| | --- |
| | license: mit |
| | tags: |
| | - medical |
| | - radiology |
| | - image-classification |
| | - pytorch |
| | - radimagenet |
| | - feature-extraction |
| | library_name: pytorch |
| | --- |
| | |
| | # RadImageNet Pre-trained Models |
| |
|
| | This repository contains pre-trained models from RadImageNet, a large-scale radiologic image dataset designed to facilitate transfer learning for medical imaging applications. |
| |
|
| | ## Model Description |
| |
|
| | RadImageNet models are convolutional neural networks pre-trained on a diverse collection of radiologic images spanning multiple modalities and anatomical regions. These models serve as powerful feature extractors for downstream medical imaging tasks. |
| |
|
| | ### Available Models |
| |
|
| | - **ResNet50.pt**: ResNet-50 architecture pre-trained on RadImageNet |
| | - **DenseNet121.pt**: DenseNet-121 architecture pre-trained on RadImageNet |
| | - **InceptionV3.pt**: Inception-V3 architecture pre-trained on RadImageNet |
| |
|
| | ## Usage |
| |
|
| | ```python |
| | import torch |
| | from huggingface_hub import hf_hub_download |
| | |
| | # Download and load a model |
| | model_path = hf_hub_download(repo_id="Lab-Rasool/RadImageNet", filename="ResNet50.pt") |
| | model = torch.load(model_path, map_location="cuda" if torch.cuda.is_available() else "cpu") |
| | model.eval() |
| | |
| | # Use for inference |
| | # ... your inference code here ... |
| | ``` |
| |
|
| | ## Preprocessing |
| |
|
| | Images should be preprocessed using standard ImageNet normalization: |
| |
|
| | ```python |
| | from torchvision import transforms |
| | |
| | preprocess = transforms.Compose([ |
| | transforms.Resize(256), |
| | transforms.CenterCrop(224), |
| | transforms.ToTensor(), |
| | transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]) |
| | ]) |
| | ``` |
| |
|
| | ## Citation |
| |
|
| | If you use these models in your research, please cite the RadImageNet paper: |
| |
|
| | ```bibtex |
| | @article{mei2022radimagenet, |
| | title={RadImageNet: An Open Radiologic Deep Learning Research Dataset for Effective Transfer Learning}, |
| | author={Mei, Xueyan and Liu, Zelong and Robson, Philip M and Marinelli, Brett and Huang, Mingqian and Doshi, Amish and Jacobi, Adam and Cao, Chendi and Link, Katherine E and Yang, Thomas and others}, |
| | journal={Radiology: Artificial Intelligence}, |
| | volume={4}, |
| | number={5}, |
| | pages={e210315}, |
| | year={2022}, |
| | publisher={Radiological Society of North America} |
| | } |
| | ``` |
| |
|
| | ## License |
| |
|
| | MIT License |
| |
|
| | Copyright (c) 2021 BMEII-AI |
| |
|
| | Permission is hereby granted, free of charge, to any person obtaining a copy |
| | of this software and associated documentation files (the "Software"), to deal |
| | in the Software without restriction, including without limitation the rights |
| | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell |
| | copies of the Software, and to permit persons to whom the Software is |
| | furnished to do so, subject to the following conditions: |
| |
|
| | The above copyright notice and this permission notice shall be included in all |
| | copies or substantial portions of the Software. |
| |
|
| | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR |
| | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, |
| | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE |
| | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER |
| | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, |
| | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE |
| | SOFTWARE. |
| |
|
| | ## Additional Information |
| |
|
| | - **Original Repository**: [BMEII-AI/RadImageNet](https://github.com/BMEII-AI/RadImageNet) |
| | - **Paper**: [RadImageNet: An Open Radiologic Deep Learning Research Dataset](https://pubs.rsna.org/doi/10.1148/ryai.210315) |
| | - **Dataset**: The RadImageNet dataset contains 1.35 million annotated radiologic images |
| |
|
