Optical coherence tomography (OCT) enables the non-invasive acquisition of high-resolution three-dimensional cross-sectional images at micrometer scale and is mainly used in the field of ophthalmology for diagnosis as well as monitoring of eye diseases. Also in other areas, such as dermatology, OCT is already well established. Due to its non-invasive nature, OCT is also employed for research studies involving animal models. Manual evaluation of OCT images of animal models is a challenging task due to the lack of imaging standards and the varying anatomy between models. In this paper, we present a deep learning algorithm for the automatic segmentation of several layers of mouse skin in OCT image data using a deep convolutional neural network (CNN). The architecture of our CNN is based on the U-net and is modified by densely-connected convolutions. We compared our adapted CNN with our previous algorithm, a combination of a random forest classification and a graph-based refinement, and a baseline U-net. The results showed that, on average, our proposed CNN outperformed our previous algorithm and the baseline U-net. In addition, a reduction of outliers could be observed through the use of densely-connected convolutions.

Purpose: The registration of medical images often suffers from missing correspondences due to inter-patient variations, pathologies and their progression leading to implausible deformations that cause misregistrations and might eliminate valuable information. Detecting non-corresponding regions simultaneously with the registration process helps generating better deformations and has been investigated thoroughly with classical iterative frameworks but rarely with deep learning-based methods. Methods: We present the joint non-correspondence segmentation and image registration network (NCR-Net), a convolutional neural network (CNN) trained on a Mumford–Shah-like functional, transferring the classical approach to the field of deep learning. NCR-Net consists of one encoding and two decoding parts allowing the network to simultaneously generate diffeomorphic deformations and segment non-correspondences. The loss function is composed of a masked image distance measure and regularization of deformation field and segmentation output. Additionally, anatomical labels are used for weak supervision of the registration task. No manual segmentations of non-correspondences are required. Results: The proposed network is evaluated on the publicly available LPBA40 dataset with artificially added stroke lesions and a longitudinal optical coherence tomography (OCT) dataset of patients with age-related macular degeneration. The LPBA40 data are used to quantitatively assess the segmentation performance of the network, and it is shown qualitatively that NCR-Net can be used for the unsupervised segmentation of pathologies in OCT images. Furthermore, NCR-Net is compared to a registration-only network and state-of-the-art registration algorithms showing that NCR-Net achieves competitive performance and superior robustness to non-correspondences. Conclusion: NCR-Net, a CNN for simultaneous image registration and unsupervised non-correspondence segmentation, is presented. Experimental results show the network’s ability to segment non-correspondence regions in an unsupervised manner and its robust registration performance even in the presence of large pathologies.

Optical coherence tomography (OCT) is an extensively used imaging tool for disease monitoring in both age-related macular degeneration (AMD) and retinal vein occlusion (RVO). However, there is limited literature on minimum requirements of OCT settings for reliable biomarker detection. This study systematically investigates both the influence of scan size and interscan distance (ISD) on disease activity detection. We analyzed 80 OCT volumes of AMD patients and 12 OCT volumes of RVO patients for the presence of subretinal fluid (SRF), intraretinal fluid (IRF), and pigment epithelium detachment (PED). All volume scans had a scan size of 6 × 6 mm and an ISD of 125 µm. We analyzed both general fluid distribution and how biomarker detection sensitivity decreases when reducing scan size or density. We found that in AMD patients, all fluids were nearly normally distributed, with most occurrences in the foveal center and concentric decrease towards the periphery. When reducing the scan size to 3 × 3 and 2 × 2 mm, disease activity detection was still high (0.98 and 0.96). Increasing ISD only slightly can already compromise biomarker detection sensitivity (0.9 for 250 µm ISD against 125 µm ISD).

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].

The evaluation of lymph node metastases plays a crucial role in achieving precise cancer staging, which in turn influences subsequent decisions regarding treatment options. The detection of lymph nodes poses challenges due to the presence of unclear boundaries and the diverse range of sizes and morphological characteristics, making it a resource-intensive process. As part of the LNQ 2023 MICCAI challenge, we propose the use of anatomical priors as a tool to address the challenges that persist in automatic mediastinal lymph node segmentation in combination with the partial annotation of the challenge training data. The model ensemble using all suggested modifications yields a Dice score of 0.6033 and segments 57% of the ground truth lymph nodes, compared to 27% when training on CT only. Segmentation accuracy is improved significantly by incorporating a probabilistic lymph node atlas in loss weighting and post-processing. The largest performance gains are achieved by oversampling fully annotated data to account for the partial annotation of the challenge training data, as well as adding additional data augmentation to address the high heterogeneity of the CT images and lymph node appearance. Our code is available at https://github.com/MICAI-IMI-UzL/LNQ2023.