Generating and analyzing Synthetic data for DeepSynthBody.org

In this project, different deep generative models will be implemented to generate synthetic medical data for deepsynthbody.org. Student want to learn and implement state-of-the-art deep generative modles for a selected medical dataset.
Master

Collecting and annotating medical domain data is time-consuming and costly. However, deep learning algorithms depend on high-quality datasets and annotations. In this regard, synthetic data shows promising directions to overcome this data deficiency problem in the medical domain. DeepSynthBody (DeepSynthBody.org) is the beginning of such an open-source project aiming at providing high-quality synthetic data.

The following sub-projects are available under this big project:

Sub-project 1: Generating medical data using Generative Adversarial Networks
Sub-project 2: Generating medical data using Diffusion Models
Sub-project 3: Conditional GANs and Diffusion models

Goal

Implement, train and evaluate different deep generative models for generating synthetic data for a selected medical dataset.

Learning outcome

  • Interdisciplinary research
  • Machine learning / AI (deep generative models)
  • Performing research with advanced deep learning algorithms

Qualifications

  • Hard-working
  • Motivated
  • Interested in learning (the rest can be learned during the thesis work)

Supervisors

  • Pål Halvorsen
  • Michael Riegler
  • Vajira Thambawita

References

 


 

Contact person