Noninvasive Risk Stratification of Prostate Cancer Patients Using Radiomic Features Derived from Magnetic Resonance Fingerprinting (MRF) and MRI

Abstract

This project is aimed at non-invasive risk stratification of prostate cancer patients achieved through development of computer assisted tools using Magnetic Resonance Imaging (MRI) and Magnetic Resonance Fingerprinting (MRF). During the current reporting period, we build upon earlier reported work and have established associations between MR parameters including T1, T2 MRF, T2w and ADC measurements, tissue compartment ratios derived from whole mount pathology within prostate cancer, prostatitis and normal prostate. We have also showed differential correlations between MRF measurements and tissue compartments within prostate cancer and prostatitis, both within central and transition zone. In addition, we have also developed novel radiomic methods for risk stratification of prostate cancer using MRI including peri-tumoral radiomics, delta radiomics and deep learning based features which have resulted in 4 journal publications.

Document Details

• Document Type: Technical Report
• Publication Date: Oct 01, 2020
• Accession Number: AD1129188

Entities

People

• Rakesh Shiradkar

Organizations

• Case Western Reserve University

Tags

Communities of Interest

• Biomedical

DTIC Thesaurus Topics

• Acquisition
• African Americans
• Artificial Intelligence Software
• Biomedical Research
• Coefficients
• Computers
• Convolutional Neural Networks
• Deep Learning
• Detection
• Diffusion Coefficient
• Diseases And Disorders
• Electronic Mail
• Histopathology
• Information Science
• Machine Learning
• Magnetic Resonance
• Magnetic Resonance Imaging
• Medical Personnel
• Neoplasms
• Neural Networks
• Physicians
• Prostate Cancer
• Risk
• Standards
• Students
• United States

Fields of Study

• Medicine
• Physics

Readers

• Medical Imaging.
• Oncology and Biomarker-Based Cancer Detection.

Technology Areas

• AI & ML