Skip to Main Content

An official website of the United States government

Computational pathological prognostic markers of colorectal cancer based on tumor heterogeneity mining and their biological interpretability

Principal Investigator
Name
Yanlong Liu
Degrees
Ph.D.
Institution
Harbin Medical University Cancer Hospital
Position Title
Professor
Email
About this CDAS Project
Study
PLCO (Learn more about this study)
Project ID
PLCOI-1520
Initial CDAS Request Approval
Apr 2, 2024
Title
Computational pathological prognostic markers of colorectal cancer based on tumor heterogeneity mining and their biological interpretability
Summary
China is the country with the highest incidence and mortality of Gastric Cancer (GC). The survival time of patients is short and the prognosis is poor. How to break through the current treatment dilemma and improve the survival rate of GC patients has become an urgent clinical problem to be solved. Whole Slide Image (WSI) scanning provides low-cost, easily accessible and informative data for the assessment of tumor heterogeneity.
In the field of Machine Learning (ML), especially with the rapid development of key technologies such as Artificial Intelligence (AI) and Deep Learning (DL), We are able to identify and extract features that are imperceptible to humans directly from biomedical data. In practice, AI-assisted medical devices have been widely used in the field of image analysis. Combining DL techniques with computational pathology of WSI opens up new opportunities for cancer research and precision medicine.
This project intends to use the Computational Pathology (CPATH) to develop a platform to directly mine the morphological features related to tumor heterogeneity in a data-driven and unbiased manner. Through this platform, accurate extraction of gastric cancer heterogeneity features can be achieved, and then precise patient stratification can be realized. To provide a basis for personalized treatment of gastric cancer, and to provide new targets and ways for the treatment of gastric cancer. At the same time, through the testing, optimization and promotion of the platform, the clinical transformation of the artificial intelligence-based computational pathology platform will be truly realized.
Aims

This project covers the whole research process from computational pathology image analysis to clinical prognosis prediction, and then to functional study of prognostic biomarkers. The aim of this study is to provide a new perspective and tool for the prognosis evaluation and personalized treatment of colorectal cancer, and provide new insights into the molecular mechanism of colorectal cancer by integrating computational pathology, bioinformatics, molecular biology, pathophysiology and other multidisciplinary methods. Specific research objectives are as follows:
① To establish a high-quality multi-class HE image annotation dataset of colorectal cancer;
② To develop a multi-modality fusion prognostic stratification model and biomarkers for colorectal cancer based on tumor heterogeneity;
③ To systematically explore and verify the biological interpretability of prognostic biomarkers for colorectal cancer.

Collaborators

Lou Shenghan，Harbin Medical University Cancer Hospital