Automatic breach detection during spine pedicle drilling based on vibroacoustic sensing

no code implementations • 27 Mar 2023 • Aidana Massalimova, Maikel Timmermans, Nicola Cavalcanti, Daniel Suter, Matthias Seibold, Fabio Carrillo, Christoph J. Laux, Reto Sutter, Mazda Farshad, Kathleen Denis, Philipp Fürnstahl

The best-performing data fusion model combined the latter two sensors with a breach recall of 98\%.

Sensor Fusion

Improved Techniques for the Conditional Generative Augmentation of Clinical Audio Data

no code implementations • 5 Nov 2022 • Mane Margaryan, Matthias Seibold, Indu Joshi, Mazda Farshad, Philipp Fürnstahl, Nassir Navab

In contrast to previously proposed fully convolutional models, the proposed model implements residual Squeeze and Excitation modules in the generator architecture.

Data Augmentation

Conditional Generative Data Augmentation for Clinical Audio Datasets

no code implementations • 22 Mar 2022 • Matthias Seibold, Armando Hoch, Mazda Farshad, Nassir Navab, Philipp Fürnstahl

In this work, we propose a novel data augmentation method for clinical audio datasets based on a conditional Wasserstein Generative Adversarial Network with Gradient Penalty (cWGAN-GP), operating on log-mel spectrograms.

Data Augmentation Generative Adversarial Network