Mask-TS Net: Mask Temperature Scaling Uncertainty Calibration for Polyp Segmentation

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhang, Yudian, Xu, Chenhao, Xu, Kaiye, Zhu, Haijiang
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917662186012672
author Zhang, Yudian
Xu, Chenhao
Xu, Kaiye
Zhu, Haijiang
author_facet Zhang, Yudian
Xu, Chenhao
Xu, Kaiye
Zhu, Haijiang
contents Lots of popular calibration methods in medical images focus on classification, but there are few comparable studies on semantic segmentation. In polyp segmentation of medical images, we find most diseased area occupies only a small portion of the entire image, resulting in previous models being not well-calibrated for lesion regions but well-calibrated for background, despite their seemingly better Expected Calibration Error (ECE) scores overall. Therefore, we proposed four-branches calibration network with Mask-Loss and Mask-TS strategies to more focus on the scaling of logits within potential lesion regions, which serves to mitigate the influence of background interference. In the experiments, we compare the existing calibration methods with the proposed Mask Temperature Scaling (Mask-TS). The results indicate that the proposed calibration network outperforms other methods both qualitatively and quantitatively.
format Preprint
id arxiv_https___arxiv_org_abs_2405_05830
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Mask-TS Net: Mask Temperature Scaling Uncertainty Calibration for Polyp Segmentation
Zhang, Yudian
Xu, Chenhao
Xu, Kaiye
Zhu, Haijiang
Computer Vision and Pattern Recognition
Lots of popular calibration methods in medical images focus on classification, but there are few comparable studies on semantic segmentation. In polyp segmentation of medical images, we find most diseased area occupies only a small portion of the entire image, resulting in previous models being not well-calibrated for lesion regions but well-calibrated for background, despite their seemingly better Expected Calibration Error (ECE) scores overall. Therefore, we proposed four-branches calibration network with Mask-Loss and Mask-TS strategies to more focus on the scaling of logits within potential lesion regions, which serves to mitigate the influence of background interference. In the experiments, we compare the existing calibration methods with the proposed Mask Temperature Scaling (Mask-TS). The results indicate that the proposed calibration network outperforms other methods both qualitatively and quantitatively.
title Mask-TS Net: Mask Temperature Scaling Uncertainty Calibration for Polyp Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.05830