mri-deep-learning-tools
Open-source libraries for MRI images processing and deep learning.
Updates in 2020: monai,torchio, medicalzooputorch, transunet virused.
Updates in 2021: torchio wrote an adds-in for monai. There is super-usefull tutorails from MONAI and NVIDIA for almost all task, that you need in MRI imaging in 2D and 3D. for
pytorch and tensorflow. Also check catalog for conteinerized solutions.
DALI augmentations just saving lots of time!
Pytorch
| Project Name | # stars | Description | Scenario |
|---|---|---|---|
| MONAI | 2200 | Medical Open Network for AI MONAI is a PyTorch-based, open-source framework for deep learning in healthcare imaging, part of PyTorch Ecosystem. |
preprocessing, classification, segmentation |
| SegNet | 1100 | SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation Vijay Badrinarayanan, Alex Kendall and Roberto Cipolla, PAMI 2017 [http://arxiv.org/abs/1511.00561] | segmentation |
| Medical Detection Toolkit | 1k | The Medical Detection Toolkit contains 2D + 3D implementations of prevalent object detectors such as Mask R-CNN, Retina Net, Retina U-Net, as well as a training and inference framework focused on dealing with medical images. | detection, segmentation |
| TorchIO | 986 | is a Python package containing a set of tools to efficiently read, preprocess, sample, augment, and write 3D medical images in deep learning applications written in PyTorch Pérez-GarcÃa et al., 2020, TorchIO: a Python library for efficient loading, preprocessing, augmentation and patch-based sampling of medical images in deep learning. |
training, preprocessing |
| DeepMedic | 859 | Efficient Multi-Scale 3D Convolutional Neural Network for Segmentation of 3D Medical Scans Project aims to offer easy access to Deep Learning for segmentation of structures of interest in biomedical 3D scans. It is a system that allows the easy creation of a 3D Convolutional Neural Network, which can be trained to detect and segment structures if corresponding ground truth labels are provided for training. The system processes NIFTI images, making its use straightforward for many biomedical tasks. | segmentation |
| MedicalTorch | 739 | is an open-source framework for PyTorch, implementing an extensive set of loaders, pre-processors and datasets for medical imaging. | preprocessing |
| MEDICAL ZOO | 724 | a 3D multi-modal medical image segmentation library in PyTorch | segmentation |
| TransUnet | 672 | Transformers (ViT) for medical image segmentation (abdominar) | segmentation |
| nipy Computational Anatomy: - dipy - mindboggle File I/O and Data Management: - nibabel Functional MRI: - Nipy - Nitime - popeye Machine Learning: - Nilearn - PyMVPA Human Electrophysiology: - MNE Data Visualisation: - niwidgets |
586 | -a community of practice devoted to the use of the Python programming language in the analysis of neuroimaging data Computational Anatomy: - Focuses on diffusion magnetic resonance imaging (dMRI) analysis. - Improves the accuracy, precision, and consistency of labeling & morphometry of brain imaging data. File I/O and Data Management: - Read / write common neuroimaging file formats. Functional MRI: - Analysis of structural and functional neuroimaging data. - Time-series analysis of neuroscience data. - Population receptive field estimation Machine Learning: - Fast and easy statistical learning on neuroimaging data. - Eases statistical learning analyses of large neuroimaging datasets. Human Electrophysiology: - Processes magnetoencephalography (MEG) and electroencephalography (EEG) data - Provides interactive plots for volumetric images. Data Visualisation: - Provides a uniform interface to existing neuroimaging software. |
- denoising, registration, reconstruction, tracking, clustering, visualization, and statistical analysis |
| Deep Cascade of Convolutional Neural Networks and Convolutioanl Recurrent Nerual Networks for MR Image Reconstruction | 227 | Reconstruct MR images from its undersampled measurements using Deep Cascade of Convolutional Neural Networks (DC-CNN) and Convolutional Recurrent Neural Networks (CRNN-MRI). This repository contains the implementation of DC-CNN using Theano and Lasagne, and CRNN-MRI using PyTorch, along with simple demos. Note that the library requires the dev version of Lasagne and Theano, as well as pygpu backend for using CUFFT Library | reconstruction |
