Multi-scale Regional Attention Deeplab3+: Multiple Myeloma Plasma Cells Segmentation in Microscopic Images

The two-stage deep model was exploited to overcome the limitation of existing methods for multiple myeloma instance segmentation...
If this code helps with your research please consider citing the following papers:

@article{bozorgpour2021multi,
  title={Multi-scale Regional Attention Deeplab3+: Multiple Myeloma Plasma Cells Segmentation in Microscopic Images},
  author={Bozorgpour, Afshin and Azad, Reza and Showkatian, Eman and Sulaiman, Alaa},
  journal={arXiv preprint arXiv:2105.06238},
  year={2021}
}

Please consider starring us, if you found it useful. Thanks

Updates

• May 25, 2021: First version released. All trained weights are available now.

Prerequisites and Run

This code has been implemented in Python language using Keras library with TensorFlow backend and tested in Ubuntu OS, though should be compatible with related environments. following environment and Library needed to run the code:

• Python 3
• Please install the requirements (pip install -r requirements.txt)

Run Demo

Please follow the below steps:
1- Put your test set inside the ./dataset/
2- Add the above path to the ./configs/config.json
3- Download the weights and put in weights folder
4- Run main.ipynb cell by cell.

Quick Overview

Results

For evaluating the performance of the proposed method, Two challenging tasks in medical image segmentation have been considered. In bellow, the results of the proposed approach are illustrated.

Performance Comparision

In order to compare the proposed method with state-of-the-art approaches on ...

Segmentation results

Model weights

You can download the learned weights from the following table.

Model	Learned weights
ADeepLab_X	Download

Query

All implementation is done by bmdeep.com. For any query please contact us for more information.

[email protected]