
  • , Hyunwoo Son
  • Cornell University
  • Gyeongsang National University
  • Pohang University of Science and Technology
  • University of California at San Diego

Research output: Contribution to journal › Article › peer-review

1 Scopus citations

Abstract

This work presents a Bayesian neural network (BNN) processor for real-time, edge-based medical applications, designed to generate high-volume inference outputs efficiently, providing rapid and precise uncertainty estimation. To address the computational and memory-intensive demands of BNNs, we propose a structured Monte Carlo (MC) dropout method and an efficient processing technique for depth-wise separable convolution, thereby minimizing both the memory and computational burdens. The proposed processor, implemented in 28-nm LP CMOS, is benchmarked on a 12-lead ECG dataset and demonstrates significant improvements in energy efficiency, achieving a 12.9× enhancement over conventional MC dropout methods. Furthermore, the processor incorporates a model-switching technique based on uncertainty estimation, resulting in 3.2× lower energy consumption while maintaining robust and accurate ECG classification, even under challenging conditions involving noise and motion artifacts.

Original languageEnglish
Pages (from-to)5459-5472
Number of pages14
JournalIEEE Transactions on Circuits and Systems
Volume73
Issue number8
DOIs
StatePublished - 1 Aug 2026

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 7 - Affordable and Clean Energy
    SDG 7 Affordable and Clean Energy

Keywords

  • Bayesian neural networks
  • ECG signal processing
  • edge applications
  • energy efficiency
  • real-time
  • reconfigurable processing elements
  • structured Monte Carlo dropout
  • uncertainty estimation

Fingerprint

Dive into the research topics of 'Bayesian Deep-Learning Processor for Real-Time Bio-Applications with Structured Monte Carlo Dropout for High-Volume Sample Generation'. Together they form a unique fingerprint.

Cite this