| SCT | 122 | Spinal Cord Toolbox (SCT) is a comprehensive, free and open-source software dedicated to the processing and analysis of spinal cord MRI data. | preprocessing, segmentation |
| TractSeg | 117 | -Tool for fast and accurate white matter bundle segmentation from Diffusion MRI. It can create bundle segmentations, segmentations of the endregions of bundles and Tract Orientation Maps (TOMs). Moreover, it can do tracking on the TOMs creating bundle-specific tractogram and do Tractometry analysis on those. | segmentation |
| Clinica Deep Learning (clinicadl) | 87 | This repository hosts the code source for reproducible experiments on automatic classification of Alzheimer's disease (AD) using anatomical MRI data. It allows to train convolutional neural networks (CNN) models. The journal version of the paper describing this work is available here. | classification |
| BrainPrep | 84 | Nicely written code for brain MRI preprocessing whole pipeline | preprocessing |
| GANCS | 84 | Compressed Sensing MRI based on Generative Adversarial Network | reconstruction |
| Layer-wise relevance propagation for explaining deep neural network decisions in MRI-based Alzheimer’s disease classification | 84 | Moritz Böhle, Fabian Eitel, Martin Weygandt, and Kerstin Ritter Preprint: https://arxiv.org/abs/1903.07317 |
classification |
| 3D-UNet-PyTorch-Implementation | 57 | The implementation of 3D UNet Proposed by Özgün Çiçek et al., Preprint:3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation. |
segmentation |
| NiftyTorch | 29 | A Python API for deploying deep neural networks for Neuroimaging research | classification, segmentation |
| DeepPipe | 29 | Manipulation with medical images - parallel model training, model optimisation and utils | model training |
| BraTS-Toolkit | 17 | Docker images for best performing BRATS tumor segmentation solutions | segmentation |
Tensorflow
| Project Name | # stars | Description | Scenario |
|---|---|---|---|
| NiftyNet | 1.2k | NiftyNet is a TensorFlow-based open-source convolutional neural networks (CNN) platform for research in medical image analysis and image-guided therapy. NiftyNet's modular structure is designed for sharing networks and pre-trained models. | classification, segmentation |
| DLTK | 1.1k | Deep Learning Toolkit (DLTK) for Medical Imaging DLTK is a neural networks toolkit written in python, on top of TensorFlow. It is developed to enable fast prototyping with a low entry threshold and ensure reproducibility in image analysis applications, with a particular focus on medical imaging. Its goal is to provide the community with state of the art methods and models and to accelerate research in this exciting field. |
classification, segmentation, super-resolution, regression |
| CNN 3D Images using Tensorflow | 184 | MRI classification task using CNN (Convolutional Neural Network) | classification |
| Graph CNNs for population graphs: classification of the ABIDE dataset | 64 | code provides a python - Tensorflow implementation of graph convolutional networks (GCNs) for semi-supervised disease prediction using population graphs | classification |
| Nobrainer | 44 | -is a deep learning framework for 3D image processing. It implements several 3D convolutional models from recent literature, methods for loading and augmenting volumetric data that can be used with any TensorFlow or Keras model, losses and metrics for 3D data, and simple utilities for model training, evaluation, prediction, and transfer learning. | preprocessing, segmentation |
| 3D-Convolutional-Network-for-Alzheimer's-Detection | 15 | This repository consists of an attempt to detect and diagnose Alzheimer's using 3D MRI T1 weighted scans from the ADNI database.It contains a data preprocessing pipeline to make the data suitable for feeding to a 3D Convnet or Voxnet followed by a Deep Neural Network definition and an exploration into all the utilities that could be required for such a task. | detection, preprocessing |