</> Code with Papers by Syntology

Papers › Long-Tailed Classification of Thorax Diseases on Chest X-Ray: A New Benchmark Study

Long-Tailed Classification of Thorax Diseases on Chest X-Ray: A New Benchmark Study

29 Aug 2022arXiv:2208.13365archive 2025-07-28

Gregory Holste, Song Wang, Ziyu Jiang, Thomas C. Shen, George Shih, Ronald M. Summers, Yifan Peng, Zhangyang Wang

Imaging exams, such as chest radiography, will yield a small set of common findings and a much larger set of uncommon findings. While a trained radiologist can learn the visual presentation of rare conditions by studying a few representative examples, teaching a machine to learn from such a "long-tailed" distribution is much more difficult, as standard methods would be easily biased toward the most frequent classes. In this paper, we present a comprehensive benchmark study of the long-tailed learning problem in the specific domain of thorax diseases on chest X-rays. We focus on learning from naturally distributed chest X-ray data, optimizing classification accuracy over not only the common "head" classes, but also the rare yet critical "tail" classes. To accomplish this, we introduce a challenging new long-tailed chest X-ray benchmark to facilitate research on developing long-tailed learning methods for medical image classification. The benchmark consists of two chest X-ray datasets for 19- and 20-way thorax disease classification, containing classes with as many as 53,000 and as few as 7 labeled training images. We evaluate both standard and state-of-the-art long-tailed learning methods on this new benchmark, analyzing which aspects of these methods are most beneficial for long-tailed medical image classification and summarizing insights for future algorithm design. The datasets, trained models, and code are available at https://github.com/VITA-Group/LongTailCXR.

PaperPDFCode

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

Code

vita-group/longtailcxr officialmentioned in papermentioned on GitHubpytorch 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

ClassificationImage ClassificationLong-tail LearningMedical Image Classificationimage-classification

Datasets

Introduced by this paper, per the archive.

MIMIC-CXR-LTNIH-CXR-LT

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Long-tail Learning MIMIC-CXR-LT Decoupling (cRT) Balanced Accuracy 0.296 #1 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Reweighted LDAM-DRW Balanced Accuracy 0.275 #2 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Class-balanced LDAM-DRW Balanced Accuracy 0.267 #3 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Reweighted LDAM Balanced Accuracy 0.243 #4 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Reweighted Focal Loss Balanced Accuracy 0.239 #5 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Decoupling (tau-norm) Balanced Accuracy 0.230 #6 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Class-balanced Softmax Balanced Accuracy 0.227 #7 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Class-balanced LDAM Balanced Accuracy 0.225 #8 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Reweighted Softmax Balanced Accuracy 0.211 #9 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Class-balanced Focal Loss Balanced Accuracy 0.191 #10 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT MixUp Balanced Accuracy 0.176 #11 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Focal Loss Balanced Accuracy 0.172 #12 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Softmax Balanced Accuracy 0.169 #13 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT Balanced-MixUp Balanced Accuracy 0.168 #14 of 15 Archive leaderboard report
Long-tail Learning MIMIC-CXR-LT LDAM Balanced Accuracy 0.165 #15 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Decoupling (cRT) Balanced Accuracy 0.294 #1 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Reweighted LDAM-DRW Balanced Accuracy 0.289 #2 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Class-balanced LDAM-DRW Balanced Accuracy 0.281 #3 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Reweighted LDAM Balanced Accuracy 0.279 #4 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Class-Balanced Softmax Balanced Accuracy 0.269 #5 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Reweighted Softmax Balanced Accuracy 0.260 #6 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Class-balanced LDAM Balanced Accuracy 0.235 #7 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Class-Balanced Focal Loss Balanced Accuracy 0.232 #8 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Decoupling (tau-norm) Balanced Accuracy 0.214 #9 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Reweighted Focal Loss Balanced Accuracy 0.197 #10 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT LDAM Balanced Accuracy 0.178 #11 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Balanced-MixUp Balanced Accuracy 0.155 #12 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Focal Loss Balanced Accuracy 0.122 #13 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT MixUp Balanced Accuracy 0.118 #14 of 15 Archive leaderboard report
Long-tail Learning NIH-CXR-LT Softmax Balanced Accuracy 0.115 #15 of 15 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