Domain-stratified Training for Cross-organ and Cross-scanner Adenocarcinoma Segmentation in the COSAS 2024 Challenge
This addresses segmentation challenges in medical imaging for adenocarcinoma across varied conditions, though it appears incremental as it builds on existing Upernet methods.
The authors tackled adenocarcinoma segmentation across different organs and scanners by training multiple Upernet-based models with organ-stratified and scanner-stratified approaches and ensembling them, achieving test scores of 0.7643 for Task 1 and 0.8354 for Task 2.
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.