paper-with-me

홈 › Papers

ASLseg: Adapting SAM in the Loop for Semi-supervised Liver Tumor Segmentation

2023-12-13 · Shiyun Chen, Li Lin, Pujin Cheng, Xiaoying Tang

Liver tumor segmentation is essential for computer-aided diagnosis, surgical planning, and prognosis evaluation. However, obtaining and maintaining a large-scale dataset with dense annotations is challenging. Semi-Supervised Learning (SSL) is a common technique to address these challenges. Recently, Segment Anything Model (SAM) has shown promising performance in some medical image segmentation tasks, but it performs poorly for liver tumor segmentation. In this paper, we propose a novel semi-supervised framework, named ASLseg, which can effectively adapt the SAM to the SSL setting and combine both domain-specific and general knowledge of liver tumors. Specifically, the segmentation model trained with a specific SSL paradigm provides the generated pseudo-labels as prompts to the fine-tuned SAM. An adaptation network is then used to refine the SAM-predictions and generate higher-quality pseudo-labels. Finally, the reliable pseudo-labels are selected to expand the labeled set for iterative training. Extensive experiments on the LiTS dataset demonstrate overwhelming performance of our ASLseg.

📄 PDF Abstract BibTeX arXiv:2312.07969

Code (0)

등록된 구현이 없습니다.

Tasks

General KnowledgeImage SegmentationMedical Image SegmentationPrognosisSegmentationSemantic SegmentationTumor Segmentation

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically
SAM 설명 없음

Similar Papers 제목 키워드 기반

Harmonizing Generalization and Specialization: Uncertainty-Informed Collaborative Learning for Semi-supervised Medical Image Segmentation

2025-12-15 · Wenjing Lu, Yi Hong, Yang Yang arxiv

Vision foundation models have demonstrated strong generalization in medical image segmentation by leveraging large-scale, heterogeneous pretraining. However, they often struggle to generalize to specialized clinical task…

Semi-supervised Medical Image SegmentationGeneral Knowledge

Semi-Supervised Active Learning for Semi-Supervised Models: Exploit Adversarial Examples With Graph-Based Virtual Labels

2021-01-01 · ICCV 2021 10 · Jiannan Guo, Haochen Shi, Yangyang Kang, Kun Kuang 외

The performance of computer vision models significantly improves with more labeled data. However, the acquisition of labeled data is limited by the high cost. To mitigate the reliance on large labeled datasets, activ…

Active Learning

Semi-Supervised Learning-Based Genetic Biomarkers Dataset for Multiple-Stage Hepatocellular Carcinoma Prediction

2026-09-15 · Ahmed Ammar Kubba, Manar Abu Talib, Jibran Sualeh Muhammad, Ali Bou Nassif 외 arxiv

Liver cancer is a complex disease responsible for a high number of deaths across the globe each year, making automated solutions for liver cancer classification urgent. The most common form of liver cancer is hepatocellu…

Cancer Classification

From Specialist to Generalist: Unlocking SAM's Learning Potential on Unlabeled Medical Images

2026-01-25 · Vi Vu, Thanh-Huy Nguyen, Tien-Thinh Nguyen, Ba-Thinh Lam 외 arxiv

Foundation models like the Segment Anything Model (SAM) show strong generalization, yet adapting them to medical images remains difficult due to domain shift, scarce labels, and the inability of Parameter-Efficient Fine-…

parameter-efficient fine-tuningMedical Image SegmentationPolyp Segmentation

Twin Neural Network Regression is a Semi-Supervised Regression Algorithm

2021-06-11 · Sebastian J. Wetzel, Roger G. Melko, Isaac Tamblyn

Twin neural network regression (TNNR) is a semi-supervised regression algorithm, it can be trained on unlabelled data points as long as other, labelled anchor data points, are present. TNNR is trained to predict differen…

regression