FocusLiteNN: High Efficiency Focus Quality Assessment for Digital Pathology

Zhongling Wang, Mahdi S. Hosseini, Adyn Miles, Konstantinos N. Plataniotis and Zhou Wang “FocusLiteNN: High Efficiency Focus Quality Assessment for Digital Pathology”, 23rd International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2020.

The PyTorch and MATLAB implementations: [GitHub].

• FocusLiteNN: High Efficiency Focus Quality Assessment for Digital Pathology
  • 1. Introduction
    • 1.1 Backgrounds
    • 1.2 Contributions
    • 1.3 Results
    • 1.4 Citation
  • 2. Dataset
    • 2.1 TCGA@Focus
    • 2.2 Focuspath Full
  • 3. Code

1. Introduction

1.1 Backgrounds

• Out-of-focus microscopy lens in digital pathology is a critical bottleneck in high-throughput Whole Slide Image scanning platforms, for which Focus Quality Assessment methods are highly desirable to help significantly accelerate the clinical workflows.
• While data-driven approaches such as Convolutional Neural Network based methods have shown great promises, they are difficult to use in practice due to their high computational complexity.

1.2 Contributions

• We propose a highly efficient CNN-based model FocusLiteNN that only has 148 paramters for Focus Quality Assessment. It maintains impressive performance and is 100x faster than ResNet50.
• We introduce a comprehensive annotated dataset TCGA@Focus, which contains 14371 pathological images with in/out focus labels.

1.3 Results

• Evaluation results on the proposed TCGA@Focus dataset
  
• Our proposed FocusLiteNN (1-kernel) model is highly efficient.
  

1.4 Citation

Please cite our paper if you find our model or the TCGA@Focus dataset useful.

1
2
3
4
5
6
7

@InProceedings{wang2020focuslitenn,
    title={FocusLiteNN: High Efficiency Focus Quality Assessment for Digital Pathology},
    author={Wang, Zhongling and Hosseini, Mahdi and Miles, Adyn and Plataniotis, Konstantinos and Wang, Zhou},
    booktitle={Medical Image Computing and Computer Assisted Intervention -- MICCAI 2020},
    year={2020},
    publisher="Springer International Publishing"
}

2. Dataset

2.1 TCGA@Focus

• Download: The dataset is available on Zenodo under a Creative Commons Attribution license: .
• Content: Contains 14371 pathological images with in/out focus labels.
• Testing: This is the testing dataset proposed and used in the paper. The specific testing images (14371 images) can be found in TCGA@Focus.txt

2.2 Focuspath Full

• Download: The dataset is available on Zenodo under a Creative Commons Attribution license:
• Content:Contains 8640 pathological images of different blur levels.
• Training: This is the training dataset used in the paper. The specific training images (5200 images) in one of the ten folds can be found in FocusPath_full_split1.txt

3. Code

The PyTorch and MATLAB implementations: [GitHub]. For more details about the code such as how to train or test the model, please refer to the GitHub page.