mlsanityData to 7 Oct 2026

Label-Assemble: Leveraging Multiple Datasets with Partial Labels

arXiv 2109.12265 ↗First posted 25 Sep 2021cs.CV cs.AIcited by 20

Not on the map. Why?

Abstract (arXiv)

The success of deep learning relies heavily on large labeled datasets, but we often only have access to several small datasets associated with partial labels. To address this problem, we propose a new initiative, "Label-Assemble", that aims to unleash the full potential of partial labels from an assembly of public datasets. We discovered that learning from negative examples facilitates both computer-aided disease diagnosis and detection. This discovery will be particularly crucial in novel disease diagnosis, where positive examples are hard to collect, yet negative examples are relatively easier to assemble. For example, assembling existing labels from NIH ChestX-ray14 (available since 2017) significantly improves the accuracy of COVID-19 diagnosis from 96.3% to 99.3%. In addition to diagnosis, assembling labels can also improve disease detection, e.g., the detection of pancreatic ductal adenocarcinoma (PDAC) can greatly benefit from leveraging the labels of Cysts and PanNets (two other types of pancreatic abnormalities), increasing sensitivity from 52.1% to 84.0% while maintaining a high specificity of 98.0%.

Key facts. Cited by 20 papers in the map (data through 7 Oct 2026). Builds on 48 of its 78 references that are in the map. First posted on arXiv 25 Sep 2021.

Builds on

48 of 78 references.

References · 48

  1. 2020End-to-End Object Detection with Transformerscited by 7,384
  2. 2016Densely Connected Convolutional Networkscited by 6,691
  3. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  4. 2020FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidencecited by 2,080
  5. 2019CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparisoncited by 1,410
  6. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,304
  7. 2021Per-Pixel Classification is Not All You Need for Semantic Segmentationcited by 1,148
  8. 2017ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseasescited by 969
  9. 2017Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learningcited by 918
  10. 2021The Medical Segmentation Decathloncited by 620
  11. 2021The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classificationcited by 507
  12. 2019The Liver Tumor Segmentation Benchmark (LiTS)cited by 504
36 more
  1. 2016Regularization With Stochastic Transformations and Perturbations for Deep Semi-Supervised Learningcited by 441
  2. 2020Propagate Yourself: Exploring Pixel-Level Consistency for Unsupervised Visual Representation Learningcited by 297
  3. 2019Med3D: Transfer Learning for 3D Medical Image Analysiscited by 282
  4. 2020COVID-Net: A Tailored Deep Convolutional Neural Network Design for Detection of COVID-19 Cases from Chest X-Ray Imagescited by 264
  5. 2014Learning with Pseudo-Ensemblescited by 241
  6. 2020Being Bayesian, Even Just a Bit, Fixes Overconfidence in ReLU Networkscited by 227
  7. 2020MedMNIST Classification Decathlon: A Lightweight AutoML Benchmark for Medical Image Analysiscited by 206
  8. 2019The state of the art in kidney and kidney tumor segmentation in contrast-enhanced CT imaging: Results of the KiTS19 Challengecited by 188
  9. 2020Hierarchical Multi-Scale Attention for Semantic Segmentationcited by 183
  10. 2019The KiTS19 Challenge Data: 300 Kidney Tumor Cases with Clinical Context, CT Semantic Segmentations, and Surgical Outcomescited by 168
  11. 2020CheXclusion: Fairness gaps in deep chest X-ray classifierscited by 153
  12. 2020Unsupervised Learning of Dense Visual Representationscited by 147
  13. 2020DoDNet: Learning to segment multi-organ and tumors from multiple partially labeled datasetscited by 94
  14. 2019Prior-aware Neural Network for Partially-Supervised Multi-Organ Segmentationcited by 88
  15. 2021XProtoNet: Diagnosis in Chest Radiography with Global and Local Explanationscited by 72
  16. 2020Multi-organ Segmentation over Partially Labeled Datasets with Multi-scale Feature Abstractioncited by 70
  17. 2018Comparison of Deep Learning Approaches for Multi-Label Chest X-Ray Classificationcited by 57
  18. 2020Interpreting Chest X-rays via CNNs that Exploit Hierarchical Disease Dependencies and Uncertainty Labelscited by 56
  19. 2017Thoracic Disease Identification and Localization with Limited Supervisioncited by 55
  20. 2020Learning from Multiple Datasets with Heterogeneous and Partial Labels for Universal Lesion Detection in CTcited by 42
  21. 2017Fast Image Processing with Fully-Convolutional Networkscited by 41
  22. 2017Learning to diagnose from scratch by exploiting dependencies among labelscited by 33
  23. 2018Learning to recognize Abnormalities in Chest X-Rays with Location-Aware Dense Networkscited by 33
  24. 2018Attention-Guided Curriculum Learning for Weakly Supervised Classification and Localization of Thoracic Diseases on Chest Radiographscited by 32
  25. 2021Cross-Dataset Collaborative Learning for Semantic Segmentation in Autonomous Drivingcited by 23
  26. 2018Weakly Supervised Medical Diagnosis and Localization from Multiple Resolutionscited by 22
  27. 2020Multi-label Thoracic Disease Image Classification with Cross-Attention Networkscited by 22
  28. 2020Taskology: Utilizing Task Relations at Scalecited by 19
  29. 2017Boosted Cascaded Convnets for Multilabel Classification of Thoracic Diseases in Chest Radiographscited by 18
  30. 2020Region Proposals for Saliency Map Refinement for Weakly-supervised Disease Localisation and Classificationcited by 15
  31. 2019GraphXNET- Chest X-Ray Classification Under Extreme Minimal Supervisioncited by 14
  32. 2020Improving Interpretability in Medical Imaging Diagnosis using Adversarial Trainingcited by 9
  33. 2021Incremental Learning for Multi-organ Segmentation with Partially Labeled Datasetscited by 9
  34. 2018Dynamic Routing on Deep Neural Network for Thoracic Disease Classification and Sensitive Area Localizationcited by 7
  35. 2020Universal Lesion Detection by Learning from Multiple Heterogeneously Labeled Datasetscited by 7
  36. 20193D Kidneys and Kidney Tumor Semantic Segmentation using Boundary-Aware Networkscited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 3; 2024: 10; 2025: 3; 2026: 3.

