# An explainable deep learning framework for biosensing data interpretation in biomedical engineering and real-time health diagnostics

## TL;DR

This paper introduces a deep learning framework that interprets biosensing data with high accuracy and transparency for health diagnostics.

## Contribution

The novel PhysioGraph Inference Network (PGIN) and Adaptive Health State Inference Mechanism (AHSIM) combine temporal graph reasoning and probabilistic modeling with physiological priors.

## Key findings

- The framework achieves up to 92.48% diagnostic accuracy and 93.65% AUC on biosensing datasets.
- It outperforms transformer-based models like RoBERTa and T5 in diagnostic performance.
- The model provides transparent uncertainty estimates suitable for clinical and wearable applications.

## Abstract

This work proposes an explainable deep learning framework to transform complex biosignal dynamics into interpretable health assessments. The core of our approach is the PhysioGraph Inference Network (PGIN), which combines temporal graph reasoning with probabilistic modeling to capture dynamic inter sensor dependencies under physiological priors.

To further enhance adaptability, an Adaptive Health State Inference Mechanism (AHSIM) is introduced to adjust diagnostic granularity based on uncertainty and signal entropy.

Evaluations on four biosensing datasets show that our framework achieves superior diagnostic accuracy (up to 92.48%) and AUC (up to 93.65%), outperforming several transformer-based baselines such as RoBERTa and T5. Furthermore, the model provides transparent uncertainty estimates, making it suitable for deployment in clinical and wearable scenarios. By integrating physiological semantics and model interpretability, our framework bridges the gap between black-box AI and trustworthy biomedical intelligence.