MedicAI

Medic-AI is a Keras based library designed for medical image analysis using machine learning techniques. Its core strengths include:

• Backend Agnostic: Compatible with tensorflow, torch, and jax backends.
• User-Friendly API: High-level interface for transformations and model creation.
• Scalable Execution: Supports training and inference on single/multi-GPU and TPU-VM setups.
• Essential Components: Includes standard metrics and losses, such as Dice.
• Optimized 3D Inference: Offers an efficient sliding-window method and callback for volumetric data

📋 Table of Contents

1. Installation
2. Features
3. Guides
4. Documentation
5. Acknowledgements
6. Citation

🛠 Installation

PyPI version:

!pip install medicai

Installing from source GitHub:

!pip install git+https://github.com/innat/medic-ai.git

📊 Features

Available Models : The following table lists the currently supported models along with their supported input modalities, primary tasks, and underlying architecture type. The model inputs can be either 3D (depth × height × width × channel) or 2D (height × width × channel).

Model	Supported Modalities	Primary Task	Architecture Type
DenseNet121	2D, 3D	Classification	CNN
DenseNet169	2D, 3D	Classification	CNN
DenseNet201	2D, 3D	Classification	CNN
ViT	2D, 3D	Classification	Transformer
Swin Transformer	2D, 3D	Classification	Transformer
DenseUNet121	2D, 3D	Segmentation	CNN
DenseUNet169	2D, 3D	Segmentation	CNN
DenseUNet201	2D, 3D	Segmentation	CNN
UNETR	2D, 3D	Segmentation	Transformer
SwinUNETR	2D, 3D	Segmentation	Transformer
TransUNet	2D, 3D	Segmentation	Transformer
SegFormer	2D, 3D	Segmentation	Transformer

Available Transformation: The following preprocessing and transformation methods are supported for volumetric data. The following layers are implemented with TensorFlow operations. It can be used in the tf.data API or a Python data generator and is fully compatible with multiple backends, tf, torch, jax in training and inference, supporting both GPUs and TPUs.

CropForeground
NormalizeIntensity
Orientation
RandCropByPosNegLabel
RandFlip
RandRotate90
RandShiftIntensity
RandSpatialCrop
Resize
ScaleIntensityRange
Spacing

💡 Guides

Segmentation: Available guides for 3D segmentation task.

Task	GitHub	Kaggle	View
Covid-19
BTCV			n/a
BraTS			n/a
Spleen

Classification: Available guides for 3D classification task.

Task (Classification)	GitHub	Kaggle
Covid-19

📚 Documentation

To learn more about model, transformation, and training, please visit official documentation: medicai/docs

🤝 Contributing

Please refer to the current roadmap for an overview of the project. Feel free to explore anything that interests you. If you have suggestions or ideas, I’d appreciate it if you could open a GitHub issue so we can discuss them further.

1. Install medicai from soruce:

!git clone https://github.com/innat/medic-ai
%cd medic-ai
!pip install keras -qU
!pip install -e .
%cd ..

Add your contribution and implement relevant test code.

1. Run test code as:

python -m pytest test/

# or, only one your new_method
python -m pytest -k new_method

🙏 Acknowledgements

This project is greatly inspired by MONAI.

📝 Citation

If you use medicai in your research or educational purposes, please cite it using the metadata from our CITATION.cff file.