Master defense Maximilian Pirson

Last week, our Master's student Maximilian Pirson successfully defended his Master's Thesis entitled "Autoencoder based Anomaly Detection and Dimension Reduction for Abdominal Pressure Prediction based on Gait". As part of his Master's degree in Medical Engineering with focus on research, he was able to work on a project at the Laboratory for Biomechanics for two semesters and carry out research work building on his Master's thesis. He trained various neural networks with simple gait recordings from the markerless motion capture system The Captury. He used a long-short term memory (LSTM)-based structure for the neural network to detect anomalies in the gait and achieve a dimensional reduction of the data. Using this autoencoder, he has also predicted intra-abdominal pressures (IAP) calculated with AnyBody Technology A/S so that gait analyses can also be used to predict IAP. His work deepens the use of AI in biomechanics and combines machine learning with relevant anatomical questions. We would like to thank you for your work and wish you all the best for the future!