DETECT: Data-Driven Evaluation of Treatments Enabled by Classification Transformers

2025-11-10

AI Summary

Chronic pain treatment evaluation relies heavily on subjective self-report scales (e.g., NRS), lacking objective, quantifiable biomarkers. To address this, we propose the first lightweight classification Transformer framework tailored for sensor data collected during daily activities. Our method jointly leverages real-world public datasets and controllably simulated patient behavioral data to end-to-end model temporal features from smartphone multimodal sensors (e.g., accelerometer, gyroscope) before and after intervention, enabling objective functional improvement assessment. Experiments demonstrate that our model achieves significantly higher accuracy than traditional machine learning and temporal CNN baselines, while maintaining low computational overhead (<1.2M parameters). It exhibits strong generalizability across diverse subjects and provides clinically interpretable predictions via attention-based feature attribution. This work establishes a deployable, personalized, and data-driven paradigm for objective, sensor-based evaluation of pain rehabilitation outcomes.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Humans and AI: Brain-Sensing and Analysis
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements

Abstract

Chronic pain is a global health challenge affecting millions of individuals, making it essential for physicians to have reliable and objective methods to measure the functional impact of clinical treatments. Traditionally used methods, like the numeric rating scale, while personalized and easy to use, are subjective due to their self-reported nature. Thus, this paper proposes DETECT (Data-Driven Evaluation of Treatments Enabled by Classification Transformers), a data-driven framework that assesses treatment success by comparing patient activities of daily life before and after treatment. We use DETECT on public benchmark datasets and simulated patient data from smartphone sensors. Our results demonstrate that DETECT is objective yet lightweight, making it a significant and novel contribution to clinical decision-making. By using DETECT, independently or together with other self-reported metrics, physicians can improve their understanding of their treatment impacts, ultimately leading to more personalized and responsive patient care.

Problem

• Assessing chronic pain treatment impact objectively through patient activity analysis
• Overcoming subjectivity limitations of traditional self-reported pain metrics
• Developing data-driven framework using smartphone sensors for clinical evaluation

Innovation

• Uses classification transformers for treatment evaluation
• Compares patient activities before and after treatment
• Analyzes smartphone sensor data from patients