Multi-domain improves out-of-distribution and data-limited scenarios for medical image analysis
Not on the map. Why?
Abstract (arXiv)
Current machine learning methods for medical image analysis primarily focus on developing models tailored for their specific tasks, utilizing data within their target domain. These specialized models tend to be data-hungry and often exhibit limitations in generalizing to out-of-distribution samples. In this work, we show that employing models that incorporate multiple domains instead of specialized ones significantly alleviates the limitations observed in specialized models. We refer to this approach as multi-domain model and compare its performance to that of specialized models. For this, we introduce the incorporation of diverse medical image domains, including different imaging modalities like X-ray, MRI, CT, and ultrasound images, as well as various viewpoints such as axial, coronal, and sagittal views. Our findings underscore the superior generalization capabilities of multi-domain models, particularly in scenarios characterized by limited data availability and out-of-distribution, frequently encountered in healthcare applications. The integration of diverse data allows multi-domain models to utilize information across domains, enhancing the overall outcomes substantially. To illustrate, for organ recognition, multi-domain model can enhance accuracy by up to 8% compared to conventional specialized models.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 51 of its 76 references that are in the map. First posted on arXiv 10 Oct 2023.
Builds on
51 of 76 references.
References · 51
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
- 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
- 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
- 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
- 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,326
- 2020Training data-efficient image transformers & distillation through attentioncited by 4,586
- 2019BioBERT: a pre-trained biomedical language representation model for biomedical text miningcited by 1,947
- 2021TransUNet: Transformers Make Strong Encoders for Medical Image Segmentationcited by 1,558
- 2019CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparisoncited by 1,423
- 2021Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentationcited by 1,222
- 2021EfficientNetV2: Smaller Models and Faster Trainingcited by 1,135
38 more
- 2020Designing Network Design Spacescited by 960
- 2021UNETR: Transformers for 3D Medical Image Segmentationcited by 959
- 2019Big Transfer (BiT): General Visual Representation Learningcited by 769
- 2021MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classificationcited by 690
- 2020Contrastive Learning of Medical Visual Representations from Paired Images and Textcited by 641
- 2021Transformer in Transformercited by 633
- 2017Learning multiple visual domains with residual adapterscited by 557
- 2022BioGPT: Generative Pre-trained Transformer for Biomedical Text Generation and Miningcited by 538
- 2023Towards Expert-Level Medical Question Answering with Large Language Modelscited by 384
- 2020Med-BERT: pre-trained contextualized embeddings on large-scale structured electronic health records for disease predictioncited by 373
- 2021Domain Adaptation for Medical Image Analysis: A Surveycited by 352
- 2021Big Self-Supervised Models Advance Medical Image Classificationcited by 320
- 2021Medical Transformer: Gated Axial-Attention for Medical Image Segmentationcited by 320
- 2019Med3D: Transfer Learning for 3D Medical Image Analysiscited by 288
- 2019BEHRT: Transformer for Electronic Health Recordscited by 263
- 2021TransFuse: Fusing Transformers and CNNs for Medical Image Segmentationcited by 258
- 2021CoTr: Efficiently Bridging CNN and Transformer for 3D Medical Image Segmentationcited by 227
- 2023BiomedGPT: A Generalist Vision-Language Foundation Model for Diverse Biomedical Taskscited by 201
- 2021Self-supervised Pretraining of Visual Features in the Wildcited by 198
- 2018Efficient parametrization of multi-domain deep neural networkscited by 168
- 2019Models Genesis: Generic Autodidactic Models for 3D Medical Image Analysiscited by 159
- 2019A deep learning system for differential diagnosis of skin diseasescited by 118
- 2023ChatCAD: Interactive Computer-Aided Diagnosis on Medical Image using Large Language Modelscited by 116
- 2017Incremental Learning Through Deep Adaptationcited by 108
- 2021Generalized Multimodal ELBOcited by 87
- 2022BioBART: Pretraining and Evaluation of A Biomedical Generative Language Modelcited by 83
- 2020Comparing to Learn: Surpassing ImageNet Pretraining on Radiographs By Comparing Image Representationscited by 69
- 2020MoCo-CXR: MoCo Pretraining Improves Representation and Transferability of Chest X-ray Modelscited by 61
- 2023Large AI Models in Health Informatics: Applications, Challenges, and the Futurecited by 61
- 2023LVM-Med: Learning Large-Scale Self-Supervised Vision Models for Medical Imaging via Second-order Graph Matchingcited by 60
- 2023Generalist Vision Foundation Models for Medical Imaging: A Case Study of Segment Anything Model on Zero-Shot Medical Segmentationcited by 51
- 2017Universal representations:The missing link between faces, text, planktons, and cat breedscited by 45
- 2019Probing Biomedical Embeddings from Language Modelscited by 39
- 2020Towards Cross-modality Medical Image Segmentation with Online Mutual Knowledge Distillationcited by 38
- 2021Improving Biomedical Pretrained Language Models with Knowledgecited by 36
- 2023ChatCAD+: Towards a Universal and Reliable Interactive CAD using LLMscited by 35
- 2023RAMM: Retrieval-augmented Biomedical Visual Question Answering with Multi-modal Pre-trainingcited by 31
- 2020Real-World Multi-Domain Data Applications for Generalizations to Clinical Settingscited by 5
50 most cited of 51 All 51 (JSON)
Built on by
Citing papers per year: 2024: 1.
1 paper.