Probabilistic Intra-Retinal Layer Segmentation in 3-D OCT Images Using Global Shape Regularization
This work addresses the need for efficient and robust segmentation in medical imaging for ophthalmology, offering an out-of-the-box solution that can handle both normal and pathological scans without preprocessing.
The paper tackles the problem of fast and accurate segmentation of 3-D OCT retinal images by introducing a probabilistic approach that models appearance and global shape variations, achieving average unsigned errors of 2.46 μm for 3-D volumes and up to 4.09 μm for 2-D scans, with segmentation taking around a minute per volume.
With the introduction of spectral-domain optical coherence tomography (OCT), resulting in a significant increase in acquisition speed, the fast and accurate segmentation of 3-D OCT scans has become evermore important. This paper presents a novel probabilistic approach, that models the appearance of retinal layers as well as the global shape variations of layer boundaries. Given an OCT scan, the full posterior distribution over segmentations is approximately inferred using a variational method enabling efficient probabilistic inference in terms of computationally tractable model components: Segmenting a full 3-D volume takes around a minute. Accurate segmentations demonstrate the benefit of using global shape regularization: We segmented 35 fovea-centered 3-D volumes with an average unsigned error of 2.46 $\pm$ 0.22 μm as well as 80 normal and 66 glaucomatous 2-D circular scans with errors of 2.92 $\pm$ 0.53 μm and 4.09 $\pm$ 0.98 μm respectively. Furthermore, we utilized the inferred posterior distribution to rate the quality of the segmentation, point out potentially erroneous regions and discriminate normal from pathological scans. No pre- or postprocessing was required and we used the same set of parameters for all data sets, underlining the robustness and out-of-the-box nature of our approach.