PCA-YOLO: A Small Liver Tumor Detection Model with Patch-Contrastive Attention

Xueyang Li, Han Xiao, Zongpeng Weng, Xinrong Hu, Danny Chen, Yiyu Shi
Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, PMLR 301:994-1007, 2026.

Abstract

Liver tumors, as one of the most common malignant tumor types, represent a significant clinical challenge, with the detection of small tumors being particularly problematic. Despite the rapid advances in deep learning (DL) offering significant support in reducing the workload of radiologists, current detection models still struggle with the detection of small tumors. This is particularly troubling as these are the cases where even experienced radiologists are more prone to errors, underscoring the critical need for improved accuracy of detection methods in this area. Addressing this critical gap, this article introduces patch-contrastive attention YOLO (PCA-YOLO), an innovative adaptation of the YOLO framework, incorporating a patch-based attention module to specifically target the detection of small liver tumors. Furthermore, we collected a specialized CT dataset focusing exclusively on small liver tumors, complemented with meticulously annotated bounding boxes, to facilitate this study. Our experimental findings demonstrate that our approach achieves a leading mean Average Precision (mAP) score of 77.2% at a 50% Intersection Over Union (IoU) threshold, surpassing all current leading detection methods tested against our specialized dataset.

Cite this Paper

BibTeX
@InProceedings{pmlr-v301-li26b, title = {PCA-YOLO: A Small Liver Tumor Detection Model with Patch-Contrastive Attention}, author = {Li, Xueyang and Xiao, Han and Weng, Zongpeng and Hu, Xinrong and Chen, Danny and Shi, Yiyu}, booktitle = {Proceedings of The 8th International Conference on Medical Imaging with Deep Learning}, pages = {994--1007}, year = {2026}, editor = {Tasdizen, Tolga and Elhabian, Shireen and Summers, Ronald and Chen, Chen and Koch, Lisa and Zhuang, Yan}, volume = {301}, series = {Proceedings of Machine Learning Research}, month = {09--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v301/main/assets/li26b/li26b.pdf}, url = {https://proceedings.mlr.press/v301/li26b.html}, abstract = {Liver tumors, as one of the most common malignant tumor types, represent a significant clinical challenge, with the detection of small tumors being particularly problematic. Despite the rapid advances in deep learning (DL) offering significant support in reducing the workload of radiologists, current detection models still struggle with the detection of small tumors. This is particularly troubling as these are the cases where even experienced radiologists are more prone to errors, underscoring the critical need for improved accuracy of detection methods in this area. Addressing this critical gap, this article introduces patch-contrastive attention YOLO (PCA-YOLO), an innovative adaptation of the YOLO framework, incorporating a patch-based attention module to specifically target the detection of small liver tumors. Furthermore, we collected a specialized CT dataset focusing exclusively on small liver tumors, complemented with meticulously annotated bounding boxes, to facilitate this study. Our experimental findings demonstrate that our approach achieves a leading mean Average Precision (mAP) score of 77.2% at a 50% Intersection Over Union (IoU) threshold, surpassing all current leading detection methods tested against our specialized dataset.} }
Endnote
%0 Conference Paper %T PCA-YOLO: A Small Liver Tumor Detection Model with Patch-Contrastive Attention %A Xueyang Li %A Han Xiao %A Zongpeng Weng %A Xinrong Hu %A Danny Chen %A Yiyu Shi %B Proceedings of The 8th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Tolga Tasdizen %E Shireen Elhabian %E Ronald Summers %E Chen Chen %E Lisa Koch %E Yan Zhuang %F pmlr-v301-li26b %I PMLR %P 994--1007 %U https://proceedings.mlr.press/v301/li26b.html %V 301 %X Liver tumors, as one of the most common malignant tumor types, represent a significant clinical challenge, with the detection of small tumors being particularly problematic. Despite the rapid advances in deep learning (DL) offering significant support in reducing the workload of radiologists, current detection models still struggle with the detection of small tumors. This is particularly troubling as these are the cases where even experienced radiologists are more prone to errors, underscoring the critical need for improved accuracy of detection methods in this area. Addressing this critical gap, this article introduces patch-contrastive attention YOLO (PCA-YOLO), an innovative adaptation of the YOLO framework, incorporating a patch-based attention module to specifically target the detection of small liver tumors. Furthermore, we collected a specialized CT dataset focusing exclusively on small liver tumors, complemented with meticulously annotated bounding boxes, to facilitate this study. Our experimental findings demonstrate that our approach achieves a leading mean Average Precision (mAP) score of 77.2% at a 50% Intersection Over Union (IoU) threshold, surpassing all current leading detection methods tested against our specialized dataset.
APA
Li, X., Xiao, H., Weng, Z., Hu, X., Chen, D. & Shi, Y.. (2026). PCA-YOLO: A Small Liver Tumor Detection Model with Patch-Contrastive Attention. Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 301:994-1007 Available from https://proceedings.mlr.press/v301/li26b.html.

Related Material