CUDLE: Learning Under Label Scarcity to Detect Cannabis Use in Uncontrolled Environments

TL;DR

CUDLE introduces a self-supervised learning framework that effectively detects cannabis use from wearable sensor data in free-living environments, reducing the need for labeled data and outperforming traditional supervised methods.

Contribution

The paper presents a novel self-supervised learning approach for cannabis detection that requires fewer labels and demonstrates improved accuracy over supervised models in real-world settings.

Findings

CUDLE achieves 73.4% accuracy, outperforming supervised methods.

Performance improves as labeled data decreases, especially with 75% fewer labels.

CUDLE reaches peak performance with fewer subjects.

Abstract

Wearable sensor systems have demonstrated a great potential for real-time, objective monitoring of physiological health to support behavioral interventions. However, obtaining accurate labels in free-living environments remains difficult due to limited human supervision and the reliance on self-labeling by patients, making data collection and supervised learning particularly challenging. To address this issue, we introduce CUDLE (Cannabis Use Detection with Label Efficiency), a novel framework that leverages self-supervised learning with real-world wearable sensor data to tackle a pressing healthcare challenge: the automatic detection of cannabis consumption in free-living environments. CUDLE identifies cannabis consumption moments using sensor-derived data through a contrastive learning framework. It first learns robust representations via a self-supervised pretext task with data augmentation. These representations are then fine-tuned in a downstream task with a shallow classifier, enabling CUDLE to outperform traditional supervised methods, especially with limited labeled data. To evaluate our approach, we conducted a clinical study with 20 cannabis users, collecting over 500 hours of wearable sensor data alongside user-reported cannabis use moments through EMA (Ecological Momentary Assessment) methods. Our extensive analysis using the collected data shows that CUDLE achieves a higher accuracy of 73.4%, compared to 71.1% for the supervised approach, with the performance gap widening as the number of labels decreases. Notably, CUDLE not only surpasses the supervised model while using 75% less labels, but also reaches peak performance with far fewer subjects.