Presentation
2 March 2022 Risk stratification of colorectal cancer using artificial intelligence based analysis of nanoscale chromatin modifications
Author Affiliations +
Abstract
Partial Wave Spectroscopy (PWS) is a nanoscale sensitive imaging technique, emerging as a potent modality for minimally invasive and cost-efficient early risk stratification for colorectal cancer. Feature engineering on the PWS map of nuclear chromatin structures, we developed 50 features extracted from the map as potentially representative of future colorectal cancer development risk. Along with feature extraction utilizing convolutional neural networks, classification model is developed with 10-fold cross validation from a multicenter data consisting of a total of 187 patients. Our newly developed feature engineering results in improved performance of the early risk stratification model for colorectal cancer.
Conference Presentation
© (2022) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Andrew S. Chang, Sravya Prabhala, Ali Daneshkhah, Hariharan Subramanian, and Vadim Backman "Risk stratification of colorectal cancer using artificial intelligence based analysis of nanoscale chromatin modifications", Proc. SPIE PC11972, Label-free Biomedical Imaging and Sensing (LBIS) 2022, PC1197208 (2 March 2022); https://doi.org/10.1117/12.2608366
Advertisement
Advertisement
KEYWORDS
Colorectal cancer

Spectroscopy

Artificial intelligence

Microscopy

Feature extraction

Convolutional neural networks

Data modeling

Back to Top