# Mitigating Biases in Surgical Operating Rooms with Geometry

arXiv: https://arxiv.org/abs/2508.08028 · HTML: https://mlsanity.com/paper/2508.08028/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 16 references that are in the map.
- First posted on arXiv 11 Aug 2025.

## Builds on (most cited first, 12 of 12)

- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness](https://mlsanity.com/paper/1811.12231/) (2018)
- [In Defense of the Triplet Loss for Person Re-Identification](https://mlsanity.com/paper/1703.07737/) (2017)
- [OpenGait: Revisiting Gait Recognition Toward Better Practicality](https://mlsanity.com/paper/2211.06597/) (2022)
- [LidarGait: Benchmarking 3D Gait Recognition with Point Clouds](https://mlsanity.com/paper/2211.10598/) (2022)
- [MVOR: A Multi-view RGB-D Operating Room Dataset for 2D and 3D Human Pose Estimation](https://mlsanity.com/paper/1808.08180/) (2018)
- [4D-OR: Semantic Scene Graphs for OR Domain Modeling](https://mlsanity.com/paper/2203.11937/) (2022)
- [Deep learning-based person re-identification methods: A survey and outlook of recent works](https://mlsanity.com/paper/2110.04764/) (2021)
- [DisguisOR: Holistic Face Anonymization for the Operating Room](https://mlsanity.com/paper/2307.14241/) (2023)
- [Know your sensORs -- A Modality Study For Surgical Action Classification](https://mlsanity.com/paper/2203.08674/) (2022)
- [Beyond Role-Based Surgical Domain Modeling: Generalizable Re-Identification in the Operating Room](https://mlsanity.com/paper/2503.13028/) (2025)

## Abstract (arXiv)

Deep neural networks are prone to learning spurious correlations, exploiting dataset-specific artifacts rather than meaningful features for prediction. In surgical operating rooms (OR), these manifest through the standardization of smocks and gowns that obscure robust identifying landmarks, introducing model bias for tasks related to modeling OR personnel. Through gradient-based saliency analysis on two public OR datasets, we reveal that CNN models succumb to such shortcuts, fixating on incidental visual cues such as footwear beneath surgical gowns, distinctive eyewear, or other role-specific identifiers. Avoiding such biases is essential for the next generation of intelligent assistance systems in the OR, which should accurately recognize personalized workflow traits, such as surgical skill level or coordination with other staff members. We address this problem by encoding personnel as 3D point cloud sequences, disentangling identity-relevant shape and motion patterns from appearance-based confounders. Our experiments demonstrate that while RGB and geometric methods achieve comparable performance on datasets with apparent simulation artifacts, RGB models suffer a 12% accuracy drop in realistic clinical settings with decreased visual diversity due to standardizations. This performance gap confirms that geometric representations capture more meaningful biometric features, providing an avenue to developing robust methods of modeling humans in the OR.