Unraveling Systematic Biases in Brain Segmentation: Insights from Synthetic Training - Archive ouverte HAL Access content directly
Conference Papers Year : 2024

Unraveling Systematic Biases in Brain Segmentation: Insights from Synthetic Training

Abstract

This study examines how the quality of ground truth labels affects brain MRI segmentation models. We investigate the potential of synthetic learning to mitigate systematic biases present in training labels. Through a validation on high-quality datasets, in the Putamen region, known for systematic segmentation errors like the inclusion of parts of the Claustrum, we demonstrate the effectiveness of the synthetic data approach in correcting these errors and enhancing segmentation accuracy. Our findings highlight the limitations of pseudo-ground truth labels derived from automated techniques and underscores the importance of precise, expert-validated labels for accurate, unbiased validation.
No file

Dates and versions

hal-04608723 , version 1 (11-06-2024)

Licence

Identifiers

  • HAL Id : hal-04608723 , version 1

Cite

Romain Valabregue, Ines Khemir, Guillaume Auzias, François Rousseau, Mehdi Ounissi. Unraveling Systematic Biases in Brain Segmentation: Insights from Synthetic Training. MIDL 2024 - Medical Imaging with Deep Learning, Jul 2024, Paris, France. ⟨hal-04608723⟩
9 View
0 Download

Share

Gmail Mastodon Facebook X LinkedIn More