20 papers.

Sort:

By area:

  1. 2023CLIP-Driven Universal Model for Organ Segmentation and Tumor Detectioncited by 224
  2. 2024MAISI: Medical AI for Synthetic Imagingcited by 86
  3. 2022Delving into Masked Autoencoders for Multi-Label Thorax Disease Classificationcited by 65
  4. 2024Towards Generalizable Tumor Synthesiscited by 65
  5. 2023AbdomenAtlas-8K: Annotating 8,000 CT Volumes for Multi-Organ Segmentation in Three Weekscited by 52
  6. 2024SELF-GUIDE: Better Task-Specific Instruction Following via Self-Synthetic Finetuningcited by 32
  7. 2025PanTS: The Pancreatic Tumor Segmentation Datasetcited by 15
  8. 2025MedSegFactory: Text-Guided Generation of Medical Image-Mask Pairscited by 13
  9. 2025Scaling Tumor Segmentation: Best Lessons from Real and Synthetic Datacited by 13
  10. 2024Analyzing Tumors by Synthesiscited by 11
  11. 2026Efficient Chest X-ray Representation Learning via Semantic-Partitioned Contrastive Learningcited by 10
  12. 2023Synthetic Data as Validationcited by 9
8 more
  1. 2024COLEP: Certifiably Robust Learning-Reasoning Conformal Prediction via Probabilistic Circuitscited by 9
  2. 2024Embracing Massive Medical Datacited by 9
  3. 2024Investigating the Quality of DermaMNIST and Fitzpatrick17k Dermatological Image Datasetscited by 7
  4. 2024Learning Low-Rank Feature for Thorax Disease Classificationcited by 4
  5. 2026RadThinking: A Dataset for Longitudinal Clinical Reasoning in Radiologycited by 4
  6. 2024Deep Mutual Learning among Partially Labeled Datasets for Multi-Organ Segmentationcited by 2
  7. 20263D-LLDM: Label-Guided 3D Latent Diffusion Model for Improving High-Resolution Synthetic MR Imaging in Hepatic Structure Segmentationcited by 2
  8. 2024Labeled-to-Unlabeled Distribution Alignment for Partially-Supervised Multi-Organ Medical Image Segmentationcited by 1