Can Foundation Models Predict Fitness for Duty?

2025-07-27

AI Summary

This study addresses the challenge of occupational fitness assessment under reduced alertness caused by alcohol, drug use, or sleep deprivation. We propose the first foundation-model-based framework for analyzing near-infrared iris images to estimate alertness. To overcome the scarcity of large-scale labeled iris data, we employ self-supervised pretraining of a foundation model followed by lightweight downstream fine-tuning, enabling end-to-end occupational fitness prediction. Our work pioneers the integration of foundation models into biometric-driven alertness assessment, effectively mitigating the small-sample bottleneck and enhancing cross-subject and cross-context generalization. Experiments demonstrate that the method achieves superior accuracy and robustness under limited labeled data, substantially reducing reliance on labor-intensive manual annotations. This approach establishes a novel paradigm for unobtrusive, non-contact occupational fitness monitoring.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Humans and AI: Other Foundations of Human Computation & AI
• Machine Learning: Other Foundations of Machine Learning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Security and Privacy: Large-scale security measurements

Abstract

Biometric capture devices have been utilised to estimate a person's alertness through near-infrared iris images, expanding their use beyond just biometric recognition. However, capturing a substantial number of corresponding images related to alcohol consumption, drug use, and sleep deprivation to create a dataset for training an AI model presents a significant challenge. Typically, a large quantity of images is required to effectively implement a deep learning approach. Currently, training downstream models with a huge number of images based on foundational models provides a real opportunity to enhance this area, thanks to the generalisation capabilities of self-supervised models. This work examines the application of deep learning and foundational models in predicting fitness for duty, which is defined as the subject condition related to determining the alertness for work.

Problem

• Predicting fitness for duty using biometric iris images
• Challenges in dataset creation for alcohol, drug, sleep effects
• Leveraging foundation models to improve alertness prediction accuracy

Innovation

• Uses near-infrared iris images for alertness estimation
• Leverages foundational models for downstream training
• Applies self-supervised learning for generalization