For his diploma thesis, engineer Zizhe Wang developed a deep learning framework that uses state-of-the-art generative adversarial networks to generate medical images of bone marrow smears.
Generative Modeling with adversarial networks is a novel branch in deep learning where synthetic data can be generated using real-world examples to train a generator that aims at ‚fooling‘ a discriminator that tries to distinguish real from synthetic data.
Zizhe used StyleGAN2-ADA to generate synthetic image data from bone marrow images with blood cancer. For his outstanding thesis, he received a 1.0 (A+) final grade.
He is going to continue his work with generative networks as a Ph.D. student at the Chair of Software Technology at the Technical University Dresden.
Well done!