The Intriguing Effect of Frequency Disentangled Learning on Medical Image Segmentation

the ICEBERG Study Group

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva


Deep models have been shown to tend to fit the target function from low to high frequencies (a phenomenon called the frequency principle of deep learning). One may hypothesize that such property can be leveraged for better training of deep learning models, in particular for segmentation tasks where annotated datasets are often small. In this paper, we exploit this property to propose a new training method based on frequency-domain disentanglement. It consists of three main stages. First, it disentangles the image into high- and low-frequency components. Then, the segmentation network model learns them separately (the approach is general and can use any segmentation network as backbone). Finally, feature fusion is performed to complete the downstream task. The method was applied to the segmentation of the red and dentate nuclei in Quantitative Susceptibility Mapping (QSM) data and to three tasks of the Medical Segmentation Decathlon (MSD) challenge under different training sample sizes. For segmenting the red and dentate nuclei and the heart, the proposed approach resulted in considerable improvements over the baseline (respectively between 8 and 16 points of Dice and between 5 and 8 points). On the other hand, there was no improvement for the spleen and the hippocampus. We believe that these intriguing results, which echo theoretical work on the frequency principle of deep learning, are of interest for discussion at the conference. The source code is publicly available at:

Idioma originalInglés
Título de la publicación alojadaMedical Imaging 2024
Subtítulo de la publicación alojadaImage Processing
EditoresOlivier Colliot, Jhimli Mitra
ISBN (versión digital)9781510671560
EstadoPublicada - 2024
Publicado de forma externa
EventoMedical Imaging 2024: Image Processing - San Diego, Estados Unidos
Duración: 19 feb. 202422 feb. 2024

Serie de la publicación

NombreProgress in Biomedical Optics and Imaging - Proceedings of SPIE
ISSN (versión impresa)1605-7422


ConferenciaMedical Imaging 2024: Image Processing
País/TerritorioEstados Unidos
CiudadSan Diego


Profundice en los temas de investigación de 'The Intriguing Effect of Frequency Disentangled Learning on Medical Image Segmentation'. En conjunto forman una huella única.

Citar esto