Hugging Face Daily Papers · · 6 min read

Disentangling Sampling from Training Budget in Class-Imbalanced CT Body Composition Segmentation

Mirrored from Hugging Face Daily Papers for archival readability. Support the source by reading on the original site.

Class imbalance is a fundamental challenge in medical image segmentation, where frequent classes typically dominate training at the expense of rare classes. Loss-based approaches mitigate imbalance by reweighting the per-pixel loss within the batch, while sampling strategies control which images enter the batch. Yet neither explicitly controls which classes appear within the batch, leaving rare-class exposure only partially rebalanced. In this work, we adopt episodic sampling from few-shot learning to promote class-balanced batch construction in a fully supervised setting. We decouple episodic sampling from its conventional metric-learning context and evaluate it in body composition segmentation in CT. We compare episodic sampling against random and weighted sampling on nine muscle and adipose tissues, derived from 210 scans of the public SAROS dataset. Training is performed under full- and low-data regimes, with additional comparisons under matched training iteration budgets. Under full-data training, all three strategies performed comparably (mean Dice 0.882 for episodic, 0.878 for random and weighted). Under low-data training, episodic sampling outperformed random and weighted (0.787 vs. 0.758 and 0.762), driven by a 12-fold difference in training iterations. Under matched training budgets, random and weighted overfit earlier, while episodic improved for approximately three times more iterations before plateauing. Our findings identify the training iteration budget as under-recognized confound in sampling strategies, motivating iteration-aware evaluation protocols for small datasets. Furthermore, the residual advantage of episodic sampling is consistent with an implicit regularization effect of class-balanced batches, offering a low-cost, model-agnostic strategy for class-imbalanced medical image segmentation. Code is available at <a href=\"https://github.com/iasonsky/episodic-sampling\" rel=\"nofollow\">https://github.com/iasonsky/episodic-sampling</a></p>\n","updatedAt":"2026-05-22T14:08:51.699Z","author":{"_id":"63d93118da4f72339247045b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63d93118da4f72339247045b/1nb8HE-TYCFQb6QB5gTZI.png","fullname":"Dimitris Karkalousos","name":"wdika","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9050236940383911},"editors":["wdika"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/63d93118da4f72339247045b/1nb8HE-TYCFQb6QB5gTZI.png"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2605.20405","authors":[{"_id":"6a10298ca53a61ce2e422f5f","user":{"_id":"6a1010c083c556fdf771096d","avatarUrl":"/avatars/0a18b45380bc9abbde7d44a9f74e5247.svg","isPro":false,"fullname":"Iason Skylitsis","user":"iasonsky","type":"user","name":"iasonsky"},"name":"Iason Skylitsis","status":"claimed_verified","statusLastChangedAt":"2026-05-22T15:59:20.819Z","hidden":false},{"_id":"6a10298ca53a61ce2e422f60","user":{"_id":"63d93118da4f72339247045b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63d93118da4f72339247045b/1nb8HE-TYCFQb6QB5gTZI.png","isPro":false,"fullname":"Dimitris Karkalousos","user":"wdika","type":"user","name":"wdika"},"name":"Dimitrios Karkalousos","status":"claimed_verified","statusLastChangedAt":"2026-05-22T15:59:17.002Z","hidden":false},{"_id":"6a10298ca53a61ce2e422f61","name":"Ivana Išgum","hidden":false}],"publishedAt":"2026-05-19T00:00:00.000Z","submittedOnDailyAt":"2026-05-22T00:00:00.000Z","title":"Disentangling Sampling from Training Budget in Class-Imbalanced CT Body Composition Segmentation","submittedOnDailyBy":{"_id":"63d93118da4f72339247045b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63d93118da4f72339247045b/1nb8HE-TYCFQb6QB5gTZI.png","isPro":false,"fullname":"Dimitris Karkalousos","user":"wdika","type":"user","name":"wdika"},"summary":"Class imbalance is a fundamental challenge in medical image segmentation, where frequent classes typically dominate training at the expense of rare classes. Loss-based approaches mitigate imbalance by reweighting the per-pixel loss within the batch, while sampling strategies control which images enter the batch. Yet neither explicitly controls which classes appear within the batch, leaving rare-class exposure only partially rebalanced. In this work, we adopt episodic sampling from few-shot learning to promote class-balanced batch construction in a fully supervised setting. We decouple episodic sampling from its conventional metric-learning context and evaluate it in body composition segmentation in CT. We compare episodic sampling against random and weighted sampling on nine muscle and adipose tissues, derived from 210 scans of the public SAROS dataset. Training is performed under full- and low-data regimes, with additional comparisons under matched training iteration budgets. Under full-data training, all three strategies performed comparably (mean Dice 0.882 for episodic, 0.878 for random and weighted). Under low-data training, episodic sampling outperformed random and weighted (0.787 vs. 0.758 and 0.762), driven by a 12-fold difference in training iterations. Under matched training budgets, random and weighted overfit earlier, while episodic improved for approximately three times more iterations before plateauing. Our findings identify the training iteration budget as under-recognized confound in sampling strategies, motivating iteration-aware evaluation protocols for small datasets. Furthermore, the residual advantage of episodic sampling is consistent with an implicit regularization effect of class-balanced batches, offering a low-cost, model-agnostic strategy for class-imbalanced medical image segmentation. Code is available at https://github.com/iasonsky/episodic-sampling.","upvotes":1,"discussionId":"6a10298ca53a61ce2e422f62","githubRepo":"https://github.com/iasonsky/episodic-sampling","githubRepoAddedBy":"user","ai_summary":"Episodic sampling from few-shot learning improves class-balanced batch construction in medical image segmentation, outperforming random and weighted sampling under low-data conditions due to reduced overfitting and extended training iterations.","ai_keywords":["class imbalance","medical image segmentation","episodic sampling","few-shot learning","body composition segmentation","CT scans","SAROS dataset","full-data training","low-data training","training iteration budget","Dice score","overfitting","implicit regularization"],"githubStars":0},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6a1010c083c556fdf771096d","avatarUrl":"/avatars/0a18b45380bc9abbde7d44a9f74e5247.svg","isPro":false,"fullname":"Iason Skylitsis","user":"iasonsky","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2605/2605.20405.md"}">
Papers
arxiv:2605.20405

