Machine learning approaches for spatial omics data analysis in digital pathology: tools and applications in genitourinary oncology

Aug 4, 2026

Upload by Ezra

Recent advances in spatial omics technologies enable in situ analysis of tissue morphology, cell composition, and biomolecule expression, accelerating computational tool development in digital pathology. This review surveys current computational methods for analyzing spatially mapped omics data from digitized histopathology slides and related materials, with emphasis on tools and applications for genitourinary oncology. It covers image-processing approaches, integration of machine learning with spatially resolved omics, and outlines limitations and future directions for clinical decision-making.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Introduction

Spatially resolved omics and machine learning opportunities

Current state and translational gap

TYPE Review PUBLISHED 29 November 2024 DOI 10.3389/fonc.2024.1465098 OPEN ACCESS EDITED BY Martin King, Harvard Medical School, United States REVIEWED BY Murat Akand, University Hospitals Leuven, Belgium Jianing Xi, Guangzhou Medical University, China *CORRESPONDENCE Sungyong You [Sungyong.You@cshs.org](Sungyong.You@cshs.org) †These authors have contributed equally to this work RECEIVED 15 July 2024 ACCEPTED 08 November 2024 PUBLISHED 29 November 2024 CITATION Kim H, Kim J, Yeon SY and You S (2024) Machine learning approaches for spatial omics data analysis in digital pathology: tools and applications in genitourinary oncology. Front. Oncol. 14:1465098 . doi: 10.3389/fonc.2024.1465098 COPYRIGHT © 2024 Kim, Kim, Yeon and You. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Machine learning approaches for spatial omics data analysis in digital pathology: tools and applications in genitourinary oncology Hojung Kim 1,2†, Jina Kim 1,3†, Su Yeon Yeon 2† and Sungyong You 1,3* 1 Department of Urology, Cedars-Sinai Medical Center, Los Angeles, CA, United States, 2 Department of Pathology, University of Illinois at Chicago, Chicago, IL, United States, 3 Department of Computational Biomedicine, Cedars-Sinai Medical Center, Los Angeles, CA, United States Recent advances in spatial omics technologies have enabled new approaches for analyzing tissue morphology, cell composition, and biomolecule expression patterns in situ. These advances are promoting the development of new computational tools and quantitative techniques in the emerging ﬁeld of digital pathology. In this review, we survey current trends in the development of computational methods for spatially mapped omics data analysis using digitized histopathology slides and supplementary materials, with an emphasis on tools and applications relevant to genitourinary oncological research. Thereview contains three sections: 1) an overview of image processing approaches for histopathology slide analysis; 2) machine learning integration with spatially resolved omics data analysis; 3) a discussion of current limitations and future directions for integration of machine learning in the clinical decisionmaking process. KEYWORDS machine learning, spatial omics, digital pathology, genitourinary, oncology 1 Introduction Over the last decade, automation and digitization of laboratory processes have slowly transformed everyday practices in hospitals. Recent advances in computational pathology, especially with machine learning (ML) suggest an imminent revolution in the clinical decision-making process (1). There has been a steady build-up of resources to further support this transition. Adoption of digital pathology is becoming more common among medical centers, and large repositories such as The Cancer Genome Atlas (TCGA) have steadily been accruing digital tissue slides complemented by multi-omics proﬁles of each Frontiers in Oncology 01 [frontiersin.org](frontiersin.org) sample ([https://www.cancer.gov/tcga](https://www.cancer.gov/tcga)). There have been ongoing research efforts to adopt ML and artiﬁcial intelligence (AI) for image analysis, with several tools already approved by the Food and Drug Administration (FDA) for use with radiologic images ([https://www.fda.gov/](https://www.fda.gov/)). Utilizing ML for digital slide images analysis could help clinicians not only with diagnosis but also with risk stratiﬁcation through predicting genetic alterations and classifying tumors based on meaningful features. Features such as microscopic mor

FAQs

What does the review focus on in digital pathology?

It surveys computational methods for analyzing spatially mapped omics data using digitized histopathology slides, highlighting tools and applications relevant to genitourinary oncology.

Which main topics does the review cover?

It covers image-processing approaches for histopathology slide analysis, machine-learning integration with spatially resolved omics analysis, and limitations plus future directions for clinical integration.

How can spatially resolved omics improve on traditional molecular testing?

By analyzing at single-cell and spatial multi-omics levels, it preserves spatial information that links expression profiles with histological context, helping detect tumor microenvironment features and support biomarker discovery.

Aug 4, 2026

Upload by Ezra