Trustworthiness of Pretrained Transformers for Lung Cancer Segmentation

arXiv:2403.13113v1 Announce Type: cross
Abstract: We assessed the trustworthiness of two self-supervision pretrained transformer models, Swin UNETR and SMIT, for fine-tuned lung (LC) tumor segmentation using 670 CT and MRI scans. We measured segmentation accuracy on two public 3D-CT datasets, robustness on CT scans of patients with COVID-19, CT scans of patients with ovarian cancer and T2-weighted MRI of men with prostate cancer, and zero-shot generalization of LC for T2-weighted MRIs. Both models demonstrated high accuracy on in-distribution data (Dice …

abstract accuracy arxiv cancer covid covid-19 cs.cv datasets eess.iv lung cancer mri patients public robustness scans segmentation supervision swin transformer transformer models transformers type