Comparator Loss: An Ordinal Contrastive Loss to Derive a Severity Score for Speech-based Health Monitoring

2025-09-22

AI Summary

This study addresses the limitation of existing binary classification or single-clinical-score prediction models for neurodegenerative diseases—such as motor neuron disease (MND)—by developing a continuous severity scoring model directly from speech signals. We propose a novel comparator loss function, introducing ordinal contrastive learning to voice-based health monitoring for the first time. Leveraging multi-source ordinal supervision—including diagnostic labels, ALS Functional Rating Scale–Revised (ALSFRS-R) scores, and recording timestamps—we enable end-to-end training. The method effectively mitigates information sparsity in small-scale clinical speech datasets. It achieves high accuracy in distinguishing patients from controls while producing a dynamic severity score strongly correlated with clinical assessments (Pearson’s *r* > 0.85). Crucially, the score robustly tracks longitudinal disease progression, offering a scalable, objective, and clinically interpretable biomarker for MND monitoring.

Technology Category

• Natural Language Processing: Speech
• Machine Learning: Multimodal Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

Monitoring the progression of neurodegenerative disease has important applications in the planning of treatment and the evaluation of future medications. Whereas much of the state-of-the-art in health monitoring from speech has been focused on classifying patients versus healthy controls, or predicting real-world health metrics, we propose here a novel measure of disease progression: the severity score. This score is derived from a model trained to minimize what we call the comparator loss. The comparator loss ensures scores follow an ordering relation, which can be based on diagnosis, clinically annotated scores, or simply the chronological order of the recordings. In addition to giving a more detailed picture than a simple discrete classification, the proposed comparator loss-based system has the potential to incorporate information from disparate health metrics, which is critical for making full use of small health-related datasets. We evaluated our proposed models based on their ability to affirmatively track the progression of patients with motor neuron disease (MND), the correlation of their output with clinical annotations such as ALSFRS-R, as well as their ability to distinguish between subjects with MND and healthy controls.

Problem

• Deriving severity scores for neurodegenerative disease progression monitoring from speech
• Establishing ordinal relationships between patient states using comparator loss function
• Tracking motor neuron disease progression and correlating with clinical assessment metrics

Innovation

• Ordinal contrastive loss for severity scoring
• Comparator loss enforces temporal ordering relations
• Integrates disparate health metrics from small datasets