Domain-stratified Training for Cross-organ and Cross-scanner Adenocarcinoma Segmentation in the COSAS 2024 Challenge

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Jiayan, Huang, Zheng, Ji, Jinbo, Kuang, Shuoyu, Xu
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929505954693120
author Jiayan, Huang
Zheng, Ji
Jinbo, Kuang
Shuoyu, Xu
author_facet Jiayan, Huang
Zheng, Ji
Jinbo, Kuang
Shuoyu, Xu
contents This manuscript presents an image segmentation algorithm developed for the Cross-Organ and Cross-Scanner Adenocarcinoma Segmentation (COSAS 2024) challenge. We adopted an organ-stratified and scanner-stratified approach to train multiple Upernet-based segmentation models and subsequently ensembled the results. Despite the challenges posed by the varying tumor characteristics across different organs and the differing imaging conditions of various scanners, our method achieved a final test score of 0.7643 for Task 1 and 0.8354 for Task 2. These results demonstrate the adaptability and efficacy of our approach across diverse conditions. Our model's ability to generalize across various datasets underscores its potential for real-world applications.
format Preprint
id arxiv_https___arxiv_org_abs_2409_12418
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Domain-stratified Training for Cross-organ and Cross-scanner Adenocarcinoma Segmentation in the COSAS 2024 Challenge
Jiayan, Huang
Zheng, Ji
Jinbo, Kuang
Shuoyu, Xu
Computer Vision and Pattern Recognition
This manuscript presents an image segmentation algorithm developed for the Cross-Organ and Cross-Scanner Adenocarcinoma Segmentation (COSAS 2024) challenge. We adopted an organ-stratified and scanner-stratified approach to train multiple Upernet-based segmentation models and subsequently ensembled the results. Despite the challenges posed by the varying tumor characteristics across different organs and the differing imaging conditions of various scanners, our method achieved a final test score of 0.7643 for Task 1 and 0.8354 for Task 2. These results demonstrate the adaptability and efficacy of our approach across diverse conditions. Our model's ability to generalize across various datasets underscores its potential for real-world applications.
title Domain-stratified Training for Cross-organ and Cross-scanner Adenocarcinoma Segmentation in the COSAS 2024 Challenge
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2409.12418