Blind Image Quality Assessment for Pathological Microscopic Image Under Screen and Immersion Scenarios
Abstract:
The high-quality pathological microscopic images are essential for physicians or pathologists to make a correct diagnosis. Image quality assessment (IQA) can quantify the visual distortion degree of images and guide the imaging system to improve image quality, thus raising the quality of pathological microscopic images. Current IQA methods are not ideal for pathological microscopy images due to their specificity. In this paper, we present deep learning-based blind image quality assessment model with saliency block and patch block for pathological microscopic images. The saliency block and patch block can handle the local and global distortions, respectively. To better capture the area of interest of pathologists when viewing pathological images, the saliency block is fine-tuned by eye movement data of pathologists. The patch block can capture lots of global information strongly related to image quality via the interaction between different image patches from different positions. The performance of the developed model is validated by the home-made Pathological Microscopic Image Quality Database under Screen and Immersion Scenarios (PMIQD-SIS) and cross-validated by the five public datasets. The results of ablation experiments demonstrate the contribution of the added blocks. The dataset and the corresponding code are publicly available at: https://github.com/mikugyf/PMIQD-SIS.
Insights
This study introduces a novel deep learning model for assessing pathological microscopic image quality. The model, incorporating saliency and patch blocks, improves diagnostic accuracy by quantifying image distortions.
Area of Science:
- Medical Imaging
- Computer Vision
- Pathology
Background:
- High-quality pathological microscopic images are crucial for accurate medical diagnosis.
- Current image quality assessment (IQA) methods lack specificity for pathological images.
- Developing specialized IQA models can enhance diagnostic capabilities.
Purpose of the Study:
- To develop a deep learning-based blind image quality assessment (IQA) model for pathological microscopic images.
- To address the limitations of existing IQA methods in the specific domain of pathology.
- To improve the overall quality of pathological images for better diagnostic outcomes.
Main Methods:
- A novel deep learning model incorporating a saliency block and a patch block was developed.
- The saliency block was fine-tuned using pathologist eye-tracking data to focus on areas of interest.
- The patch block was designed to capture global image quality information through patch interactions.
Main Results:
- The model demonstrated strong performance in assessing pathological image quality, validated on a custom database (PMIQD-SIS) and public datasets.
- Ablation experiments confirmed the significant contribution of the saliency and patch blocks.
- The developed model effectively quantifies visual distortions in pathological images.
Conclusions:
- The proposed deep learning model offers an effective solution for blind image quality assessment in pathological microscopy.
- The integration of saliency and patch blocks enhances the model's ability to handle local and global image distortions.
- This work provides a valuable tool for improving pathological image quality and supporting accurate diagnoses.
More Related Videos
09:31High-Speed Ultraviolet Photoacoustic Microscopy for Histological Imaging with Virtual-Staining assisted by Deep Learning
Published on: April 28, 2022
14:09Quantitative Optical Microscopy: Measurement of Cellular Biophysical Features with a Standard Optical Microscope
Published on: April 7, 2014
