Tyee: A Unified, Modular, and Fully-Integrated Configurable Toolkit for Intelligent Physiological Health Care

2025-12-27

AI Summary

Deep learning for physiological signal analysis faces critical challenges including heterogeneous data formats, inconsistent preprocessing, fragmented model pipelines, and poor experimental reproducibility. To address these, we propose PhysioToolkit—the first unified, configurable toolkit specifically designed for intelligent physiological health analytics. It introduces three key innovations: (1) a multimodal unified data interface; (2) a modular, extensible architecture; and (3) an end-to-end reproducible workflow configuration mechanism, enabling flexible modeling and standardized experimentation across 12 physiological signal types. The toolkit integrates a multimodal preprocessing framework, a configurable deep learning pipeline, and an experiment management module. Evaluated on 13 benchmark datasets, PhysioToolkit matches or surpasses state-of-the-art methods—with SOTA performance achieved on 12 datasets. The open-source implementation is actively maintained, significantly enhancing reproducibility and engineering efficiency in physiological AI research.

Technology Category

• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

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
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Deep learning has shown great promise in physiological signal analysis, yet its progress is hindered by heterogeneous data formats, inconsistent preprocessing strategies, fragmented model pipelines, and non-reproducible experimental setups. To address these limitations, we present Tyee, a unified, modular, and fully-integrated configurable toolkit designed for intelligent physiological healthcare. Tyee introduces three key innovations: (1) a unified data interface and configurable preprocessing pipeline for 12 kinds of signal modalities; (2) a modular and extensible architecture enabling flexible integration and rapid prototyping across tasks; and (3) end-to-end workflow configuration, promoting reproducible and scalable experimentation. Tyee demonstrates consistent practical effectiveness and generalizability, outperforming or matching baselines across all evaluated tasks (with state-of-the-art results on 12 of 13 datasets). The Tyee toolkit is released at https://github.com/SmileHnu/Tyee and actively maintained.

Problem

• Addresses heterogeneous data formats and inconsistent preprocessing in physiological signal analysis
• Solves fragmented model pipelines and non-reproducible experimental setups in deep learning
• Provides a unified toolkit for intelligent healthcare to enhance generalizability and scalability

Innovation

• Unified data interface for 12 signal modalities
• Modular architecture for flexible integration and prototyping
• End-to-end workflow configuration for reproducibility