AI-Powered Tool Improves Cancer Tissue Analysis

This process is labor-intensive, time-consuming, and often limited to small sample areas, leaving room for variability and inefficiency. Now, a new computational approach can extract detailed cellular and spatial information from tumor tissue slides with unprecedented accuracy and scalability.

Scientists at the Icahn School of Medicine at Mount Sinai (New York City, NY, USA) have developed MARQO, an artificial intelligence (AI)-powered image analysis tool designed to streamline the evaluation of immunohistochemistry (IHC) and immunofluorescence (IF) images. These staining methods are widely used to detect immune cells and biomarkers in cancer tissues, and MARQO enhances their analysis by fully integrating whole-slide processing. The platform works across multiple staining technologies, improving reproducibility and enabling comparisons between studies.

MARQO addresses three major challenges in digital pathology. Unlike tools that require chopping slides into patches or costly computing clusters, MARQO processes intact slides in minutes using standard GPUs. It supports a wide range of IHC and IF stains, facilitating cross-study analysis, and it automatically identifies positive cells, records coordinates and marker intensities, and passes results to pathologists for validation. This preserves expert oversight while automating the most laborious steps.

The study, published in Nature Biomedical Engineering, highlights MARQO’s potential to transform cancer research workflows. While currently designed for research purposes, its compatibility with standard clinical staining methods suggests that future applications in diagnostic pathology are possible. The team plans to improve the user interface, expand spatial analysis tools, and adapt the platform for high-performance computing to support studies involving millions of tissue slides.

By accelerating image analysis, MARQO could play a vital role in biomarker discovery, improving predictions of which patients might benefit from specific treatments, and enhancing the precision of cancer diagnostics. The ability to quickly generate structured data from complex slides represents a significant step toward more efficient and personalized oncology.

“We designed MARQO to fill a major gap in the field: turning complex whole-slide images into usable, structured data quickly and consistently,” said Sacha Gnjatic, PhD, senior author of the study. “By automating the heavy lifting, we let experts focus on interpretation and discovery. This platform could accelerate biomarker discovery, improve how we predict which patients will benefit from specific treatments, and ultimately support the development of more precise cancer diagnostics.”

Related Links:
Icahn School of Medicine