S039 - Equivariant and Denoising CNNs to Decouple Intensity and Spatial Features for Motion Tracking in Fetal Brain MRI
Benjamin Billot, Daniel Moyer, Neerav Karani, Malte Hoffmann, Esra Abaci Turk, Ellen Grant, Polina Golland
Equivariance in convolutional neural networks (CNN) has been a long-sought property, as it would ensure robustness to expected effects in the data. Convolutional filters are by nature translation-equivariant, and rotation-equivariant kernels were proposed recently. While these filters can be paired with learnable weights to form equivariant networks (E-CNN), we show here that such E-CNNs have a limited learning capacity, which makes them fragile against even slight changes in intensity distribution. This sensitivity to intensity changes presents a major challenge in medical imaging where many noise sources can randomly corrupt the data, even for consecutive scans of the same subject. Here, we propose a hybrid architecture that successively decouples intensity and spatial features: we first remove irrelevant noise in the data with a denoising CNN, and then use an E-CNN to extract robust spatial features. We demonstrate our method for motion tracking in fetal brain MRI, where it considerably outperforms standard CNNs and E-CNNs.
Schedule: Tuesday, July 11: Posters — 10:30–12:00 & 15:00–16:00
Poster location: T42