About ETRO  |  News  |  Events  |  Vacancies  |  Contact  
Home Research Education Industry Publications About ETRO

Master theses

Current and past ideas and concepts for Master Theses.

Analysing cellular dynamics from single-cell segmentation in microscopic images


Billions of euros are invested annually into the clinical evaluation of new drug compounds, yet only a few drugs are eventually approved for market access. Automated analysis of cellular dynamics provides a mean to monitor the effect of external factors in living cells, and is a key technique to streamline the development of new drugs. The first step in such analysis consist in detecting and separating each individual cell from the other cells in a given microscopic image, a process known as single-cell image segmentation. The total number of images and cells can be substantially large, and therefore the segmentation task should be done automatically by a program with minimal human intervention.

This thesis presents the application of Computer Vision and Machine Learning techniques as a mean to automate this process. It is part of a multi-disciplinary research collaboration between the Electronics and Informatics Department (VUB-ETRO) and the Structural Biology group (VUB-SBB), and involves a combination of biology experiments, data acquisition, image processing, computer vision & machine learning.

Kind of work

The proposed research is focused on automatic segmentation of individual cells in time-lapse microscopy images. Currently, there is a number of image processing solutions [1,2,3] to the cells segmentation problem, broadly organized in two groups: unsupervised and supervised segmentation.
The student will compare some of these solutions in the context of unsupervised image segmentation, and will acquire experience in running related solution scripts in local servers, high performance computing (HPC) nodes, and/or in cloud-alike (big data) environments.

Framework of the Thesis

[1] Alioscha-Perez, Mitchel, Ronnie Willaert, and Hichem Sahli. "A segmentation framework for phase contrast and fluorescence microscopy images." International Journal of Pattern Recognition and Artificial Intelligence 28.07 (2014): 1460013.
[2] Ronneberger, Olaf and Fischer, Philipp and Brox, Thomas. U-net: Convolutional networks for biomedical image segmentation. International Conference on Medical image computing and computer-assisted intervention. 2015 234 241. Springer.
[3] Kraus OZ, Ba JL, Frey BJ. Classifying and segmenting microscopy images with deep multiple instance learning. Bioinformatics. 201632(12):i52-i59. doi:10.1093/bioinformatics/btw252.

In collaboration with Prof. Ronnie Willaert (SBB) office E.4.01 (building E)

Expected Student Profile

  • Following a MSc in a field related to one or more of the following: biomedical engineering, electrical engineering, computer science, applied mathematics
  • Basic knowledge of image processing (data structure, reading, etc.)
  • Willing to work with matlab and python (keras+tensorflow), in HPC clusters and big data platforms


Prof. Hichem Sahli

+32 (0)2 629 2916

more info


Dr. Mitchel Perez Gonzalez

+32 (0)2 629 2986

more info

- Contact person




- Contact person

- Thesis proposals

- ETRO Courses

- Contact person

- Spin-offs

- Know How

- Journals

- Conferences

- Books

- Vacancies

- News

- Events

- Press


ETRO Department

Tel: +32 2 629 29 30

©2019 • Vrije Universiteit Brussel • ETRO Dept. • Pleinlaan 2 • 1050 Brussels • Tel: +32 2 629 2930 (secretariat) • Fax: +32 2 629 2883 • WebmasterDisclaimer