Mitigating False Predictions In Unreasonable Body Regions
This addresses a critical issue for clinical deployment of AI in medical imaging by mitigating generalization failures due to field-of-view limitations, though it is an incremental improvement focused on a specific domain.
The paper tackles the problem of false predictions in 3D medical image segmentation when models trained with limited field of view are applied to unseen body regions, proposing a novel loss function that reduces false positive tumor predictions by up to 85% and improves segmentation performance.
Despite considerable strides in developing deep learning models for 3D medical image segmentation, the challenge of effectively generalizing across diverse image distributions persists. While domain generalization is acknowledged as vital for robust application in clinical settings, the challenges stemming from training with a limited Field of View (FOV) remain unaddressed. This limitation leads to false predictions when applied to body regions beyond the FOV of the training data. In response to this problem, we propose a novel loss function that penalizes predictions in implausible body regions, applicable in both single-dataset and multi-dataset training schemes. It is realized with a Body Part Regression model that generates axial slice positional scores. Through comprehensive evaluation using a test set featuring varying FOVs, our approach demonstrates remarkable improvements in generalization capabilities. It effectively mitigates false positive tumor predictions up to 85% and significantly enhances overall segmentation performance.