Dense Segmentation in Selected Dimensions: Application to Retinal Optical Coherence Tomography

Bart Liefers, Cristina González-Gonzalo, Caroline Klaver, Bram Van Ginneken, Clara I. Sánchez

Research output: Contribution to journalConference articleAcademicpeer-review

8 Citations (Scopus)


We present a novel convolutional neural network architecture designed for dense segmentation in a subset of the dimensions of the input data. The architecture takes an N-dimensional image as input, and produces a label for every pixel in M output dimensions, where 0 < M < N. Large context is incorporated by an encoder-decoder structure, while funneling shortcut subnetworks provide precise localization. We demonstrate applicability of the architecture on two problems in retinal optical coherence tomography: segmentation of geographic atrophy and segmentation of retinal layers. Performance is compared against two baseline methods, that leave out either the encoderdecoder structure or the shortcut subnetworks. For segmentation of geographic atrophy, an average Dice score of 0:49±0:21 was obtained, compared to 0:46±0:22 and 0:28±0:19 for the baseline methods, respectively. For the layer-segmentation task, the proposed architecture achieved a mean absolute error of 1:305±0:547 pixels compared to 1:967±0:841 and 2:166±0:886 for the baseline methods.

Original languageEnglish
Pages (from-to)337-346
Number of pages10
JournalProceedings of Machine Learning Research
Publication statusPublished - 2019
Event2nd International Conference on Medical Imaging with Deep Learning, MIDL 2019 - London, United Kingdom
Duration: 8 Jul 201910 Jul 2019

Bibliographical note

Publisher Copyright:
© 2019 B. Liefers, C. González-Gonzalo, C. Klaver, B. van Ginneken & C. Sánchez.


Dive into the research topics of 'Dense Segmentation in Selected Dimensions: Application to Retinal Optical Coherence Tomography'. Together they form a unique fingerprint.

Cite this