Automated Machine Learning in Medical Image Segmentation

Master thesis (2021)

Authors

C.S.O. van Gruijthuijsen Mechanical Engineering

Contributors

W. Pan Robot Dynamics - Mechanical, Maritime and Materials Engineering (supervisor 1)
Martijn Starmans Erasmus MC (supervisor 1)
Karin van Garderen Erasmus MC (supervisor 1)
Stefan Klein Erasmus MC (supervisor 2)
M. Wisse Robot Dynamics - Mechanical, Maritime and Materials Engineering (coach)
Faculty
Mechanical Engineering
Copyright: © 2021 Coen van Gruijthuijsen
Automated Machine Learning Convolutional Neural Networks (CNNs) Hyperparameter Optimization Semantic Segmentation
More Info
expand_more

Published Date

05-11-2021

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Mechanical Engineering

Abstract

Semantic Segmentation of medical images are used to improve diagnosis and treatment. In recent years, the application of machine learning methods are increasingly used. However, the design of these models is difficult and time-consuming. In this thesis, we investigated the automation of this process using an Automated Machine Learning (AutoML) algorithm called Bayesian Optimization with HyperBand (BOHB). We found several hyperparameters significantly influencing the performance and confirmed that BOHB is applicable to these kind of problems. However, we did not find a significant difference between BOHB and Random Search on a small search, stressing the need of matching the search space to the search strategy. Finally, our approach was only slightly worse than the current state-of-the-art, which shows that AutoML has potential in this field.