# FairTune: A Bias-Aware Fine-Tuning Framework Towards Fair Heart Rate Prediction from PPG

arXiv: https://arxiv.org/abs/2509.16491 · HTML: https://mlsanity.com/paper/2509.16491/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 39 references that are in the map.
- First posted on arXiv 20 Sep 2025.

## Builds on (most cited first, 15 of 15)

- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization](https://mlsanity.com/paper/1911.08731/) (2019)
- [Mitigating Unwanted Biases with Adversarial Learning](https://mlsanity.com/paper/1801.07593/) (2018)
- [Exploring the Limits of Large Scale Pre-training](https://mlsanity.com/paper/2110.02095/) (2021)
- [When Foundation Model Meets Federated Learning: Motivations, Challenges, and Future Directions](https://mlsanity.com/paper/2306.15546/) (2023)
- [Foundation Model for Advancing Healthcare: Challenges, Opportunities, and Future Directions](https://mlsanity.com/paper/2404.03264/) (2024)
- [Overwriting Pretrained Bias with Finetuning Data](https://mlsanity.com/paper/2303.06167/) (2023)
- [Multimodal Clinical Benchmark for Emergency Care (MC-BEC): A Comprehensive Benchmark for Evaluating Foundation Models in Emergency Medicine](https://mlsanity.com/paper/2311.04937/) (2023)
- [Erasing the Bias: Fine-Tuning Foundation Models for Semi-Supervised Learning](https://mlsanity.com/paper/2405.11756/) (2024)
- [Personalized Adapter for Large Meteorology Model on Devices: Towards Weather Foundation Models](https://mlsanity.com/paper/2405.20348/) (2024)
- [GPT-PPG: A GPT-based Foundation Model for Photoplethysmography Signals](https://mlsanity.com/paper/2503.08015/) (2025)
- [Fairness and Bias Mitigation in Computer Vision: A Survey](https://mlsanity.com/paper/2408.02464/) (2024)
- [A Scoping Review of Deep Learning Methods for Photoplethysmography Data](https://mlsanity.com/paper/2401.12783/) (2024)
- [Bias Mitigation in Fine-tuning Pre-trained Models for Enhanced Fairness and Efficiency](https://mlsanity.com/paper/2403.00625/) (2024)
- [Improving Machine Learning Based Sepsis Diagnosis Using Heart Rate Variability](https://mlsanity.com/paper/2408.02683/) (2024)

## Built on by (most cited first, 2 of 2)

- [Generalist vs Specialist Time Series Foundation Models: Investigating Potential Emergent Behaviors in Assessing Human Health Using PPG Signals](https://mlsanity.com/paper/2510.14254/) (2025)
- [Wav2Arrest 2.0: Long-Horizon Cardiac Arrest Prediction with Time-to-Event Modeling, Identity-Invariance, and Pseudo-Lab Alignment](https://mlsanity.com/paper/2509.21695/) (2025)

## Abstract (arXiv)

Foundation models pretrained on physiological data such as photoplethysmography (PPG) signals are increasingly used to improve heart rate (HR) prediction across diverse settings. Fine-tuning these models for local deployment is often seen as a practical and scalable strategy. However, its impact on demographic fairness particularly under domain shifts remains underexplored. We fine-tune PPG-GPT a transformer-based foundation model pretrained on intensive care unit (ICU) data across three heterogeneous datasets (ICU, wearable, smartphone) and systematically evaluate the effects on HR prediction accuracy and gender fairness. While fine-tuning substantially reduces mean absolute error (up to 80%), it can simultaneously widen fairness gaps, especially in larger models and under significant distributional characteristics shifts. To address this, we introduce FairTune, a bias-aware fine-tuning framework in which we benchmark three mitigation strategies: class weighting based on inverse group frequency (IF), Group Distributionally Robust Optimization (GroupDRO), and adversarial debiasing (ADV). We find that IF and GroupDRO significantly reduce fairness gaps without compromising accuracy, with effectiveness varying by deployment domain. Representation analyses further reveal that mitigation techniques reshape internal embeddings to reduce demographic clustering. Our findings highlight that fairness does not emerge as a natural byproduct of fine-tuning and that explicit mitigation is essential for equitable deployment of physiological foundation models.