Sign in Subscribe
Medical research

Whole Slide Image Registration in Pathology Informatics: A Review of Current Approaches and Challenges

Strategic Insights

— 1 min read

Researchers from the University of Warwick have published a comprehensive review of whole slide image registration in pathology informatics. The study highlights the importance of WSI registration in analyzing the tumor microenvironment (TME) in histopathology, particularly in identifying neighboring nuclei along the Z-axis for creating a 3D image or identifying subclasses of cells in the TME. The research examines current deep learning methods used for WSI registration, emphasizing their diverse methodologies and limitations.

Key Takeaways:

  • Whole slide image (WSI) registration is a crucial task in pathology informatics for analyzing the tumor microenvironment (TME) in histopathology.
  • The goal of WSI registration is to identify neighboring nuclei along the Z-axis for creating a 3D image or identifying subclasses of cells in the TME.
  • Current approaches to WSI registration are limited by factors such as the gigapixel size of images, variations in appearance between differently stained tissues, and changes in structure and morphology between non-consecutive sections.
  • Deep learning methods have been employed for WSI registration, but their diverse methodologies and limitations make it challenging to evaluate their performance.
  • The study identifies open challenges and potential future trends in this area of research, including the development of more robust and accurate registration algorithms.
  • The authors acknowledge the need for a comprehensive understanding of the available approaches and their application for various purposes.
  • Abdullah Alsalemi and his team at the Tissue Image Analytics (TIA) Centre, University of Warwick, are investigating current deep learning methods used for WSI registration.
  • The study was published in the Journal of Pathology Informatics and is titled "From traditional to deep learning approaches in whole slide image registration: A methodological review."

Statistics:

  • The tumor microenvironment (TME) is a critical area of study in histopathology, essential for understanding cancer progression and treatment outcomes.
  • Over 18 million Americans are diagnosed with cancer each year, highlighting the need for accurate and efficient pathology informatics tools. (Source: American Cancer Society)
  • Histopathology images are typically gigapixel in size, requiring significant computational resources for analysis and registration.
  • The availability of high-quality datasets and tools is a significant challenge in the field of pathology informatics.

Sources:

  • "From traditional to deep learning approaches in whole slide image registration: A methodological review." Journal of Pathology Informatics, 2025;19:100512.
  • NewsRx. Reports Summarize Pathology Informatics Findings from University of Warwick (From traditional to deep learning approaches in whole slide image registration: A methodological review). Information Technology Newsweekly. November 4, 2025; p 607.

Read more