Disentangling Sampling from Training Budget in Class-Imbalanced CT Body Composition Segmentation

Published on May 19
· Submitted by
Dimitris Karkalousos
on May 22

Abstract

Episodic sampling from few-shot learning improves class-balanced batch construction in medical image segmentation, outperforming random and weighted sampling under low-data conditions due to reduced overfitting and extended training iterations.

AI-generated summary

Class imbalance is a fundamental challenge in medical image segmentation, where frequent classes typically dominate training at the expense of rare classes. Loss-based approaches mitigate imbalance by reweighting the per-pixel loss within the batch, while sampling strategies control which images enter the batch. Yet neither explicitly controls which classes appear within the batch, leaving rare-class exposure only partially rebalanced. In this work, we adopt episodic sampling from few-shot learning to promote class-balanced batch construction in a fully supervised setting. We decouple episodic sampling from its conventional metric-learning context and evaluate it in body composition segmentation in CT. We compare episodic sampling against random and weighted sampling on nine muscle and adipose tissues, derived from 210 scans of the public SAROS dataset. Training is performed under full- and low-data regimes, with additional comparisons under matched training iteration budgets. Under full-data training, all three strategies performed comparably (mean Dice 0.882 for episodic, 0.878 for random and weighted). Under low-data training, episodic sampling outperformed random and weighted (0.787 vs. 0.758 and 0.762), driven by a 12-fold difference in training iterations. Under matched training budgets, random and weighted overfit earlier, while episodic improved for approximately three times more iterations before plateauing. Our findings identify the training iteration budget as under-recognized confound in sampling strategies, motivating iteration-aware evaluation protocols for small datasets. Furthermore, the residual advantage of episodic sampling is consistent with an implicit regularization effect of class-balanced batches, offering a low-cost, model-agnostic strategy for class-imbalanced medical image segmentation. Code is available at https://github.com/iasonsky/episodic-sampling.

Community

Paper author Paper submitter about 12 hours ago

Class imbalance is a fundamental challenge in medical image segmentation, where frequent classes typically dominate training at the expense of rare classes. Loss-based approaches mitigate imbalance by reweighting the per-pixel loss within the batch, while sampling strategies control which images enter the batch. Yet neither explicitly controls which classes appear within the batch, leaving rare-class exposure only partially rebalanced. In this work, we adopt episodic sampling from few-shot learning to promote class-balanced batch construction in a fully supervised setting. We decouple episodic sampling from its conventional metric-learning context and evaluate it in body composition segmentation in CT. We compare episodic sampling against random and weighted sampling on nine muscle and adipose tissues, derived from 210 scans of the public SAROS dataset. Training is performed under full- and low-data regimes, with additional comparisons under matched training iteration budgets. Under full-data training, all three strategies performed comparably (mean Dice 0.882 for episodic, 0.878 for random and weighted). Under low-data training, episodic sampling outperformed random and weighted (0.787 vs. 0.758 and 0.762), driven by a 12-fold difference in training iterations. Under matched training budgets, random and weighted overfit earlier, while episodic improved for approximately three times more iterations before plateauing. Our findings identify the training iteration budget as under-recognized confound in sampling strategies, motivating iteration-aware evaluation protocols for small datasets. Furthermore, the residual advantage of episodic sampling is consistent with an implicit regularization effect of class-balanced batches, offering a low-cost, model-agnostic strategy for class-imbalanced medical image segmentation. Code is available at https://github.com/iasonsky/episodic-sampling

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images

· Sign up or log in to comment

Get this paper in your agent:

hf papers read 2605.20405
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2605.20405 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2605.20405 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2605.20405 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

Discussion (0)

Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.

Sign in →

No comments yet. Sign in and be the first to say something.

More from Hugging Face Daily Papers