Search

Diabetes is a chronic disease characterized by excess sugar in the blood and affects 422 million people worldwide, including 3.3 million in France. One of the frequent complications of diabetes is diabetic retinopathy (DR): it is the leading cause of blindness in the working population of developed countries. As a result, ophthalmology is on the verge of a revolution in screening, diagnosing, and managing of pathologies. This upheaval is led by the arrival of technologies based on artificial intelligence. The "Evaluation intelligente de la r\'etinopathie diab\'etique" (EviRed) project uses artificial intelligence to answer a medical need: replacing the current classification of diabetic retinopathy which is mainly based on outdated fundus photography and providing an insufficient prediction precision. EviRed exploits modern fundus imaging devices and artificial intelligence to properly integrate the vast amount of data they provide with other available medical data of the patient. The goal is to improve diagnosis and prediction and help ophthalmologists to make better decisions during diabetic retinopathy follow-up. In this study, we investigate the fusion of different modalities acquired simultaneously with a PLEXElite 9000 (Carl Zeiss Meditec Inc. Dublin, California, USA), namely 3-D structural optical coherence tomography (OCT), 3-D OCT angiography (OCTA) and 2-D Line Scanning Ophthalmoscope (LSO), for the automatic detection of proliferative DR., Comment: Abstract

Diabetic Retinopathy (DR) is a severe complication of diabetes that can cause blindness. Although effective treatments exist (notably laser) to slow the progression of the disease and prevent blindness, the best treatment remains prevention through regular check-ups (at least once a year) with an ophthalmologist. Optical Coherence Tomography Angiography (OCTA) allows for the visualization of the retinal vascularization, and the choroid at the microvascular level in great detail. This allows doctors to diagnose DR with more precision. In recent years, algorithms for DR diagnosis have emerged along with the development of deep learning and the improvement of computer hardware. However, these usually focus on retina photography. There are no current methods that can automatically analyze DR using Ultra-Wide OCTA (UW-OCTA). The Diabetic Retinopathy Analysis Challenge 2022 (DRAC22) provides a standardized UW-OCTA dataset to train and test the effectiveness of various algorithms on three tasks: lesions segmentation, quality assessment, and DR grading. In this paper, we will present our solutions for the three tasks of the DRAC22 challenge. The obtained results are promising and have allowed us to position ourselves in the TOP 5 of the segmentation task, the TOP 4 of the quality assessment task, and the TOP 3 of the DR grading task. The code is available at \url{https://github.com/Mostafa-EHD/Diabetic_Retinopathy_OCTA}.

Longitudinal imaging is able to capture both static anatomical structures and dynamic changes in disease progression towards earlier and better patient-specific pathology management. However, conventional approaches for detecting diabetic retinopathy (DR) rarely take advantage of longitudinal information to improve DR analysis. In this work, we investigate the benefit of exploiting self-supervised learning with a longitudinal nature for DR diagnosis purposes. We compare different longitudinal self-supervised learning (LSSL) methods to model the disease progression from longitudinal retinal color fundus photographs (CFP) to detect early DR severity changes using a pair of consecutive exams. The experiments were conducted on a longitudinal DR screening dataset with or without those trained encoders (LSSL) acting as a longitudinal pretext task. Results achieve an AUC of 0.875 for the baseline (model trained from scratch) and an AUC of 0.96 (95% CI: 0.9593-0.9655 DeLong test) with a p-value < 2.2e-16 on early fusion using a simple ResNet alike architecture with frozen LSSL weights, suggesting that the LSSL latent space enables to encode the dynamic of DR progression., Accepted preprint for presentation at MICCAI-OMIA

International audience; In the last decades, large datasets of fundus photographs have been collected in diabetic retinopathy (DR) screening networks. Through deep learning, these datasets were used to train automatic detectors for DR and a few other frequent pathologies, with the goal to automate screening. One challenge limits the adoption of such systems so far: automatic detectors ignore rare conditions that ophthalmologists currently detect, such as papilledema or anterior ischemic optic neuropathy. The reason is that standard deep learning requires too many examples of these conditions. However, this limitation can be addressed with few-shot learning, a machine learning paradigm where a classifier has to generalize to a new category not seen in training, given only a few examples of this category. This paper presents a new few-shot learning framework that extends convolutional neural networks (CNNs), trained for frequent conditions, with an unsupervised probabilistic model for rare condition detection. It is based on the observation that CNNs often perceive photographs containing the same anomalies as similar, even though these CNNs were trained to detect unrelated conditions. This observation was based on the t-SNE visualization tool, which we decided to incorporate in our probabilistic model. Experiments on a dataset of 164,660 screening examinations from the OPHDIAT screening network show that 37 conditions, out of 41, can be detected with an area under the ROC curve (AUC) greater than 0.8 (average AUC: 0.938). In particular, this framework significantly outperforms other frameworks for detecting rare conditions, including multitask learning, transfer learning and Siamese networks, another few-shot learning solution. We expect these richer predictions to trigger the adoption of automated eye pathology screening, which will revolutionize clinical practice in ophthalmology.

