A Clinically Explainable AI-Based Grading System for Age-Related Macular Degeneration Using Optical Coherence Tomography.

We propose an automated, explainable artificial intelligence (xAI) system for age-related macular degeneration (AMD) diagnosis. Mimicking the physician's perceptions, the proposed xAI system is capable of deriving clinically meaningful features from optical coherence tomography (OCT) B-scan images to differentiate between a normal retina, different grades of AMD (early, intermediate, geographic atrophy (GA), inactive wet or active neovascular disease [exudative or wet AMD]), and non-AMD diseases. Particularly, we extract retinal OCT-based clinical imaging markers that are correlated with the progression of AMD, which include: (i) subretinal tissue, sub-retinal pigment epithelial tissue, intraretinal fluid, subretinal fluid, and choroidal hypertransmission detection using a DeepLabV3+ network; (ii) detection of merged retina layers using a novel convolutional neural network model; (iii) drusen detection based on 2D curvature analysis; (iv) estimation of retinal layers' thickness, and first-order and higher-order reflectivity features. Those clinical features are used to grade a retinal OCT in a hierarchical decision tree process. The first step looks for severe disruption of retinal layers' indicative of advanced AMD. These cases are analyzed further to diagnose GA, inactive wet AMD, active wet AMD, and non-AMD diseases. Less severe cases are analyzed using a different pipeline to identify OCT with AMD-specific pathology, which is graded as intermediate-stage or early-stage AMD. The remainder is classified as either being a normal retina or having other non-AMD pathology. The proposed system in the multi-way classification task, evaluated on 1285 OCT images, achieved 90.82% accuracy. These promising results demonstrated the capability to automatically distinguish between normal eyes and all AMD grades in addition to non-AMD diseases.

Precise higher-order reflectivity and morphology models for early diagnosis of diabetic retinopathy using OCT images.

This study proposes a novel computer assisted diagnostic (CAD) system for early diagnosis of diabetic retinopathy (DR) using optical coherence tomography (OCT) B-scans. The CAD system is based on fusing novel OCT markers that describe both the morphology/anatomy and the reflectivity of retinal layers to improve DR diagnosis. This system separates retinal layers automatically using a segmentation approach based on an adaptive appearance and their prior shape information. High-order morphological and novel reflectivity markers are extracted from individual segmented layers. Namely, the morphological markers are layer thickness and tortuosity while the reflectivity markers are the 1st-order reflectivity of the layer in addition to local and global high-order reflectivity based on Markov-Gibbs random field (MGRF) and gray-level co-occurrence matrix (GLCM), respectively. The extracted image-derived markers are represented using cumulative distribution function (CDF) descriptors. The constructed CDFs are then described using their statistical measures, i.e., the 10th through 90th percentiles with a 10% increment. For individual layer classification, each extracted descriptor of a given layer is fed to a support vector machine (SVM) classifier with a linear kernel. The results of the four classifiers are then fused using a backpropagation neural network (BNN) to diagnose each retinal layer. For global subject diagnosis, classification outputs (probabilities) of the twelve layers are fused using another BNN to make the final diagnosis of the B-scan. This system is validated and tested on 130 patients, with two scans for both eyes (i.e. 260 OCT images), with a balanced number of normal and DR subjects using different validation metrics: 2-folds, 4-folds, 10-folds, and leave-one-subject-out (LOSO) cross-validation approaches. The performance of the proposed system was evaluated using sensitivity, specificity, F1-score, and accuracy metrics. The system's performance after the fusion of these different markers showed better performance compared with individual markers and other machine learning fusion methods. Namely, it achieved [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text], respectively, using the LOSO cross-validation technique. The reported results, based on the integration of morphology and reflectivity markers and by using state-of-the-art machine learning classifications, demonstrate the ability of the proposed system to diagnose the DR early.

Genetic profiling of CAH Egyptian children: rapid guide to clinical interpretation of common mutations.

OBJECTIVES: The prevalence of CAH in Egypt is reported to be ten times more than that of the worldwide prevalence. The study aimed at genetic screening of children diagnosed with 21-alpha hydroxylase deficiency congenital adrenal hyperplasia (21OHD-CAH). In addition, the study offers a rapid and easy guide for clinical reporting of common mutations for endocrinologists. METHODS: A cohort of 174 unrelated Egyptian children with 21OHD-CAH were screened for 11 common CYP21A2 gene mutations using a strip hybridization assay, and then, bioinformatics analysis was done to report the pathogenicity of the common mutations for clinical classification. RESULTS: The most common mutations were I2 splice and p.Q318X. Deletions/conversions comprised 45.9% of the cohort, whereas 7.4% of the cases were negative for all mutations. The least positively detected point mutations were p.P453S, cluster E6, p.R483P, and p.L307FS, which were detected in fewer than 5% of cases. CONCLUSION: Strip hybridization assay is a rapid screening tool for the diagnosis of CAH. The authors hypothesized an easy and rapid scheme for clinical interpretation of the strip results to gain the highest value of the strip in diagnosis.

High frequency of splice site mutation in 21-hydroxylase deficiency children.

PURPOSE: Steroid 21-hydroxylase deficiency (21-OHD) is the common type of congenital adrenal hyperplasia (CAH) caused by defects in the CYP21A2 gene, as an autosomal recessive disease, genetic analysis has a prominent role in its diagnosis. Our objectives were to determine the prevalence of common mutations in a group of Egyptian patients with 21-OHD and their families using rapid methods, and also to detect the rate of deletion, duplication and conversions in CYP21A2 gene. METHODS: Rapid detection methods were used: allele-specific PCR for c.293-13A>G (g.659A>G), c.518T>A (p.I172N) variants and c.332_339del (8-bp deletion in exon 3), and real-time, quantitative PCR assay was used to detect deletion in the CYP21A2 gene. 29 Egyptian patients, 38 family members, and 20 healthy controls were all included in the study. RESULTS: The frequency of c.293-13A>G splice mutation was reported in 96.6 % cases, G allele had 2.5-folds higher risk to develop CAH than other alleles. The c.518T>A mutation was reported in 69 % cases, children carrying the mutant allele were 2.1 times more risk. The most frequent combined mutations detected were c.293-13A/C>G/c.518T>A in 58.6 % cases. CONCLUSION: The genetic analysis of the splice site mutation c.293-13A>G and c.518T>A variant can be used as good biomarkers for early detection of cases and carriers in 21-OHD CAH Egyptian children, since the methods used have rapid turnaround time.