Awesome-Medical-Vision-Language-Learning
收藏资源简介:
该合集专注于收集和索引医学视觉语言学习领域的公共数据集,涵盖医学影像模态如胸部X光、CT、超声、MRI等,并提供数据集的年份、图像数量、文本信息及相关论文链接。同时,合集还包括相关调查、教程和视觉语言预训练方法,旨在为研究人员提供全面的资源索引。
This collection focuses on collecting and indexing public datasets in the field of medical vision-language learning, covering medical imaging modalities such as chest X-rays, CT scans, ultrasound, MRI and other modalities. It provides the release year, number of images, textual information and relevant paper links for each dataset. Additionally, the collection includes relevant surveys, tutorials and vision-language pre-training methods, aiming to offer comprehensive resource indexes for researchers.
数据集详情总结
数据集列表
该页面整理了4个医学视觉语言数据集,涵盖年份、模态、图像数量和文本数量:
| 数据集 | 年份 | 模态 | 图像数量 | 文本数量 |
|---|---|---|---|---|
| MIMIC-CXR | 2019 | 胸部X光 | 377,110 | 227,827 |
| CheXpert | 2019 | 胸部X光 | 224,316 | 224,316 |
| ROCO | 2018 | CT、超声、X光、透视、PET、乳腺摄影、MRI、血管造影、PET-CT | 81,825 | 81,825 |
| MedICaT | 2020 | CT、超声、X光、透视、PET、乳腺摄影、MRI、血管造影、PET-CT | 217,060 | 217,060 |
综述论文
- 2022年《VLP: A Survey on Vision-Language Pre-training》(arXiv)
- 2022年《Vision-Language Pre-training: Basics, Recent Advances, and Future Trends》(arXiv)
- 2022年《Beyond Medical Imaging: A Review of Multimodal Deep Learning in Radiology》(techrxiv)
教程
- 2022年ACL《Vision-Language Pretraining: Current Trends and the Future》
- 2022年CVPR《Recent Advances in Vision-and-Language Pre-training》
视觉语言预训练
文本编码器
| 编码器 | 年份 | 语料库 |
|---|---|---|
| BioBERT | 2020 | PubMed |
| ClinicalBERT | 2019 | MIMIC-III |
| PubMedBERT | 2022 | PubMed |
| CXR-BERT | 2022 | PubMed+MIMIC-III/CXR |
训练方法(按年份)
2023年:
- PMC-CLIP(arXiv)
- BiomedCLIP(arXiv)
- Vision-Language Modelling for Radiological Imaging and Reports in the Low Data Regime(MIDL)
- PTUnifier(arXiv)
- MRM(ICLR)
- BioViL-T(CVPR)
- MedKLIP(arXiv)
2022年:
- MGCA(NIPS)
- MedCLIP(EMNLP)
- M3AE(MICCAI)
- Breaking with Fixed Set Pathology Recognition through Report-Guided Contrastive Training(MICCAI)
- ARL(MM)
- MedViLL(JHB1)
- REFERS(Nature Machine Intelligence)
- BioViL(ECCV)
- LoVT(ECCV)
2021年:
- Local-MI(MICCAI)
- GLoRIA(ICCV)
- Self-supervised Image-text Pre-training With Mixed Data In Chest X-rays(arXiv)
2020年:
- A Comparison of Pre-trained Vision-and-Language Models(BIBM)
- ConVIRT(MLHC)
2018年:
- Unsupervised Multimodal Representation Learning across Medical Images and Reports(NIPS workshop)
使用方法(按年份)
2023年:
- Medical Image Understanding with Pretrained Vision Language Models(ICLR)
2022年:
- Adapting Pretrained Vision-Language Foundational Models to Medical Imaging Domains(NIPS workshop)
2021年:
- PubMedCLIP(arXiv)
视觉语言任务
分割
- LViT(2022,arXiv)
生成
- RoentGen(2022,arXiv)
更多相关论文可参考:Awesome-Multimodal-Applications-In-Medical-Imaging




