Related Experiment Video
Updated: Oct 24, 2025
07:54
Heterogeneity Mapping of Protein Expression in Tumors using Quantitative Immunofluorescence
Published on: October 25, 2011
18.8K
Hierarchical cancer heterogeneity analysis based on histopathological imaging features
Mingyang Ren1, Qingzhao Zhang2, Sanguo Zhang1
1School of Mathematics Sciences, University of Chinese Academy of Sciences, Key Laboratory of Big Data Mining and Knowledge Management, Chinese Academy of Sciences, Beijing, China.
Biometrics
|August 14, 2021
Summary
This study introduces a novel supervised heterogeneity analysis for cancer research using hierarchical histopathological imaging features. The method offers improved prediction and stability for cancer outcome modeling.
Area of Science:
- Oncology
- Computational Pathology
- Bioinformatics
Background:
- Supervised heterogeneity analysis is crucial in cancer research, traditionally relying on clinical, demographic, or molecular data.
- Histopathological imaging features from biopsies are emerging as powerful tools for modeling cancer outcomes.
- Existing analyses using imaging features have not addressed hierarchical structures.
Purpose of the Study:
- To perform the first supervised cancer heterogeneity analysis with a hierarchical structure.
- To integrate two types of histopathological imaging features: biologically informed and automated.
- To develop a novel penalization approach for this hierarchical analysis.
Main Methods:
- Extraction of two types of histopathological imaging features (biologically informed and automated).
- Development of a novel penalization approach inspired by, but distinct from, penalized fusion and sparse group penalization.
- Application of the method to lung adenocarcinoma data.
Main Results:
- The developed penalization approach demonstrated satisfactory statistical and numerical properties.
- The hierarchical analysis identified a heterogeneity structure distinct from alternative methods.
- The model achieved satisfactory prediction and stability performance on lung adenocarcinoma data.
Conclusions:
- The novel method successfully integrates hierarchical histopathological imaging features for supervised cancer heterogeneity analysis.
- This approach provides a more refined understanding of cancer heterogeneity compared to traditional methods.
- The findings suggest significant potential for improved cancer outcome prediction and personalized treatment strategies.
