Automated segmentation of visceral organs and tissues in whole body MRI.
Principal Investigator:
Dr Sergios Gatidis
Approved Research ID:
40040
Approval date:
June 18th 2019
Lay summary
In this research project we want to build a program that can find organs automatically in medical images. In the UK Biobank imaging study, thousands of volunteers are examined using Magnetic Resonance Imaging (MRI). The resulting data can give insights into processes within the human body. However, it would take too much time to analyze all the data manually. Thus, an automated analysis program is necessary. In our project we will use machine learning methods in order to create a tool for automated detection and delineation of organs on MRI. In order to do so, we will train a machine learning algorithm by presenting data that were already analyzed by humans. As a next step, this algorithm can learn to fulfill the respective task automatically. We hope that this project will contribute to a faster and better analysis of MRI data in the context of the UK Biobank MR study.