Invited Paper: Feature-to-Classifier Co-Design for Mixed-Signal Smart Flexible Wearables for Healthcare at the Extreme Edge

2025-08-27

AI Summary

Flexible electronics (FE) face critical bottlenecks in wearable healthcare applications—including limited chip area and power budget, low integration density, and large feature sizes—hindering co-optimization of analog front-ends, feature extraction, and classifiers; particularly, ADCs and digital feature processing dominate hardware overhead. This paper proposes an end-to-end mixed-signal co-design framework tailored for extreme-edge scenarios: (1) it pioneers analog-domain feature extraction directly on flexible substrates, eliminating costly ADCs and digital computation; and (2) it introduces a hardware-aware neural architecture search (NAS)-driven mechanism for joint feature selection and classifier optimization. Evaluated on medical benchmark datasets, the framework achieves high recognition accuracy while reducing system area by over 60% and significantly lowering power consumption. It enables disposable, ultra-low-power flexible health monitoring systems.

Technology Category

• Machine Learning: Hardware-aware ML
• Search and Optimization: Mixed Discrete/Continuous Search
• Humans and AI: Brain-Sensing and Analysis

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Flexible Electronics (FE) offer a promising alternative to rigid silicon-based hardware for wearable healthcare devices, enabling lightweight, conformable, and low-cost systems. However, their limited integration density and large feature sizes impose strict area and power constraints, making ML-based healthcare systems-integrating analog frontend, feature extraction and classifier-particularly challenging. Existing FE solutions often neglect potential system-wide solutions and focus on the classifier, overlooking the substantial hardware cost of feature extraction and Analog-to-Digital Converters (ADCs)-both major contributors to area and power consumption. In this work, we present a holistic mixed-signal feature-to-classifier co-design framework for flexible smart wearable systems. To the best of our knowledge, we design the first analog feature extractors in FE, significantly reducing feature extraction cost. We further propose an hardware-aware NAS-inspired feature selection strategy within ML training, enabling efficient, application-specific designs. Our evaluation on healthcare benchmarks shows our approach delivers highly accurate, ultra-area-efficient flexible systems-ideal for disposable, low-power wearable monitoring.

Problem

• Co-designing mixed-signal systems for flexible healthcare wearables
• Reducing hardware costs of feature extraction and ADCs
• Overcoming area and power constraints in flexible electronics

Innovation

• Analog feature extractors in flexible electronics
• Hardware-aware NAS-inspired feature selection strategy
• Holistic mixed-signal feature-to-classifier co-design framework