Computational Pathology in Ovarian Cancer
Principal Investigator
Name
Sandra Orsulic
Degrees
PhD
Institution
Regents University of California Los Angeles
Position Title
Professor
Email
About this CDAS Project
Study
PLCO
(Learn more about this study)
Project ID
PLCOI-1132
Initial CDAS Request Approval
Dec 19, 2022
Title
Computational Pathology in Ovarian Cancer
Summary
Our goal is to identify high-grade serous ovarian carcinoma image features as biomarkers of clinical outcomes or specific molecular characteristics, such as aneuploidy, genomic instability, and homologous recombination defects. We will extract image features from the PLCO H&E pathology images (40x) of ovarian cancers and analyze the data as a validation dataset for computational image features that we have identified as putative biomarkers of clinical outcomes and molecular characteristics in the ovarian TCGA dataset and our institutional datasets.
Aims
1. Extract pyradiomics features from 40x whole slide images of high-grade serous ovarian carcinomas and use statistical methods to select relevant pyradiomics features
3. Use statistical methods to correlate the image features with clinical variables and molecular data
4. Compare the data from the PLCO cohort with the data from the TCGA cohort and our institutional cohort
Collaborators
Arkadiusz Gertych - Cedars-Sinai Medical Center
Amarjot Singh - University of California Los Angeles