Related Experiment Video
Updated: Oct 21, 2025
11:00
Visualization, Quantification, and Mapping of Immune Cell Populations in the Tumor Microenvironment
Published on: March 25, 2020
17.4K
Improving Algorithm for the Alignment of Consecutive, Whole-Slide, Immunohistochemical Section Images.
Cher-Wei Liang1,2,3, Ruey-Feng Chang4,5,6, Pei-Wei Fang1
1Department of Pathology, Fu Jen Catholic University Hospital, Fu Jen Catholic University, New Taipei City, Taiwan.
Journal of Pathology Informatics
|September 3, 2021
Summary
This study presents an automated graph-based method for aligning whole-slide immunohistochemical (IHC) tissue sections, improving 3D reconstruction accuracy. The new approach significantly reduces alignment errors compared to previous methods.
Area of Science:
- Histopathology
- Computational Biology
- Medical Imaging
Background:
- Accurate alignment of histopathology tissue sections is crucial for proteome topology interpretation and 3D tissue reconstruction.
- Automated and robust methods for aligning non-globally stained immunohistochemical (IHC) sections remain a challenge.
Purpose of the Study:
- To assess the feasibility of multidimensional graph-based image registration for aligning serial-section and whole-slide IHC images.
- To develop an automated and robust method for precise histopathology slide alignment.
Main Methods:
- An automated, patch graph-based registration method was developed for whole-slide IHC sections (×10 magnification).
- The method involved initial rigid registration, followed by nonlinear registration, multidimensional graph-based registration of segmented patches, and patch fusion.
- Performance was evaluated using the Hausdorff distance between continuous image slices.
Main Results:
- The automated method successfully aligned images from five tissue types using 21 different IHC antibodies.
- The average Hausdorff distance achieved was 48.93 μm (SD 14.94 μm).
- This represents a significant improvement over previous methods (average Hausdorff distance of 93.89 μm).
Conclusions:
- The developed method is effective for cell-level resolution alignment of whole-slide tissue sections.
- This advancement is expected to facilitate further progress in proteome topology screening and 3D tissue reconstruction.
