Color standardization method and system for whole slide imaging based on spectral sensing

Area of Science:

• Digital Pathology
• Computational Imaging
• Biomedical Engineering

Background:

• Whole slide imaging (WSI) is crucial for digital pathology.
• Color variations from imaging devices and staining processes introduce significant variability in WSI analysis.
• This variability impacts the accuracy and reproducibility of image analysis performed by pathologists.

Purpose of the Study:

• To develop and validate a novel color standardization method and system for whole slide imaging.
• To address the challenge of color variations in WSI to stabilize and improve image analysis.
• To enhance the performance of automated image analysis tasks in digital pathology.

Main Methods:

• Developed a color standardization method utilizing RGB imaging and multispectral sensing with fewer bands (16 bands) compared to conventional methods (60 bands).
• Integrated a high-speed spectral sensing module into the system.
• Evaluated the method's performance on nucleus detection and H&E component analysis.

Main Results:

• Demonstrated improved performance in nucleus detection following color standardization, eliminating the need for training datasets required by conventional methods.
• Achieved H&E component detection performance equivalent to conventional 60-band methods.
• The measurement process was accelerated by over 255 times compared to conventional approaches.

Conclusions:

• The developed color standardization method effectively stabilizes whole slide imaging analysis.
• This approach significantly enhances nucleus detection accuracy and efficiency.
• The system offers a faster and more accessible solution for digital pathology image analysis, comparable to traditional methods.