Papers › CaraNet: Context Axial Reverse Attention Network for Segmentation of Small Medical Objects

CaraNet: Context Axial Reverse Attention Network for Segmentation of Small Medical Objects

16 Aug 2021arXiv:2108.07368archive 2025-07-28

Ange Lou, Shuyue Guan, Hanseok Ko, Murray Loew

Segmenting medical images accurately and reliably is important for disease diagnosis and treatment. It is a challenging task because of the wide variety of objects' sizes, shapes, and scanning modalities. Recently, many convolutional neural networks (CNN) have been designed for segmentation tasks and achieved great success. Few studies, however, have fully considered the sizes of objects, and thus most demonstrate poor performance for small objects segmentation. This can have a significant impact on the early detection of diseases. This paper proposes a Context Axial Reserve Attention Network (CaraNet) to improve the segmentation performance on small objects compared with several recent state-of-the-art models. We test our CaraNet on brain tumor (BraTS 2018) and polyp (Kvasir-SEG, CVC-ColonDB, CVC-ClinicDB, CVC-300, and ETIS-LaribPolypDB) segmentation datasets. Our CaraNet achieves the top-rank mean Dice segmentation accuracy, and results show a distinct advantage of CaraNet in the segmentation of small medical objects.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

AngeLouCN/CaraNet officialmentioned in paperpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Medical Image SegmentationSegmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Medical Image Segmentation CVC-ClinicDB CaraNet Average MAE 0.007 #25 of 48 Archive leaderboard report
Medical Image Segmentation CVC-ClinicDB CaraNet S-Measure 0.954 #25 of 48 Archive leaderboard report
Medical Image Segmentation CVC-ClinicDB CaraNet mIoU 0.887 #25 of 48 Archive leaderboard report
Medical Image Segmentation CVC-ClinicDB CaraNet max E-Measure 0.991 #25 of 48 Archive leaderboard report
Medical Image Segmentation CVC-ClinicDB CaraNet mean Dice 0.936 #25 of 48 Archive leaderboard report
Medical Image Segmentation CVC-ColonDB CaraNet Average MAE 0.042 #18 of 25 Archive leaderboard report
Medical Image Segmentation CVC-ColonDB CaraNet S-Measure 0.853 #18 of 25 Archive leaderboard report
Medical Image Segmentation CVC-ColonDB CaraNet mIoU 0.689 #18 of 25 Archive leaderboard report
Medical Image Segmentation CVC-ColonDB CaraNet max E-Measure 0.902 #18 of 25 Archive leaderboard report
Medical Image Segmentation CVC-ColonDB CaraNet mean Dice 0.773 #18 of 25 Archive leaderboard report
Medical Image Segmentation ETIS-LARIBPOLYPDB CaraNet Average MAE 0.017 #15 of 25 Archive leaderboard report
Medical Image Segmentation ETIS-LARIBPOLYPDB CaraNet S-Measure 0.868 #15 of 25 Archive leaderboard report
Medical Image Segmentation ETIS-LARIBPOLYPDB CaraNet mIoU 0.672 #15 of 25 Archive leaderboard report
Medical Image Segmentation ETIS-LARIBPOLYPDB CaraNet max E-Measure 0.894 #15 of 25 Archive leaderboard report
Medical Image Segmentation ETIS-LARIBPOLYPDB CaraNet mean Dice 0.747 #15 of 25 Archive leaderboard report
Medical Image Segmentation Kvasir-SEG CaraNet Average MAE 0.023 #25 of 58 Archive leaderboard report
Medical Image Segmentation Kvasir-SEG CaraNet S-Measure 0.929 #25 of 58 Archive leaderboard report
Medical Image Segmentation Kvasir-SEG CaraNet mIoU 0.865 #25 of 58 Archive leaderboard report
Medical Image Segmentation Kvasir-SEG CaraNet max E-Measure 0.968 #25 of 58 Archive leaderboard report
Medical Image Segmentation Kvasir-SEG CaraNet mean Dice 0.918 #25 of 58 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections