Biomedical Imaging Informatics
Imaging informatics methods describe, store, analyze, share, and visualize images and their associated data to refine knowledge and inform decision makers. Researchers at the Bio-MIBLab have developed novel imaging informatics methods for traditional histopathological images and for nanoparticle-tagged tissue images. Notably, we have developed methods to handle “Big Data” in the form of whole-slide images (WSIs), each of which may contain billions of pixels.
Currently, tumor biopsy analysis, called histopathology, is a common clinical procedure for diagnosing cancer presence, type, and progression. While diagnosing patients using biopsy slides, pathologists manually assess nuclear characteristics. However, making decisions manually from a slide with millions of nuclei can be time-consuming and subjective. Researchers have proposed clinical decision support systems (CDSSs) that can help in decision making. However, these systems have limited reproducibility. The development of robust CDSSs for biopsy images faces several informatics challenges: (1) Lack of robust segmentation methods, (2) Semantic gap between quantitative information and pathologists’ knowledge, (3) Lack of batch-invariant imaging informatics methods, (4) Lack of knowledge models for capturing informative patterns in large biopsy slides, and (5) Lack of guidelines for optimizing and validating diagnostic models. We have conducted advanced imaging informatics research to overcome these challenges and developed novel methods to extract information from biopsy images, to model knowledge embedded in large histopathological datasets, such as The Cancer Genome Atlas (TCGA), and to assist decision making with biological and clinical validation.
- Feature Extraction
- Modeling and Prediction
This pipeline is summarized in the following publication: