Abstract: Shape-based Segmentation of Retinal Layers and Fluids in OCT Image Data

Optical coherence tomography (OCT) offers non-invasive imaging of the retina and has been well established in the field of ophthalmology for several decades. Based on its high-resolution cross-sectional images, OCT supports diagnosis of various eye diseases. For clinical examination and treatment planning, automated segmentation of individual retinal layers and pathologies is helpful. Retinal layers follow a strict topology that is not addressed by most state-of-the-art methods. While graph-based methods can be used to correct for topology errors, their application is costly and complex, especially in the presence of pathologies. In this work, we propose a segmentation method that uses shape information of retinal layers for improved topology preservation while providing a simple applicability. For this purpose, we use a multi-task framework based on the U-Net architecture that integrates regression of the shape information of the retinal layers in addition to pixel-wise classification. On the one hand, shape-based regression of retinal layers provides spatial regularization and enables the generation of plausible segmentations. On the other hand, the simultaneous classification results in sharper delineations of pathological structures. Furthermore, a task consistency enables semi-supervised training, allowing the use of unlabeled image data for training. A comprehensive evaluation of our segmentation method is performed using OCT image data from patients with diabetic macular edema. Both an ablation study and comparison with other methods have shown the advantages of our proposed segmentation framework. Learning multiple tasks via separate paths leads to higher generalization and prevents overfitting of network parameters. Simultaneously, feature representations preferred by both tasks are learned. Results also show improved preservation of retinal topology while maintaining sharp pathology delineation [1].

Authors and Affiliations

1. German Research Center for Artificial Intelligence, Lübeck, Germany

   Timo Kepp & Heinz Handels

2. Institute of Medical Informatics, University of Lübeck, Lübeck, Germany

   Timo Kepp, Julia Andresen & Heinz Handels

3. Department of Ophthalmology, University of Kiel, Kiel, Germany

   Claus von der Burchard & Johann Roider

4. Institute of Biomedical Optics, University of Lübeck, Lübeck, Germany

   Gereon Hüttmann

Corresponding author

Correspondence to Timo Kepp .

Editors and Affiliations

1. Peter L. Reichertz Institut für Medizinische, Informatik der TU Braunschweig und der Medizinischen Hochschule Hannover, Braunschweig, Niedersachsen, Deutschland

   Thomas M. Deserno

2. Institut für Medizinische Informatik, Universität zu Lübeck, Lübeck, Schleswig-Holstein, Deutschland

   Heinz Handels

3. Lehrstuhl für Mustererkennung, Friedrich-Alexander-Universität, Erlangen, Bayern, Deutschland

   Andreas Maier

4. Medical Image Computing, E230, Deutsches Krebsforschungszentrum (DKFZ), Heidelberg, Baden-Württemberg, Deutschland

   Klaus Maier-Hein

5. Fakultät für Informatik und Mathematik, Ostbayerische Technische Hochschule Regensburg, Regensburg, Deutschland

   Christoph Palm

6. Institut für Medizinische Informatik, Charité – Universitätsmedizin Berlin, Berlin, Berlin, Deutschland

   Thomas Tolxdorff

Reprints and permissions

© 2023 Der/die Autor(en), exklusiv lizenziert an Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature

Policies and ethics