Ursula Schmidt-Erfurth

Robbie Holland, Oliver Leingang, Hrvoje Bogunović et al.

Deep learning has potential to automate screening, monitoring and grading of disease in medical images. Pretraining with contrastive learning enables models to extract robust and generalisable features from natural image datasets, facilitating label-efficient downstream image analysis. However, the direct application of conventional contrastive methods to medical datasets introduces two domain-specific issues. Firstly, several image transformations which have been shown to be crucial for effective contrastive learning do not translate from the natural image to the medical image domain. Secondly, the assumption made by conventional methods, that any two images are dissimilar, is systematically misleading in medical datasets depicting the same anatomy and disease. This is exacerbated in longitudinal image datasets that repeatedly image the same patient cohort to monitor their disease progression over time. In this paper we tackle these issues by extending conventional contrastive frameworks with a novel metadata-enhanced strategy. Our approach employs widely available patient metadata to approximate the true set of inter-image contrastive relationships. To this end we employ records for patient identity, eye position (i.e. left or right) and time series information. In experiments using two large longitudinal datasets containing 170,427 retinal OCT images of 7,912 patients with age-related macular degeneration (AMD), we evaluate the utility of using metadata to incorporate the temporal dynamics of disease progression into pretraining. Our metadata-enhanced approach outperforms both standard contrastive methods and a retinal image foundation model in five out of six image-level downstream tasks related to AMD. Due to its modularity, our method can be quickly and cost-effectively tested to establish the potential benefits of including available metadata in contrastive pretraining.

Taha Emre, Marzieh Oghbaie, Arunava Chakravarty et al.

In the field of medical imaging, 3D deep learning models play a crucial role in building powerful predictive models of disease progression. However, the size of these models presents significant challenges, both in terms of computational resources and data requirements. Moreover, achieving high-quality pretraining of 3D models proves to be even more challenging. To address these issues, hybrid 2.5D approaches provide an effective solution for utilizing 3D volumetric data efficiently using 2D models. Combining 2D and 3D techniques offers a promising avenue for optimizing performance while minimizing memory requirements. In this paper, we explore 2.5D architectures based on a combination of convolutional neural networks (CNNs), long short-term memory (LSTM), and Transformers. In addition, leveraging the benefits of recent non-contrastive pretraining approaches in 2D, we enhanced the performance and data efficiency of 2.5D techniques even further. We demonstrate the effectiveness of architectures and associated pretraining on a task of predicting progression to wet age-related macular degeneration (AMD) within a six-month period on two large longitudinal OCT datasets.

Robbie Holland, Oliver Leingang, Christopher Holmes et al.

Age-related macular degeneration (AMD) is the leading cause of blindness in the elderly. Current grading systems based on imaging biomarkers only coarsely group disease stages into broad categories and are unable to predict future disease progression. It is widely believed that this is due to their focus on a single point in time, disregarding the dynamic nature of the disease. In this work, we present the first method to automatically discover biomarkers that capture temporal dynamics of disease progression. Our method represents patient time series as trajectories in a latent feature space built with contrastive learning. Then, individual trajectories are partitioned into atomic sub-sequences that encode transitions between disease states. These are clustered using a newly introduced distance metric. In quantitative experiments we found our method yields temporal biomarkers that are predictive of conversion to late AMD. Furthermore, these clusters were highly interpretable to ophthalmologists who confirmed that many of the clusters represent dynamics that have previously been linked to the progression of AMD, even though they are currently not included in any clinical grading system.

Arunava Chakravarty, Taha Emre, Oliver Leingang et al.

The lack of reliable biomarkers makes predicting the conversion from intermediate to neovascular age-related macular degeneration (iAMD, nAMD) a challenging task. We develop a Deep Learning (DL) model to predict the future risk of conversion of an eye from iAMD to nAMD from its current OCT scan. Although eye clinics generate vast amounts of longitudinal OCT scans to monitor AMD progression, only a small subset can be manually labeled for supervised DL. To address this issue, we propose Morph-SSL, a novel Self-supervised Learning (SSL) method for longitudinal data. It uses pairs of unlabelled OCT scans from different visits and involves morphing the scan from the previous visit to the next. The Decoder predicts the transformation for morphing and ensures a smooth feature manifold that can generate intermediate scans between visits through linear interpolation. Next, the Morph-SSL trained features are input to a Classifier which is trained in a supervised manner to model the cumulative probability distribution of the time to conversion with a sigmoidal function. Morph-SSL was trained on unlabelled scans of 399 eyes (3570 visits). The Classifier was evaluated with a five-fold cross-validation on 2418 scans from 343 eyes with clinical labels of the conversion date. The Morph-SSL features achieved an AUC of 0.766 in predicting the conversion to nAMD within the next 6 months, outperforming the same network when trained end-to-end from scratch or pre-trained with popular SSL methods. Automated prediction of the future risk of nAMD onset can enable timely treatment and individualized AMD management.

Dmitrii Lachinov, Arunava Chakravarty, Christoph Grechenig et al.

Robust forecasting of the future anatomical changes inflicted by an ongoing disease is an extremely challenging task that is out of grasp even for experienced healthcare professionals. Such a capability, however, is of great importance since it can improve patient management by providing information on the speed of disease progression already at the admission stage, or it can enrich the clinical trials with fast progressors and avoid the need for control arms by the means of digital twins. In this work, we develop a deep learning method that models the evolution of age-related disease by processing a single medical scan and providing a segmentation of the target anatomy at a requested future point in time. Our method represents a time-invariant physical process and solves a large-scale problem of modeling temporal pixel-level changes utilizing NeuralODEs. In addition, we demonstrate the approaches to incorporate the prior domain-specific constraints into our method and define temporal Dice loss for learning temporal objectives. To evaluate the applicability of our approach across different age-related diseases and imaging modalities, we developed and tested the proposed method on the datasets with 967 retinal OCT volumes of 100 patients with Geographic Atrophy, and 2823 brain MRI volumes of 633 patients with Alzheimer's Disease. For Geographic Atrophy, the proposed method outperformed the related baseline models in the atrophy growth prediction. For Alzheimer's Disease, the proposed method demonstrated remarkable performance in predicting the brain ventricle changes induced by the disease, achieving the state-of-the-art result on TADPOLE challenge.

Botond Fazekas, Dmitrii Lachinov, Guilherme Aresta et al.

Bruch's membrane (BM) segmentation on optical coherence tomography (OCT) is a pivotal step for the diagnosis and follow-up of age-related macular degeneration (AMD), one of the leading causes of blindness in the developed world. Automated BM segmentation methods exist, but they usually do not account for the anatomical coherence of the results, neither provide feedback on the confidence of the prediction. These factors limit the applicability of these systems in real-world scenarios. With this in mind, we propose an end-to-end deep learning method for automated BM segmentation in AMD patients. An Attention U-Net is trained to output a probability density function of the BM position, while taking into account the natural curvature of the surface. Besides the surface position, the method also estimates an A-scan wise uncertainty measure of the segmentation output. Subsequently, the A-scans with high uncertainty are interpolated using thin plate splines (TPS). We tested our method with ablation studies on an internal dataset with 138 patients covering all three AMD stages, and achieved a mean absolute localization error of 4.10 um. In addition, the proposed segmentation method was compared against the state-of-the-art methods and showed a superior performance on an external publicly available dataset from a different patient cohort and OCT device, demonstrating strong generalization ability.

Botond Fazekas, Guilherme Aresta, Dmitrii Lachinov et al.

Optical coherence tomography (OCT) is a non-invasive 3D modality widely used in ophthalmology for imaging the retina. Achieving automated, anatomically coherent retinal layer segmentation on OCT is important for the detection and monitoring of different retinal diseases, like Age-related Macular Disease (AMD) or Diabetic Retinopathy. However, the majority of state-of-the-art layer segmentation methods are based on purely supervised deep-learning, requiring a large amount of pixel-level annotated data that is expensive and hard to obtain. With this in mind, we introduce a semi-supervised paradigm into the retinal layer segmentation task that makes use of the information present in large-scale unlabeled datasets as well as anatomical priors. In particular, a novel fully differentiable approach is used for converting surface position regression into a pixel-wise structured segmentation, allowing to use both 1D surface and 2D layer representations in a coupled fashion to train the model. In particular, these 2D segmentations are used as anatomical factors that, together with learned style factors, compose disentangled representations used for reconstructing the input image. In parallel, we propose a set of anatomical priors to improve network training when a limited amount of labeled data is available. We demonstrate on the real-world dataset of scans with intermediate and wet-AMD that our method outperforms state-of-the-art when using our full training set, but more importantly largely exceeds state-of-the-art when it is trained with a fraction of the labeled data.

Thomas Schlegl, Heiko Stino, Michael Niederleithner et al.

The automatic detection and localization of anatomical features in retinal imaging data are relevant for many aspects. In this work, we follow a data-centric approach to optimize classifier training for optic nerve head detection and localization in optical coherence tomography en face images of the retina. We examine the effect of domain knowledge driven spatial complexity reduction on the resulting optic nerve head segmentation and localization performance. We present a machine learning approach for segmenting optic nerve head in 2D en face projections of 3D widefield swept source optical coherence tomography scans that enables the automated assessment of large amounts of data. Evaluation on manually annotated 2D en face images of the retina demonstrates that training of a standard U-Net can yield improved optic nerve head segmentation and localization performance when the underlying pixel-level binary classification task is spatially relaxed through domain knowledge.

Robbie Holland, Thomas R. P. Taylor, Christopher Holmes et al.

Clinicians spend a significant amount of time reviewing medical images and transcribing their findings regarding patient diagnosis, referral and treatment in text form. Vision-language models (VLMs), which automatically interpret images and summarize their findings as text, have enormous potential to alleviate clinical workloads and increase patient access to high-quality medical care. While foundational models have stirred considerable interest in the medical community, it is unclear whether their general capabilities translate to real-world clinical utility. In this work, we demonstrate that OpenAI's ChatGPT-4o model, in addition to two foundation VLMs designed for medical use, markedly underperform compared to practicing ophthalmologists on specialist tasks crucial to the care of patients with age-related macular degeneration (AMD). To address this, we initially identified the essential capabilities required for image-based clinical decision-making, and then developed a curriculum to selectively train VLMs in these skills. The resulting model, RetinaVLM, can be instructed to write reports that significantly outperform those written by leading foundation medical VLMs and ChatGPT-4o in disease staging (F1 score of 0.63 vs. 0.33) and patient referral (0.67 vs. 0.50), and approaches the diagnostic performance of junior ophthalmologists (who achieve 0.77 and 0.78 on the respective tasks). Furthermore, in a single-blind reader study two senior ophthalmologists with up to 32 years of experience found RetinaVLM's reports were found to be substantially more accurate than those by ChatGPT-4o (64.3% vs. 14.3%). These results reinforce that our curriculum-based approach provides a blueprint towards specializing foundation medical VLMs for real-world clinical tasks.

Marzieh Oghbaie, Teresa Araujo, Taha Emre et al.

The automatic classification of 3D medical data is memory-intensive. Also, variations in the number of slices between samples is common. Naïve solutions such as subsampling can solve these problems, but at the cost of potentially eliminating relevant diagnosis information. Transformers have shown promising performance for sequential data analysis. However, their application for long sequences is data, computationally, and memory demanding. In this paper, we propose an end-to-end Transformer-based framework that allows to classify volumetric data of variable length in an efficient fashion. Particularly, by randomizing the input volume-wise resolution(#slices) during training, we enhance the capacity of the learnable positional embedding assigned to each volume slice. Consequently, the accumulated positional information in each positional embedding can be generalized to the neighbouring slices, even for high-resolution volumes at the test time. By doing so, the model will be more robust to variable volume length and amenable to different computational budgets. We evaluated the proposed approach in retinal OCT volume classification and achieved 21.96% average improvement in balanced accuracy on a 9-class diagnostic task, compared to state-of-the-art video transformers. Our findings show that varying the volume-wise resolution of the input during training results in more informative volume representation as compared to training with fixed number of slices per volume.

Arunava Chakravarty, Taha Emre, Dmitrii Lachinov et al.

Predicting future disease progression risk from medical images is challenging due to patient heterogeneity, and subtle or unknown imaging biomarkers. Moreover, deep learning (DL) methods for survival analysis are susceptible to image domain shifts across scanners. We tackle these issues in the task of predicting late dry Age-related Macular Degeneration (dAMD) onset from retinal OCT scans. We propose a novel DL method for survival prediction to jointly predict from the current scan a risk score, inversely related to time-to-conversion, and the probability of conversion within a time interval $t$. It uses a family of parallel hyperplanes generated by parameterizing the bias term as a function of $t$. In addition, we develop unsupervised losses based on intra-subject image pairs to ensure that risk scores increase over time and that future conversion predictions are consistent with AMD stage prediction using actual scans of future visits. Such losses enable data-efficient fine-tuning of the trained model on new unlabeled datasets acquired with a different scanner. Extensive evaluation on two large datasets acquired with different scanners resulted in a mean AUROCs of 0.82 for Dataset-1 and 0.83 for Dataset-2, across prediction intervals of 6,12 and 24 months.

Taha Emre, Arunava Chakravarty, Thomas Pinetz et al.

Temporally aware image representations are crucial for capturing disease progression in 3D volumes of longitudinal medical datasets. However, recent state-of-the-art self-supervised learning approaches like Masked Autoencoding (MAE), despite their strong representation learning capabilities, lack temporal awareness. In this paper, we propose STAMP (Stochastic Temporal Autoencoder with Masked Pretraining), a Siamese MAE framework that encodes temporal information through a stochastic process by conditioning on the time difference between the 2 input volumes. Unlike deterministic Siamese approaches, which compare scans from different time points but fail to account for the inherent uncertainty in disease evolution, STAMP learns temporal dynamics stochastically by reframing the MAE reconstruction loss as a conditional variational inference objective. We evaluated STAMP on two OCT and one MRI datasets with multiple visits per patient. STAMP pretrained ViT models outperformed both existing temporal MAE methods and foundation models on different late stage Age-Related Macular Degeneration and Alzheimer's Disease progression prediction which require models to learn the underlying non-deterministic temporal dynamics of the diseases.

Taha Emre, Arunava Chakravarty, Antoine Rivail et al.

Recent contrastive learning methods achieved state-of-the-art in low label regimes. However, the training requires large batch sizes and heavy augmentations to create multiple views of an image. With non-contrastive methods, the negatives are implicitly incorporated in the loss, allowing different images and modalities as pairs. Although the meta-information (i.e., age, sex) in medical imaging is abundant, the annotations are noisy and prone to class imbalance. In this work, we exploited already existing temporal information (different visits from a patient) in a longitudinal optical coherence tomography (OCT) dataset using temporally informed non-contrastive loss (TINC) without increasing complexity and need for negative pairs. Moreover, our novel pair-forming scheme can avoid heavy augmentations and implicitly incorporates the temporal information in the pairs. Finally, these representations learned from the pretraining are more successful in predicting disease progression where the temporal information is crucial for the downstream task. More specifically, our model outperforms existing models in predicting the risk of conversion within a time frame from intermediate age-related macular degeneration (AMD) to the late wet-AMD stage.

José Morano, Guilherme Aresta, Dmitrii Lachinov et al.

Deep learning has become a valuable tool for the automation of certain medical image segmentation tasks, significantly relieving the workload of medical specialists. Some of these tasks require segmentation to be performed on a subset of the input dimensions, the most common case being 3D-to-2D. However, the performance of existing methods is strongly conditioned by the amount of labeled data available, as there is currently no data efficient method, e.g. transfer learning, that has been validated on these tasks. In this work, we propose a novel convolutional neural network (CNN) and self-supervised learning (SSL) method for label-efficient 3D-to-2D segmentation. The CNN is composed of a 3D encoder and a 2D decoder connected by novel 3D-to-2D blocks. The SSL method consists of reconstructing image pairs of modalities with different dimensionality. The approach has been validated in two tasks with clinical relevance: the en-face segmentation of geographic atrophy and reticular pseudodrusen in optical coherence tomography. Results on different datasets demonstrate that the proposed CNN significantly improves the state of the art in scenarios with limited labeled data by up to 8% in Dice score. Moreover, the proposed SSL method allows further improvement of this performance by up to 23%, and we show that the SSL is beneficial regardless of the network architecture.

José Morano, Botond Fazekas, Emese Sükei et al.

Artificial intelligence (AI) has become a fundamental tool for assisting clinicians in analyzing ophthalmic images, such as optical coherence tomography (OCT). However, developing AI models often requires extensive annotation, and existing models tend to underperform on independent, unseen data. Foundation models (FMs), large AI models trained on vast unlabeled datasets, have shown promise in overcoming these challenges. Nonetheless, available FMs for ophthalmology lack extensive validation, especially for segmentation tasks, and focus on a single imaging modality. In this context, we propose MIRAGE, a novel multimodal FM for the analysis of OCT and scanning laser ophthalmoscopy (SLO) images. Additionally, we propose a new evaluation benchmark with OCT/SLO classification and segmentation tasks. The comparison with general and specialized FMs and segmentation methods shows the superiority of MIRAGE in both types of tasks, highlighting its suitability as a basis for the development of robust AI systems for retinal OCT image analysis. Both MIRAGE and the evaluation benchmark are publicly available: https://github.com/j-morano/MIRAGE.

Ronald Fecso, José Morano, Ursula Schmidt-Erfurth et al.

The rise of imaging techniques such as optical coherence tomography (OCT) and advances in deep learning (DL) have enabled clinicians and researchers to streamline retinal disease staging. A popular DL approach is self-supervised learning (SSL), where models learn from vast amounts of unlabeled data, avoiding costly annotation. SSL has allowed the development of foundation models (FMs), large models that can be used for a variety of downstream tasks. However, existing FMs for OCT, trained solely on image data, lack a comprehensive and robust semantic understanding of images, as evidenced by their downstream performance (especially for complex tasks), and thus require supervised fine-tuning (which may be unfeasible) to better adapt to specific applications and populations. To address this, we propose RetFiner, an SSL vision-language refinement scheme that improves the representations of existing FMs and enables their efficient and direct adaptation to specific populations for improved downstream performance. Our method uses a diverse set of training objectives which take advantage of the rich supervisory signal found in textual data. We tested RetFiner on the retinal FMs RETFound, UrFound, and VisionFM, showing significant improvements in linear probing performance on seven highly diverse OCT classification tasks, with an average increase of 5.8, 3.9, and 2.1 percentage points over their baselines, respectively. Our code and model weights are publicly available at https://github.com/ronnief1/RetFiner.

Robbie Holland, Rebecca Kaye, Ahmed M. Hagag et al.

Diseases are currently managed by grading systems, where patients are stratified by grading systems into stages that indicate patient risk and guide clinical management. However, these broad categories typically lack prognostic value, and proposals for new biomarkers are currently limited to anecdotal observations. In this work, we introduce a deep-learning-based biomarker proposal system for the purpose of accelerating biomarker discovery in age-related macular degeneration (AMD). It works by first training a neural network using self-supervised contrastive learning to discover, without any clinical annotations, features relating to both known and unknown AMD biomarkers present in 46,496 retinal optical coherence tomography (OCT) images. To interpret the discovered biomarkers, we partition the images into 30 subsets, termed clusters, that contain similar features. We then conduct two parallel 1.5-hour semi-structured interviews with two independent teams of retinal specialists that describe each cluster in clinical language. Overall, both teams independently identified clearly distinct characteristics in 27 of 30 clusters, of which 23 were related to AMD. Seven were recognised as known biomarkers already used in established grading systems and 16 depicted biomarker combinations or subtypes that are either not yet used in grading systems, were only recently proposed, or were unknown. Clusters separated incomplete from complete retinal atrophy, intraretinal from subretinal fluid and thick from thin choroids, and in simulation outperformed clinically-used grading systems in prognostic value. Overall, contrastive learning enabled the automatic proposal of AMD biomarkers that go beyond the set used by clinically established grading systems. Ultimately, we envision that equipping clinicians with discovery-oriented deep-learning tools can accelerate discovery of novel prognostic biomarkers.

José Morano, Guilherme Aresta, Christoph Grechenig et al.

The use of multimodal imaging has led to significant improvements in the diagnosis and treatment of many diseases. Similar to clinical practice, some works have demonstrated the benefits of multimodal fusion for automatic segmentation and classification using deep learning-based methods. However, current segmentation methods are limited to fusion of modalities with the same dimensionality (e.g., 3D+3D, 2D+2D), which is not always possible, and the fusion strategies implemented by classification methods are incompatible with localization tasks. In this work, we propose a novel deep learning-based framework for the fusion of multimodal data with heterogeneous dimensionality (e.g., 3D+2D) that is compatible with localization tasks. The proposed framework extracts the features of the different modalities and projects them into the common feature subspace. The projected features are then fused and further processed to obtain the final prediction. The framework was validated on the following tasks: segmentation of geographic atrophy (GA), a late-stage manifestation of age-related macular degeneration, and segmentation of retinal blood vessels (RBV) in multimodal retinal imaging. Our results show that the proposed method outperforms the state-of-the-art monomodal methods on GA and RBV segmentation by up to 3.10% and 4.64% Dice, respectively.

Taha Emre, Arunava Chakravarty, Antoine Rivail et al.

Self-supervised learning (SSL) has emerged as a powerful technique for improving the efficiency and effectiveness of deep learning models. Contrastive methods are a prominent family of SSL that extract similar representations of two augmented views of an image while pushing away others in the representation space as negatives. However, the state-of-the-art contrastive methods require large batch sizes and augmentations designed for natural images that are impractical for 3D medical images. To address these limitations, we propose a new longitudinal SSL method, 3DTINC, based on non-contrastive learning. It is designed to learn perturbation-invariant features for 3D optical coherence tomography (OCT) volumes, using augmentations specifically designed for OCT. We introduce a new non-contrastive similarity loss term that learns temporal information implicitly from intra-patient scans acquired at different times. Our experiments show that this temporal information is crucial for predicting progression of retinal diseases, such as age-related macular degeneration (AMD). After pretraining with 3DTINC, we evaluated the learned representations and the prognostic models on two large-scale longitudinal datasets of retinal OCTs where we predict the conversion to wet-AMD within a six months interval. Our results demonstrate that each component of our contributions is crucial for learning meaningful representations useful in predicting disease progression from longitudinal volumetric scans.

Chengzhi Shen, Martin J. Menten, Hrvoje Bogunović et al.

Analyzing temporal developments is crucial for the accurate prognosis of many medical conditions. Temporal changes that occur over short time scales are key to assessing the health of physiological functions, such as the cardiac cycle. Moreover, tracking longer term developments that occur over months or years in evolving processes, such as age-related macular degeneration (AMD), is essential for accurate prognosis. Despite the importance of both short and long term analysis to clinical decision making, they remain understudied in medical deep learning. State of the art methods for spatiotemporal representation learning, developed for short natural videos, prioritize the detection of temporal constants rather than temporal developments. Moreover, they do not account for varying time intervals between acquisitions, which are essential for contextualizing observed changes. To address these issues, we propose two approaches. First, we combine clip-level contrastive learning with a novel temporal embedding to adapt to irregular time series. Second, we propose masking and predicting latent frame representations of the temporal sequence. Our two approaches outperform all prior methods on temporally-dependent tasks including cardiac output estimation and three prognostic AMD tasks. Overall, this enables the automated analysis of temporal patterns which are typically overlooked in applications of deep learning to medicine.

Botond Fazekas, Guilherme Aresta, Philipp Seeböck et al.

Optical coherence tomography (OCT) is widely used for diagnosing and monitoring retinal diseases, such as age-related macular degeneration (AMD). The segmentation of biomarkers such as layers and lesions is essential for patient diagnosis and follow-up. Recently, semi-supervised learning has shown promise in improving retinal segmentation performance. However, existing methods often produce anatomically implausible segmentations, fail to effectively model layer-lesion interactions, and lack guarantees on topological correctness. To address these limitations, we propose a novel semi-supervised model that introduces a fully differentiable biomarker topology engine to enforce anatomically correct segmentation of lesions and layers. This enables joint learning with bidirectional influence between layers and lesions, leveraging unlabeled and diverse partially labeled datasets. Our model learns a disentangled representation, separating spatial and style factors. This approach enables more realistic layer segmentations and improves lesion segmentation, while strictly enforcing lesion location in their anatomically plausible positions relative to the segmented layers. We evaluate the proposed model on public and internal datasets of OCT scans and show that it outperforms the current state-of-the-art in both lesion and layer segmentation, while demonstrating the ability to generalize layer segmentation to pathological cases using partially annotated training data. Our results demonstrate the potential of using anatomical constraints in semi-supervised learning for accurate, robust, and trustworthy retinal biomarker segmentation.

Taha Emre, Arunava Chakravarty, Dmitrii Lachinov et al.

Contrastive pretraining provides robust representations by ensuring their invariance to different image transformations while simultaneously preventing representational collapse. Equivariant contrastive learning, on the other hand, provides representations sensitive to specific image transformations while remaining invariant to others. By introducing equivariance to time-induced transformations, such as disease-related anatomical changes in longitudinal imaging, the model can effectively capture such changes in the representation space. In this work, we propose a Time-equivariant Contrastive Learning (TC) method. First, an encoder embeds two unlabeled scans from different time points of the same patient into the representation space. Next, a temporal equivariance module is trained to predict the representation of a later visit based on the representation from one of the previous visits and the corresponding time interval with a novel regularization loss term while preserving the invariance property to irrelevant image transformations. On a large longitudinal dataset, our model clearly outperforms existing equivariant contrastive methods in predicting progression from intermediate age-related macular degeneration (AMD) to advanced wet-AMD within a specified time-window.

Anna Breger, Felix Goldbach, Bianca S. Gerendas et al.

Optical coherence tomography angiography (OCTA) is a novel noninvasive imaging modality for visualization of retinal blood flow in the human retina. Using specific OCTA imaging biomarkers for the identification of pathologies, automated image segmentations of the blood vessels can improve subsequent analysis and diagnosis. We present a novel segmentation method for vessel density identification based on frequency representations of the image, in particular, using so-called Gabor filter banks. The algorithm is evaluated qualitatively and quantitatively on an OCTA image in-house data set from $10$ eyes acquired by a Cirrus HD-OCT device. Qualitatively, the segmentation outcomes received very good visual evaluation feedback by experts. Quantitatively, we compared the resulting vessel density values with automated in-built values provided by the device. The results underline the visual evaluation. For the evaluation of the FAZ identification substep, manual annotations of $2$ expert graders were used, showing that our results coincide well in visual and quantitative manners. Lastly, we suggest the computation of adaptive local vessel density maps that allow straightforward analysis of retinal blood flow in a local manner.

Dmitrii Lachinov, Philipp Seeboeck, Julia Mai et al.

In medical imaging, there are clinically relevant segmentation tasks where the output mask is a projection to a subset of input image dimensions. In this work, we propose a novel convolutional neural network architecture that can effectively learn to produce a lower-dimensional segmentation mask than the input image. The network restores encoded representation only in a subset of input spatial dimensions and keeps the representation unchanged in the others. The newly proposed projective skip-connections allow linking the encoder and decoder in a UNet-like structure. We evaluated the proposed method on two clinically relevant tasks in retinal Optical Coherence Tomography (OCT): geographic atrophy and retinal blood vessel segmentation. The proposed method outperformed the current state-of-the-art approaches on all the OCT datasets used, consisting of 3D volumes and corresponding 2D en-face masks. The proposed architecture fills the methodological gap between image classification and ND image segmentation.

Rhona Asgari, Sebastian Waldstein, Ferdinand Schlanitz et al.

The presence of drusen is the main hallmark of early/intermediate age-related macular degeneration (AMD). Therefore, automated drusen segmentation is an important step in image-guided management of AMD. There are two common approaches to drusen segmentation. In the first, the drusen are segmented directly as a binary classification task. In the second approach, the surrounding retinal layers (outer boundary retinal pigment epithelium (OBRPE) and Bruch's membrane (BM)) are segmented and the remaining space between these two layers is extracted as drusen. In this work, we extend the standard U-Net architecture with spatial pyramid pooling components to introduce global feature context. We apply the model to the task of segmenting drusen together with BM and OBRPE. The proposed network was trained and evaluated on a longitudinal OCT dataset of 425 scans from 38 patients with early/intermediate AMD. This preliminary study showed that the proposed network consistently outperformed the standard U-net model.

Antoine Rivail, Ursula Schmidt-Erfurth, Wolf-Dieter Vogl et al.

Longitudinal imaging is capable of capturing the static ana\-to\-mi\-cal structures and the dynamic changes of the morphology resulting from aging or disease progression. Self-supervised learning allows to learn new representation from available large unlabelled data without any expert knowledge. We propose a deep learning self-supervised approach to model disease progression from longitudinal retinal optical coherence tomography (OCT). Our self-supervised model takes benefit from a generic time-related task, by learning to estimate the time interval between pairs of scans acquired from the same patient. This task is (i) easy to implement, (ii) allows to use irregularly sampled data, (iii) is tolerant to poor registration, and (iv) does not rely on additional annotations. This novel method learns a representation that focuses on progression specific information only, which can be transferred to other types of longitudinal problems. We transfer the learnt representation to a clinically highly relevant task of predicting the onset of an advanced stage of age-related macular degeneration within a given time interval based on a single OCT scan. The boost in prediction accuracy, in comparison to a network learned from scratch or transferred from traditional tasks, demonstrates that our pretrained self-supervised representation learns a clinically meaningful information.

José Ignacio Orlando, Anna Breger, Hrvoje Bogunović et al.

Segmenting anatomical structures such as the photoreceptor layer in retinal optical coherence tomography (OCT) scans is challenging in pathological scenarios. Supervised deep learning models trained with standard loss functions are usually able to characterize only the most common disease appeareance from a training set, resulting in suboptimal performance and poor generalization when dealing with unseen lesions. In this paper we propose to overcome this limitation by means of an augmented target loss function framework. We introduce a novel amplified-target loss that explicitly penalizes errors within the central area of the input images, based on the observation that most of the challenging disease appeareance is usually located in this area. We experimentally validated our approach using a data set with OCT scans of patients with macular diseases. We observe increased performance compared to the models that use only the standard losses. Our proposed loss function strongly supports the segmentation model to better distinguish photoreceptors in highly pathological scenarios.

Rhona Asgari, José Ignacio Orlando, Sebastian Waldstein et al.

Automated drusen segmentation in retinal optical coherence tomography (OCT) scans is relevant for understanding age-related macular degeneration (AMD) risk and progression. This task is usually performed by segmenting the top/bottom anatomical interfaces that define drusen, the outer boundary of the retinal pigment epithelium (OBRPE) and the Bruch's membrane (BM), respectively. In this paper we propose a novel multi-decoder architecture that tackles drusen segmentation as a multitask problem. Instead of training a multiclass model for OBRPE/BM segmentation, we use one decoder per target class and an extra one aiming for the area between the layers. We also introduce connections between each class-specific branch and the additional decoder to increase the regularization effect of this surrogate task. We validated our approach on private/public data sets with 166 early/intermediate AMD Spectralis, and 200 AMD and control Bioptigen OCT volumes, respectively. Our method consistently outperformed several baselines in both layer and drusen segmentation evaluations.

Philipp Seeböck, José Ignacio Orlando, Thomas Schlegl et al.

Diagnosis and treatment guidance are aided by detecting relevant biomarkers in medical images. Although supervised deep learning can perform accurate segmentation of pathological areas, it is limited by requiring a-priori definitions of these regions, large-scale annotations, and a representative patient cohort in the training set. In contrast, anomaly detection is not limited to specific definitions of pathologies and allows for training on healthy samples without annotation. Anomalous regions can then serve as candidates for biomarker discovery. Knowledge about normal anatomical structure brings implicit information for detecting anomalies. We propose to take advantage of this property using bayesian deep learning, based on the assumption that epistemic uncertainties will correlate with anatomical deviations from a normal training set. A Bayesian U-Net is trained on a well-defined healthy environment using weak labels of healthy anatomy produced by existing methods. At test time, we capture epistemic uncertainty estimates of our model using Monte Carlo dropout. A novel post-processing technique is then applied to exploit these estimates and transfer their layered appearance to smooth blob-shaped segmentations of the anomalies. We experimentally validated this approach in retinal optical coherence tomography (OCT) images, using weak labels of retinal layers. Our method achieved a Dice index of 0.789 in an independent anomaly test set of age-related macular degeneration (AMD) cases. The resulting segmentations allowed very high accuracy for separating healthy and diseased cases with late wet AMD, dry geographic atrophy (GA), diabetic macular edema (DME) and retinal vein occlusion (RVO). Finally, we qualitatively observed that our approach can also detect other deviations in normal scans such as cut edge artifacts.

Philipp Seeböck, David Romo-Bucheli, Sebastian Waldstein et al.

Optical coherence tomography (OCT) has become the most important imaging modality in ophthalmology. A substantial amount of research has recently been devoted to the development of machine learning (ML) models for the identification and quantification of pathological features in OCT images. Among the several sources of variability the ML models have to deal with, a major factor is the acquisition device, which can limit the ML model's generalizability. In this paper, we propose to reduce the image variability across different OCT devices (Spectralis and Cirrus) by using CycleGAN, an unsupervised unpaired image transformation algorithm. The usefulness of this approach is evaluated in the setting of retinal fluid segmentation, namely intraretinal cystoid fluid (IRC) and subretinal fluid (SRF). First, we train a segmentation model on images acquired with a source OCT device. Then we evaluate the model on (1) source, (2) target and (3) transformed versions of the target OCT images. The presented transformation strategy shows an F1 score of 0.4 (0.51) for IRC (SRF) segmentations. Compared with traditional transformation approaches, this means an F1 score gain of 0.2 (0.12).

José Ignacio Orlando, Philipp Seeböck, Hrvoje Bogunović et al.

In this paper, we introduce a Bayesian deep learning based model for segmenting the photoreceptor layer in pathological OCT scans. Our architecture provides accurate segmentations of the photoreceptor layer and produces pixel-wise epistemic uncertainty maps that highlight potential areas of pathologies or segmentation errors. We empirically evaluated this approach in two sets of pathological OCT scans of patients with age-related macular degeneration, retinal vein oclussion and diabetic macular edema, improving the performance of the baseline U-Net both in terms of the Dice index and the area under the precision/recall curve. We also observed that the uncertainty estimates were inversely correlated with the model performance, underlying its utility for highlighting areas where manual inspection/correction might be needed.

Anna Breger, Jose Ignacio Orlando, Pavol Harar et al.

The use of orthogonal projections on high-dimensional input and target data in learning frameworks is studied. First, we investigate the relations between two standard objectives in dimension reduction, preservation of variance and of pairwise relative distances. Investigations of their asymptotic correlation as well as numerical experiments show that a projection does usually not satisfy both objectives at once. In a standard classification problem we determine projections on the input data that balance the objectives and compare subsequent results. Next, we extend our application of orthogonal projections to deep learning tasks and introduce a general framework of augmented target loss functions. These loss functions integrate additional information via transformations and projections of the target data. In two supervised learning problems, clinical image segmentation and music information classification, the application of our proposed augmented target loss functions increase the accuracy.

Philipp Seeböck, Sebastian M. Waldstein, Sophie Klimscha et al.

The identification and quantification of markers in medical images is critical for diagnosis, prognosis, and disease management. Supervised machine learning enables the detection and exploitation of findings that are known a priori after annotation of training examples by experts. However, supervision does not scale well, due to the amount of necessary training examples, and the limitation of the marker vocabulary to known entities. In this proof-of-concept study, we propose unsupervised identification of anomalies as candidates for markers in retinal Optical Coherence Tomography (OCT) imaging data without a constraint to a priori definitions. We identify and categorize marker candidates occurring frequently in the data, and demonstrate that these markers show predictive value in the task of detecting disease. A careful qualitative analysis of the identified data driven markers reveals how their quantifiable occurrence aligns with our current understanding of disease course, in early- and late age-related macular degeneration (AMD) patients. A multi-scale deep denoising autoencoder is trained on healthy images, and a one-class support vector machine identifies anomalies in new data. Clustering in the anomalies identifies stable categories. Using these markers to classify healthy-, early AMD- and late AMD cases yields an accuracy of 81.40%. In a second binary classification experiment on a publicly available data set (healthy vs. intermediate AMD) the model achieves an area under the ROC curve of 0.944.

Thomas Schlegl, Hrvoje Bogunovic, Sophie Klimscha et al.

The automatic detection of disease related entities in retinal imaging data is relevant for disease- and treatment monitoring. It enables the quantitative assessment of large amounts of data and the corresponding study of disease characteristics. The presence of hyperreflective foci (HRF) is related to disease progression in various retinal diseases. Manual identification of HRF in spectral-domain optical coherence tomography (SD-OCT) scans is error-prone and tedious. We present a fully automated machine learning approach for segmenting HRF in SD-OCT scans. Evaluation on annotated OCT images of the retina demonstrates that a residual U-Net allows to segment HRF with high accuracy. As our dataset comprised data from different retinal diseases including age-related macular degeneration, diabetic macular edema and retinal vein occlusion, the algorithm can safely be applied in all of them though different pathophysiological origins are known.

Thomas Schlegl, Philipp Seeböck, Sebastian M. Waldstein et al.

Obtaining models that capture imaging markers relevant for disease progression and treatment monitoring is challenging. Models are typically based on large amounts of data with annotated examples of known markers aiming at automating detection. High annotation effort and the limitation to a vocabulary of known markers limit the power of such approaches. Here, we perform unsupervised learning to identify anomalies in imaging data as candidates for markers. We propose AnoGAN, a deep convolutional generative adversarial network to learn a manifold of normal anatomical variability, accompanying a novel anomaly scoring scheme based on the mapping from image space to a latent space. Applied to new data, the model labels anomalies, and scores image patches indicating their fit into the learned distribution. Results on optical coherence tomography images of the retina demonstrate that the approach correctly identifies anomalous images, such as images containing retinal fluid or hyperreflective foci.

Philipp Seeböck, Sebastian Waldstein, Sophie Klimscha et al.

The identification and quantification of markers in medical images is critical for diagnosis, prognosis and management of patients in clinical practice. Supervised- or weakly supervised training enables the detection of findings that are known a priori. It does not scale well, and a priori definition limits the vocabulary of markers to known entities reducing the accuracy of diagnosis and prognosis. Here, we propose the identification of anomalies in large-scale medical imaging data using healthy examples as a reference. We detect and categorize candidates for anomaly findings untypical for the observed data. A deep convolutional autoencoder is trained on healthy retinal images. The learned model generates a new feature representation, and the distribution of healthy retinal patches is estimated by a one-class support vector machine. Results demonstrate that we can identify pathologic regions in images without using expert annotations. A subsequent clustering categorizes findings into clinically meaningful classes. In addition the learned features outperform standard embedding approaches in a classification task.