Back to all Internships




Université Libre de Bruxelles

3D Electron Microscopy Image Segmentation of Golgi Apparatus, Mitochondria and Endoplasmic Reticulum

Company description

LISA - Image is involved in research and development in two main topics of applied sciences : image analysis and computer graphics.

In the field of image analysis and pattern recognition, LISA - Image develops new methods for object segmentation in digital images, based on classification and decision theory, as well as for tracking mobile objects in image sequences. Developed algorithms and applications cover 2D and 3D problems and are most of the time related to medical problems such as: automatic in vitro 2D & 3D cell tracking for characterization of cell migration and proliferation, characterization of pigmented skin lesions, tissue-based biomarkers, pulmonary emphysema and articulation cartilage. We are currently offering a number of intership and master thesis projects. 

Job description

The internship project will address the problem of analysis of high-resolution electron microscopy 3D images of cells. In these images, cellular organelles are clearly resolved simultaneously. However, the organelles of interest are easily visually recognizable only when inspected in 3D volumes. The most important cellular organelles that we aim to inspect in this project are: Golgi apparatus; mitochondria and endoplasmic reticulum. The ultimate goal is to develop automated method for segmentation and quantification of important features of these organelles. This is a challenging task given the presence of noise in the images, size of the images, and the organelle’s complex shape. All organelles of interest have a similar shape: a tubular or a flattened tubular. However, their size, intensity and the tubular shape characteristics are variable. The complex task of cellular organelles automated segmentation can be broken down to several more simple tasks: 3D visualization of data, software development of a manual labeling tool, denoising of images, automated segmentation of tubular structures of interest using manually labeled data, quantification of features of interest.

The applicant should satisfy the following requirements: choice of image processing subjects, experience in programing languages (preferably MATLAB or ImageJ or C++ or Python).  


Searched Profile

  • Experience and interest in computer vision (image processing & machine learning)
  • Programing skills: preferably experience or eagerness to learn Matlab or C++ or Python or ImageJ
  • Good knowledge of English
  • Creative
  • Well-organized

We offer

We offer an oportunity to work in a stimulating enviroment in one of the best Belgian universities in a highly popular field of computer vision. Additionally, a multidisciplinary topic of the internship is motivativating, interesting, and offers a good experiance for vast oportunities in the job market for both academia and industry. You will be guided and helped by experienced universtiy researchers, and you will have oportunity to work with highest quality images made using the lastest technology biomedical scanners. 

Apply for this job
Only for undergraduate students

Articles you may want to read

Similar internships