The purpose of this study is to evaluate the performance of the OphtAI system for the automatic detection of referable diabetic retinopathy (DR) and the automatic assessment of DR severity using color fundus photography. OphtAI relies on ensembles of convolutional neural networks trained to recognize eye laterality, detect referable DR and assess DR severity. The system can either process single images or full examination records. To document the automatic diagnoses, accurate heatmaps are generated. The system was developed and validated using a dataset of 763,848 images from 164,660 screening procedures from the OPHDIAT screening program. For comparison purposes, it was also evaluated in the public Messidor-2 dataset. Referable DR can be detected with an area under the ROC curve of AUC = 0.989 in the Messidor-2 dataset, using the University of Iowa's reference standard (95% CI: 0.984-0.994). This is significantly better than the only AI system authorized by the FDA, evaluated in the exact same conditions (AUC = 0.980). OphtAI can also detect vision-threatening DR with an AUC of 0.997 (95% CI: 0.996-0.998) and proliferative DR with an AUC of 0.997 (95% CI: 0.995-0.999). The system runs in 0.3 seconds using a graphics processing unit and less than 2 seconds without. OphtAI is safer, faster and more comprehensive than the only AI system authorized by the FDA so far. Instant DR diagnosis is now possible, which is expected to streamline DR screening and to give easy access to DR screening to more diabetic patients.

International audience; Many surgical computer-aided projects ( CAD ) have emerged in recent years, but none interested in cataract surgery . The aim of our study was to develop a method able to recognize in real time the video stream, identified surgical activities and classify them in a database. We propose to analyze the video stream using a new method describing different levels of granularity : procedure, phases, steps and activities.

International audience; In this paper, we address the problem of medical diagnosis aid through content based image retrieval methods (CBIR). We propose to characterize images without extracting local features, by using global information extracted from the image Bidimensional Empirical Mode Decomposition (BEMD). This method decompose image into a set of functions named Bidimensional Intrinsic Mode Functions (BIMF) and a residue. The generalized Gaussian density function (GGD) is used for modelling the coefficients derived from each BIMF, and to measure similarity between images we compute the similarity between GGDs by using the Kullback-Leibler Divergence (KLD). Retrieval efficiency is given for different databases including a diabetic retinopathy , mammography and a face database. Results are promising: the retrieval efficiency is higher than 95% for some cases .

In this paper, we address the problem of medical ddiagnosis aid through content based image retrieval methods. We propose to characterize images without extracting local features, by using global information extracted from the image Bidimensional Empirical Mode Decomposition (BEMD). This method decompose image into a set of functions named Intrinsic Mode Functions (IMF) and a residu. The generalized Gaussian density function (GGD) is used to represent the coefficients derived from each IMF, and the Kullback–Leibler Distance (KLD) compute the similarity between GGDs. Retrieval efficiency is given for different databases including a diabetic retinopathy, and a face database. Results are promising: the retrieval efficiency is higher than 85% for some cases.

International audience; Purpose: We propose an automatic method to detect microaneurysms in retina photographs. Automating this task, which is currently performed manually, would bring more objectivity and reproducibility in the current clinical practice. Methods: The screening method follows several steps. As a template of microaneurysms we use a generalized Gaussian. This template and retina Images are decomposed in several subbands by using the wavelet transform. We propose to detect the lesions by locally matching the template in a selection of subbands. To improve the method performance, we have searched for the best adapted wavelet within the lifting scheme framework. Results: Results are evaluated on 120 retinal images analyzed by an expert and the optimal wavelet is compared to different conventional mother wavelets. Depending on the imaging modality (color photographs, green filtered photographs and angiographs), microaneurysms were detected with a sensitivity of respectively 89.62%, 90.24% and 93.74% and a positive predictive value of respectively 89.50%, 89.75% and 91.67%, which is better than previously published methods. Conclusions: In this study, we have proposed a new method for detecting microaneurysms in retina. Given its simplicity and relative genericity, this method may be an answer for large screening of Diabetic Retinopathy.

International audience; Purpose: we propose to use different Case Based Reasoning (CBR) methods to retrieve diabetic patient files: should an ophthalmologist have doubts on his diagnosis, he can send the available data about his patient to the system, that selects the most similar cases, along with their medical interpretations, in order to assist his diagnosis. Methods: the system is applied to a diabetic retinopathy (DR) database, consisting of 67 patient files. Each patient file consists of up to 20 retinal images and up to 11 information fields about the patient's medical history; each patient has been graded according to its disease severity level (ICDRS classification). The objective is to retrieve patient files with the same grade. Information about each patient is incomplete and heterogeneous (there are both digital images and text information), so we proposed three different retrieval methods suitable to manage both issues: the first one is based on decision trees, the second one on a Bayesian network, and the third one on the DSmT theory. Images are characterized by their digital content: a feature vector is extracted from their wavelet transform to describe their textural content, and microaneurysms are detected. The retrieval methods are assessed by the mean precision at 5, i.e. the mean percentage of cases relevant for a query, among the topmost 5 results. Results: a mean precision at 5 of 81.0%, 70.4% and 81.8% was achieved with these three methods, respectively. The scores range from 58.9% for the less frequent grade to 87.4% for the most frequent; they range from 42.1% for sparse cases (10% of available data) to 91.2% for comprehensive cases. Conclusions: retrieving entire patient files, we can achieve a significantly higher precision than simply retrieving similar images (a mean precision at 5 of 46.1% is achieved in the latter case). The proposed multimodal retrieval methods are precise enough to be useful in a DR diagnosis aid system.

International audience; The Télé Orthoptie pour Déficients Visuels (TéoDéVi) project, involves developing a platform which allows an orthoptist to propose remote training exercises, while undertaking normal consultations, thanks to a distance learning system. TéoDéVi is managed jointly by the computer science department at ENST-Bretagne (Ecole Nationale Supérieure des Télécommunications de Bretagne) and the LATIM (Laboratoire de Traitement de l’Information Médicale, INSERM U650) which is a research laboratory composed of the Brest CHU (Teaching Hospital), U.B.O (Université de Bretagne Occidentale) and ENST-Bretagne.