Related Experiment Video
Updated: Dec 28, 2025
08:40
Quantitation of Protein Expression and Co-localization Using Multiplexed Immuno-histochemical Staining and Multispectral Imaging
Published on: April 8, 2016
13.3K
Marker controlled superpixel nuclei segmentation and automatic counting on immunohistochemistry staining images
Jie Shu1,2, Jingxin Liu3, Yongmei Zhang1
1School of Information Science and Technology, North China University of Technology.
Bioinformatics (Oxford, England)
|February 20, 2020
Summary
This study introduces an automated method for segmenting and counting nuclei in histopathological images, overcoming manual counting limitations. The new approach achieves high accuracy in both segmentation and counting, aiding cancer diagnosis.
Area of Science:
- Computational pathology
- Digital image analysis
- Biomedical imaging
Background:
- Manual nuclei counting in histopathology is time-consuming and subjective.
- Accurate nuclei quantification is crucial for cancer diagnosis and research.
Purpose of the Study:
- To develop and validate an automated method for nuclei segmentation and counting in histopathological images.
- To improve the efficiency and objectivity of nuclei analysis.
Main Methods:
- A novel segmentation method using superpixel segmentation with superseeds for improved nucleus detection.
- A fusing method to reduce stain variations while preserving nucleus contour information.
- Implementation as an ImageJ plugin with freely available source code.
Main Results:
- Achieved the highest mean F1-score of 0.668 in segmentation accuracy compared to five existing methods.
- Demonstrated a high correlation (R² = 0.901, P < 0.001) between automated and manual nuclei counts on a large dataset.
- The ImageJ plugin tool achieved a Data Science Bowl score of 0.331 ± 0.006.
Conclusions:
- The proposed method offers an accurate and automated solution for nuclei segmentation and counting.
- This tool can enhance the objectivity and efficiency of histopathological image analysis for cancer diagnosis.
- The freely available ImageJ plugin facilitates wider adoption and research.
