C-GRASP: Clinically-Grounded Reasoning for Affective Signal Processing

TL;DR

C-GRASP is a novel, clinically-grounded reasoning pipeline that enhances HRV interpretation by mitigating physiological hallucinations and incorporating individualized baselines, leading to more transparent and accurate emotion classification.

Contribution

It introduces a guardrailed, reasoning-based framework with a Z-score hierarchy to improve HRV analysis and reduce population bias in affective signal processing.

Findings

Achieved 37.3% accuracy in 4-class emotion classification.

Attained a Clinical Reasoning Consistency score of 69.6%.

Ablation studies highlight the importance of the individualized Delta Z-score module.

Abstract

Heart rate variability (HRV) is a pivotal noninvasive marker for autonomic monitoring; however, applying Large Language Models (LLMs) to HRV interpretation is hindered by physiological hallucinations. These include respiratory sinus arrhythmia (RSA) contamination, short-data instability in nonlinear metrics, and the neglect of individualized baselines in favor of population norms. We propose C-GRASP (Clinically-Grounded Reasoning for Affective Signal Processing), a guardrailed RAG-enhanced pipeline that decomposes HRV interpretation into eight traceable reasoning steps. Central to C-GRASP is a Z-score Priority Hierarchy that enforces the weighting of individualized baseline shifts over normative statistics. The system effectively mitigates spectral hallucinations through automated RSA-aware guardrails, preventing contamination of frequency-domain indices. Evaluated on 414 trials from the DREAMER dataset, C-GRASP integrated with high-scale reasoning models (e.g., MedGemma3-thinking) achieved superior performance in 4-class emotion classification (37.3% accuracy) and a Clinical Reasoning Consistency (CRC) score of 69.6%. Ablation studies confirm that the individualized Delta Z-score module serves as the critical logical anchor, preventing the "population bias" common in native LLMs. Ultimately, C-GRASP transitions affective computing from black-box classification to transparent, evidence-based clinical decision support, paving the way for safer AI integration in biomedical engineering.