Stud.IP Uni Oldenburg
University of Oldenburg
29.11.2021 22:17:51
ThesisTopics

Personal details

Title Data deconvolution for biotech applications
Description

The Amir team is working on the development of a scanning microscopy technique to support biochemists in the evaluation of the structure of proteins.

The extraction of the results requires a post-processing step taking advantage of a neural network (NN).

In order to do so, the training of the NN is required. The aim of the thesis is the implementation of the workflow that is necessary to achieve the training and the comparaison of the NN prediction with confirmation dataset.

Home institution Department of Computing Science
Type of work practical / application-focused
Type of thesis Bachelor's or Master's degree
Author Dr. Matthieu Gaudet
Status available
Problem statement

The tasks of the student will be to:

  1. set up the structure of the NN

  2. train the NN using already an existing dataset

  3. run the tests to evaluate the efficiency of the trained NN

Requirement
  • First Experience in setting up a neural network

  • Programming in python (PyTorch)

Created 18/10/21

Study data

Departments No departments assigned
Degree programmes
  • Master's Programme Computing Science
  • Bachelor's Programme Computing Science
Assigned courses No courses assigned
Contact person