Convolutional autoencoders to process 4D gynaecological data

Finished: 2020-08-25

MSc assignment

In the GynIUS project we are collecting a huge data set of pelvic floor ultrasounds of female patients visiting a clincic with pelvic floor complaints. However the understanding of how to interpret this data is limited. In the gyneacological field there is a debate on how to determine what type of problem there is in the data.  Since we have a big dataset with patients with different symptoms we can look if we can use unsupervised deep learining to find relevant clustering of patients groups. And use this clusters to get a better insight into the different pelvic floor patient groups. 

The goal of the project is to use (un)supervised deep learning to find clinically relevant parameters in pelvic floor ultrasound data.