Modularization of deep networks allows cross-modality reuse: lesson learnt

Fu W, Husvogt L, Ploner S, Fujimoto JG, Maier A (2020)

Publication Type: Conference contribution

Publication year: 2020

Journal

Informatik aktuell Gesellschaft für Informatik (GI)

Publisher: Springer

Pages Range: 274-279

Conference Proceedings Title: Informatik aktuell

Event location: Berlin DE

ISBN: 9783658292669

DOI: 10.1007/978-3-658-29267-6_61

Abstract

Fundus photography and Optical Coherence Tomography Angiography (OCT-A) are two commonly used modalities in ophthalmic imaging. With the development of deep learning algorithms, fundus image processing, especially retinal vessel segmentation, has been extensively studied. Built upon the known operator theory, interpretable deep network pipelines with well-defined modules have been constructed on fundus images. In this work, we firstly train a modularized network pipeline for the task of retinal vessel segmentation on the fundus database DRIVE. The pretrained preprocessing module from the pipeline is then directly transferred onto OCT-A data for image quality enhancement without further fine-tuning. Output images show that the preprocessing net can balance the contrast, suppress noise and thereby produce vessel trees with improved connectivity in both image modalities. The visual impression is confirmed by an observer study with five OCT-A experts. Statistics of the grades by the experts indicate that the transferred module improves both the image quality and the diagnostic quality. Our work provides an example that modules within network pipelines that are built upon the known operator theory facilitate cross-modality reuse without additional training or transfer learning.

Authors with CRIS profile

Weilin Fu Lehrstuhl für Informatik 5 (Mustererkennung) Lennart Husvogt Lehrstuhl für Informatik 5 (Mustererkennung) Stefan Ploner Lehrstuhl für Informatik 5 (Mustererkennung) Andreas Maier Lehrstuhl für Informatik 5 (Mustererkennung)

Involved external institutions

Massachusetts Institute of Technology (MIT) US United States (USA) (US)

How to cite

APA:

Fu, W., Husvogt, L., Ploner, S., Fujimoto, J.G., & Maier, A. (2020). Modularization of deep networks allows cross-modality reuse: lesson learnt. In Thomas Tolxdorff, Thomas M. Deserno, Heinz Handels, Andreas Maier, Klaus H. Maier-Hein, Christoph Palm (Eds.), Informatik aktuell (pp. 274-279). Berlin, DE: Springer.

MLA:

Fu, Weilin, et al. "Modularization of deep networks allows cross-modality reuse: lesson learnt." Proceedings of the International workshop on Algorithmen - Systeme - Anwendungen, 2020, Berlin Ed. Thomas Tolxdorff, Thomas M. Deserno, Heinz Handels, Andreas Maier, Klaus H. Maier-Hein, Christoph Palm, Springer, 2020. 274-279.

BibTeX: Download