Key result
Machine learning models incorporating race reduce pulse oximetry prediction error by ~85%.
Why the study?
Pulse oximetry exhibits racial biases in blood oxygen saturation measurements due to differences in skin color, which can lead to misdiagnosis and improper treatment.
Do machine learning models incorporating race improve the accuracy of pulse oximetry predictions compared to baseline readings?
Observational
Do machine learning models incorporating race improve the accuracy of pulse oximetry predictions compared to baseline readings?
Absolute Event Rate: 4.72% vs 30.76%
Machine learning models incorporating race as a parameter significantly improve the accuracy of pulse oximetry measurements, potentially reducing racial bias in clinical assessments.
Race-inclusive ML models warrant prospective testing for bias reduction; does not yet support clinical adoption.
Pulse oximetry is a widely used non-invasive method for estimating arterial oxygen saturation (SaO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>). However, it has been shown to exhibit racial biases in its blood oxygen saturation (SpO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>) measurements, primarily due to differences in skin color, which can lead to misdiagnosis and improper treatment. This study utilizes the publicly available BOLD dataset on PhysioNet to investigate these biases. It employs machine learning models, including Random Forest, Gradient Boosting, and XGBoost regressors, to predict SpO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> from SpO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> readings, incorporating interpersonal parameters such as race, with an initial focus on the most distinguishable groups by skin color: Black and White. The developed models significantly improve prediction accuracy, reducing the Mean Squared Error from 30.76 (baseline) to 4.72. SHAP analysis underscores the critical role of race as a surrogate for skin tone in enhancing predictions. These findings emphasize the need for equitable signal processing techniques for medical devices to decrease the bias and improve health outcomes across diverse populations.
No takes yet. Share an insight, caveat, or question.
Karli et al. (2024) conducted an observational in Pulse oximetry racial bias. Machine learning models incorporating race vs. Baseline SpO2 readings was evaluated on Prediction accuracy (Mean Squared Error). Machine learning models incorporating race significantly improved pulse oximetry prediction accuracy, reducing the Mean Squared Error from 30.76 at baseline to 4.72.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: