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.

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.

"Exactly one year ago, we spoke about the fact that, in future, it will be possible to diagnose diabetes from the eye using automatic digital retinal screening, without the assistance of an ophthalmologist – 12 months on, MedUni Vienna is right in the middle of this digital revolution." These were the words used by Ursula Schmidt-Erfurth, Head of MedUni Vienna's Department of Ophthalmology and Optometrics to open today's press conference about the ART-2018 Specialist Meeting on new developments in retinal therapy, which is to take place on 1 December. One of the latest developments is automatic diabetes screening, which was recently implemented at MedUni Vienna. Patients flock to the Department to undergo this retinal examination to detect any diabetic changes. It takes just a few minutes and is completely non-invasive" Essentially this technique can detect all stages of diabetic retinal disease – high-resolution digital retinal images with two million pixels are taken and analyzed within seconds – but Big Data offers even more potential: nowadays it is already possible to diagnose an additional 50 other diseases in this way. Diabetes is just the start. And MedUni Vienna is among the global leaders in this digital revolution. The Division of Cardiology led by Christian Hengstenberg within the Department of Medicine II is working on how digital retinal analysis can also be used in future for the early diagnosis of cardiovascular diseases. "This AI medicine is'super human'," emphasizes Schmidt-Erfurth. "The algorithms are quicker and more accurate.

AI is poised to revolutionize medicine. An overview of the field, with selected applications in ophthalmology. From the back of the eye to the front, artificial intelligence (AI) is expected to give ophthalmologists new automated tools for diagnosing and treating ocular diseases. This transformation is being driven in part by a recent surge in attention to AI's medical potential from big players in the digital world like Google and IBM. But, in ophthalmic AI circles, com puterized analytics are being viewed as the path toward more efficient and more objective ways to interpret the flood of images that modern eye care practices produce, according to ophthalmologists involved in these efforts. The most immediately promising computer algorithms are in the field of retinal diseases.

The light-sensitive layer found at the back of a person's eyes contains more than just cells that detect shadows and light -- it also contains information about the health of a person's entire body. And now, artificial intelligence can glean this information from a single snapshot, new research suggests. The new AI algorithm, which analyzes images of this light-sensitive layer of the eye, called the retina, could one day provide on the spot diagnoses of various ailments from diabetes to autoimmune and neurodegenerative diseases, the researchers claim. The AI algorithm was presented by Dr. Ursula Schmidt-Erfurth, the director of the ophthalmology department at the Medical University of Vienna, earlier this month at a scientific meeting in Vienna. Research on the algorithm was published Dec. 8 in the journal Ophthalmology.