DigiTIL, a Computational Histomorphometric Predictor of Disease Recurrence and Overall Survival for p16-Positive Oropharyngeal Squamous Cell Carcinoma

Abstract

In this project, we propose to develop DigiTIL, a novel computational tool for risk stratification of p16-positive oropharyngeal squamous cell carcinoma (OPSCC). The tumor microenvironment on digitized H and E images will be characterized using image-extracted features and predictors will be built using machine learning approaches. DigiTIL will be independently validated on a dataset from 7 institutions across the US and two well documented Clinical Trials (RTOG-5022 and RTOG-0129). The predictions made by DigiTIL will be compared against clinical/pathological variables and human estimations. Finally, DigiTIL will be used to identify possible population specific morphologic differences in the tissue phenotype of OPSCC, e.g., between African/Caucasian Americans and Veterans/Non-Veterans.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 2023
Accession Number
AD1207634

Entities

People

  • Germán Corredor

Organizations

  • Emory University

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Abstracts
  • Biological Markers
  • Biomedical Research
  • Breast Cancer
  • Cells
  • Clinical Trials
  • Diseases
  • Image Processing
  • Lymphocytes
  • Machine Learning
  • Maryland
  • Medical Personnel
  • Neoplasms
  • Papillomavirus Infections
  • Statistical Analysis
  • Teamwork
  • United States
  • Universities

Fields of Study

  • Medicine

Readers

  • Oncology and Biomarker-Based Cancer Detection.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks