Introduction

The HistoMaps project, funded by the Medical Research Council (MRC), is a research project led by Dr. Raza at TIA Centre. This project aims to revolutionise the way we analyse and understand the tumour microenvironment (TME) in cancer tissues using advanced computational techniques. Cancer diagnosis and treatment heavily rely on the analysis of tissue samples stained with various dyes to highlight different cellular components. Traditional methods involve examining these stained samples under a microscope, which can be time-consuming and subject to human error. The complexity and vast size of whole slide images (WSIs) present significant challenges for analysis, especially when dealing with diverse types of staining.

HistoMaps aims to overcome these challenges by developing new algorithms that can analyse WSIs in a stain-agnostic manner. This means the algorithms will be able to process images stained with different dyes, such as Haematoxylin and Eosin (HE), Immunohistochemistry (IHC), and Multiplexed Immunofluorescence (MxIF), without losing essential spatial information. The goal is to create comprehensive "maps" of the tumour microenvironment that can reveal clinically relevant traits and patterns.

The HistoMaps project promises to significantly enhance our understanding of the tumour microenvironment, which is crucial for improving cancer diagnosis and treatment. By providing a detailed and accurate analysis of WSIs, the project aims to assist oncologists in making more informed decisions regarding therapeutic interventions. Additionally, the insights gained from HistoMaps can help pharmaceutical companies develop new targets for cancer treatment.

HistoMaps represents a significant step forward in the field of computational pathology. By leveraging advanced image analysis techniques, the project aims to unlock new insights into the tumour microenvironment, leading to better cancer diagnosis, treatment, and research. This innovative approach has the potential to transform the way we understand and combat cancer, making a lasting impact on patient care and outcomes.

Collaborators

The project is a collaborative effort involving experts from various fields, including computer science, pathology, and oncology. Key collaborators include Prof Lawrence Young from Warwick Medical School, Prof David Snead from University Hospitals Coventry and Warwickshire, and Prof Nick James from the Institute of Cancer Research. The project also benefits from the support of Nvidia, which provides technological expertise and resources for optimising the algorithms.

Key Objectives

- Design and develop new algorithms that can efficiently analyse WSIs, reducing redundant information while preserving crucial data, including spatial relationships.

- Create methodologies that can analyse diverse types of staining, making the analysis applicable to both brightfield and multiplexed microscopy.

- Develop methods to align images from different staining techniques, allowing the creation of unified HistoMaps.

- Link the features identified in HistoMaps to clinical variables such as mutations, survival rates, and response to therapy.

Alignment with NHS Long term plan, NHS or wider government priorities

The HistoMaps project aims to revolutionise cancer diagnosis and treatment by developing advanced algorithms for analysing whole slide images (WSIs) of tumour tissues. By creating stain-agnostic feature representations, these algorithms will efficiently process images stained with various dyes, such as Haematoxylin and Eosin (HE), Immunohistochemistry (IHC), and Multiplexed Immunofluorescence (MxIF), without losing essential spatial information. This innovative approach will provide comprehensive maps of the tumour microenvironment, revealing clinically relevant traits and patterns. The project aligns with the NHS Long Term Plan by improving cancer diagnosis, leveraging technology, enhancing efficiency, supporting personalised medicine, addressing health inequalities, fostering collaboration, and ensuring cost-effectiveness. Through collaboration with experts and industry partners like Nvidia, HistoMaps aims to transform cancer research and patient care, leading to better outcomes and more informed therapeutic decisions.

Project Partners

- NVIDIA Limited (UK)

Publications

Al-Rubaian, A. et al. (2024) ‘Cell Maps Representation for Lung Adenocarcinoma Growth Patterns Classification in Whole Slide Images’, in 2024 IEEE International Symposium on Biomedical Imaging (ISBI). IEEE, pp. 1–5. Available at: https://doi.org/10.1109/ISBI56570.2024.10635418.

Al-Rubaian, A. et al. (2025) ‘CellOMaps: A Compact Representation for Robust Classification of Lung Adenocarcinoma Growth Patterns’. Available at: http://arxiv.org/abs/2501.08094.

Lv, J., Antonowicz, S.S. and Raza, S.E.A. (2025) ‘Deep Learning Based Segmentation of Blood Vessels from H&E Stained Oesophageal Adenocarcinoma Whole-Slide Images’. Available at:http://arxiv.org/abs/2501.12323

Jeyasangar, A., Alsalemi, A. and Raza, S.E.A. (2024) ‘Nuclei-Location Based Point Set Registration of Multi-stained Whole Slide Images’, in, pp. 372–386. Available at: https://doi.org/10.1007/978-3-031-66955-2_26.