mlsanityData to 7 Oct 2026

Warm Start Active Learning with Proxy Labels & Selection via Semi-Supervised Fine-Tuning

arXiv 2209.06285 ↗First posted 13 Sep 2022cs.CVcited by 21

Not on the map. Why?

Abstract (arXiv)

Which volume to annotate next is a challenging problem in building medical imaging datasets for deep learning. One of the promising methods to approach this question is active learning (AL). However, AL has been a hard nut to crack in terms of which AL algorithm and acquisition functions are most useful for which datasets. Also, the problem is exacerbated with which volumes to label first when there is zero labeled data to start with. This is known as the cold start problem in AL. We propose two novel strategies for AL specifically for 3D image segmentation. First, we tackle the cold start problem by proposing a proxy task and then utilizing uncertainty generated from the proxy task to rank the unlabeled data to be annotated. Second, we craft a two-stage learning framework for each active iteration where the unlabeled data is also used in the second stage as a semi-supervised fine-tuning strategy. We show the promise of our approach on two well-known large public datasets from medical segmentation decathlon. The results indicate that the initial selection of data and semi-supervised framework both showed significant improvement for several AL strategies.

Key facts. Cited by 21 papers in the map (data through 7 Oct 2026). Builds on 23 of its 33 references that are in the map. First posted on arXiv 13 Sep 2022.

Builds on

23 of 33 references.

References · 23

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  4. 2017Deep Bayesian Active Learning with Image Datacited by 812
  5. 2019Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Boundscited by 602
  6. 2021Self-Supervised Pre-Training of Swin Transformers for 3D Medical Image Analysiscited by 418
  7. 2019Uncertainty-aware Self-ensembling Model for Semi-supervised 3D Left Atrium Segmentationcited by 378
  8. 2019A large annotated medical image dataset for the development and evaluation of segmentation algorithmscited by 371
  9. 2020Shape-aware Semi-supervised 3D Semantic Segmentation for Medical Imagescited by 152
  10. 2017Suggestive Annotation: A Deep Active Learning Framework for Biomedical Image Segmentationcited by 145
  11. 2020Self-Supervision with Superpixels: Training Few-shot Medical Image Segmentation without Annotationcited by 128
  12. 2020Cold-start Active Learning through Self-supervised Language Modelingcited by 128
11 more
  1. 2020Annotation-efficient deep learning for automatic medical image segmentationcited by 79
  2. 20183D Semi-Supervised Learning with Uncertainty-Aware Multi-View Co-Trainingcited by 57
  3. 2021DiNTS: Differentiable Neural Network Topology Search for 3D Medical Image Segmentationcited by 47
  4. 2021Diminishing Uncertainty within the Training Pool: Active Learning for Medical Image Segmentationcited by 44
  5. 2021Reducing Label Effort: Self-Supervised meets Active Learningcited by 42
  6. 2019Rethinking deep active learning: Using unlabeled data at model trainingcited by 39
  7. 2018Cost-Sensitive Active Learning for Intracranial Hemorrhage Detectioncited by 28
  8. 2018Large-Scale Visual Active Learning with Deep Probabilistic Ensemblescited by 13
  9. 2020Semi-supervised Active Learning for Instance Segmentation via Scoring Predictionscited by 7
  10. 2021When Deep Learners Change Their Mind: Learning Dynamics for Active Learningcited by 6
  11. 2021The Power of Proxy Data and Proxy Networks for Hyper-Parameter Optimization in Medical Image Segmentationcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 13; 2024: 3; 2025: 2; 2026: 2.

21 papers.

Sort:

  1. 2023AbdomenAtlas-8K: Annotating 8,000 CT Volumes for Multi-Organ Segmentation in Three Weekscited by 52
  2. 2023A comprehensive survey on deep active learning in medical image analysiscited by 32
  3. 2022Making Your First Choice: To Address Cold Start Problem in Vision Active Learningcited by 25
  4. 2023Active learning for medical image segmentation with stochastic batchescited by 20
  5. 2023Brainomaly: Unsupervised Neurologic Disease Detection Utilizing Unannotated T1-weighted Brain MR Imagescited by 9
  6. 2023COLosSAL: A Benchmark for Cold-start Active Learning for 3D Medical Image Segmentationcited by 9
  7. 2023BaSAL: Size-Balanced Warm Start Active Learning for LiDAR Semantic Segmentationcited by 8
  8. 2023You Never Get a Second Chance To Make a Good First Impression: Seeding Active Learning for 3D Semantic Segmentationcited by 6
  9. 2023Data efficient deep learning for medical image analysis: A surveycited by 6
  10. 2024A Review of Bayesian Uncertainty Quantification in Deep Probabilistic Image Segmentationcited by 6
  11. 2023OpenAL: An Efficient Deep Active Learning Framework for Open-Set Pathology Image Classificationcited by 5
  12. 2023Hybrid Representation-Enhanced Sampling for Bayesian Active Learning in Musculoskeletal Segmentation of Lower Extremitiescited by 4
9 more
  1. 2023Correlation-aware active learning for surgery video segmentationcited by 4
  2. 2023Test-time augmentation-based active learning and self-training for label-efficient segmentationcited by 3
  3. 2023EdgeAL: An Edge Estimation Based Active Learning Approach for OCT Segmentationcited by 2
  4. 2024ProtoAL: Interpretable Deep Active Learning with prototypes for medical imagingcited by 2
  5. 2024Modular Deep Active Learning Framework for Image Annotation: A Technical Report for the Ophthalmo-AI Projectcited by 0
  6. 2025Deep Learning Approaches for Medical Imaging Under Varying Degrees of Label Availability: A Comprehensive Surveycited by 0
  7. 2025Is Complete Labeling Necessary? Understanding Active Learning in Longitudinal Medical Imagingcited by 0
  8. 2026Active Transfer Bagging: A New Approach for Accelerated Active Learning Acquisition of Data by Combined Transfer Learning and Bagging Based Modelscited by 0
  9. 2026Dataset-Aware Cold-Start Active Learning for Annotation-Efficient 3D Medical Image Segmentationcited by 0