Related Experiment Video
Updated: Jun 30, 2026
07:31
Establishment of Pancreatic Cancer-Derived Tumor Organoids and Fibroblasts From Fresh Tissue
Published on: May 26, 2023
Foundation Models in Cancer Pathology: Techniques, Applications, and Future Directions
Bo Zhang1, Victor Yu Cui2, Tong Wu3
1Center of Clinical Big Data and Analytics of the Second Affiliated Hospital and the School of Public Health, Zhejiang University School of Medicine, Hangzhou 310058, China.
Research (Washington, D.C.)
|June 29, 2026
Summary
Computational pathology foundation models (CPathFMs) offer reusable representations for cancer analysis, moving beyond task-specific AI. Further research is needed to establish their clinical value and ensure broad applicability.
Area of Science:
- Digital pathology
- Artificial intelligence in medicine
- Machine learning for healthcare
Background:
- Conventional deep learning in computational pathology requires extensive annotations and is task-specific.
- Recent advancements include self-supervised learning, transformers, and large-scale pretraining.
- Computational pathology foundation models (CPathFMs) learn transferable representations from diverse pathology data.
Purpose of the Study:
- To review the development and applications of CPathFMs.
- To identify challenges and future directions for CPathFM deployment in clinical settings.
Main Methods:
- Examination of data resources, backbone architectures, representation levels, input modalities, and pretraining strategies for CPathFMs.
- Analysis of CPathFM applications in cancer pathology tasks.
- Discussion of limitations and ethical considerations.
Main Results:
- CPathFMs leverage diverse data and architectures for transferable representation learning.
- Applications span tumor detection, grading, biomarker prediction, prognosis, and report generation.
- Key challenges include data representation, generalization, standardization, costs, and ethics.
Conclusions:
- CPathFMs represent a shift towards reusable AI in pathology.
- Clinical utility requires improved performance, generalizability, standardized evaluation, and real-world validation.
Related Concept Videos
02:43Mouse Models of Cancer Study
Mice have long served as models for studying human biology and pathology because of their phylogenetic and physiological similarity with humans. They are also easy to maintain and breed in the laboratory, and hence, many inbred strains are now available for research. Studies on mice have contributed immeasurably to our understanding of cancer biology.
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
The development of transgenic, knockout, and knock-in mice has led to an exponential increase in their use as model organisms in research,...
02:12What is Cancer?
Cells and tissues must meticulously coordinate their activities for the normal functioning of the human body. Therefore, they exhibit socially responsible behavior - resting, growing, dividing, differentiating, or dying - for the organism’s benefit. Cancer arises when cells divide uncontrollably and invade other tissues or organs.
Although people have known about cancer for centuries, it was only in 1761 that Giovanni Morgagni of Padua performed a detailed autopsy of patients who died from...
Although people have known about cancer for centuries, it was only in 1761 that Giovanni Morgagni of Padua performed a detailed autopsy of patients who died from...