UIdataGB: Multi-Class ultrasound images dataset for gallbladder disease detection.

Artificial Intelligence (AI) allows computers to self-develop decision-making algorithms through huge data analysis. In medical investigations, using computers to automatically diagnose diseases is a promising area of research that could change healthcare strategies worldwide. However, it can be challenging to reproduce or/and compare various approaches due to the often-limited datasets comprising medical images. Since there is no open access dataset for the Gallbladder (GB) organ, we introduce, in this study, a large dataset that includes 10,692 GB Ultrasound Images (UI) acquired at high resolution from 1,782 individuals. These UI include many disease types related to the GB, and they are organized around nine important anatomical landmarks. The data in this collection can be used to train machine learning (ML) and deep learning (DL) models for computer-aided detection of GB diseases. It can also help academics conduct comparative studies and test out novel techniques for analyzing UI to explore the medical domain of GB diseases. The objective is then to help move medical imaging forward so that patients get better treatment.

Detection of Gallbladder Disease Types Using Deep Learning: An Informative Medical Method.

Nowadays, despite all the conducted research and the provided efforts in advancing the healthcare sector, there is a strong need to rapidly and efficiently diagnose various diseases. The complexity of some disease mechanisms on one side and the dramatic life-saving potential on the other side raise big challenges for the development of tools for the early detection and diagnosis of diseases. Deep learning (DL), an area of artificial intelligence (AI), can be an informative medical tomography method that can aid in the early diagnosis of gallbladder (GB) disease based on ultrasound images (UI). Many researchers considered the classification of only one disease of the GB. In this work, we successfully managed to apply a deep neural network (DNN)-based classification model to a rich built database in order to detect nine diseases at once and to determine the type of disease using UI. In the first step, we built a balanced database composed of 10,692 UI of the GB organ from 1782 patients. These images were carefully collected from three hospitals over roughly three years and then classified by professionals. In the second step, we preprocessed and enhanced the dataset images in order to achieve the segmentation step. Finally, we applied and then compared four DNN models to analyze and classify these images in order to detect nine GB disease types. All the models produced good results in detecting GB diseases; the best was the MobileNet model, with an accuracy of 98.35%.

MEFV mutations in Tunisian patients suffering from familial Mediterranean fever.

OBJECTIVES: To identify the frequency and distribution of familial Mediterranean fever (FMF) gene (MEFV) mutations in Tunisian patients. PATIENTS AND METHODS: This study was performed in the Genetic Department of Tunis University Hospital. A clinical diagnosis of FMF was made according to published criteria. Mutation screening of the MEFV gene was performed in the Human Genetic Laboratory of the "Faculté de Medecine de Tunis" for 8 mutations including the 5 most common known mutations M694V, V726A, M694l, M680l, and E148Q. The tests performed were polymerase chain reaction (PCR) restriction-digestion for M694V, V726A, M680l, R761H, E148Q; amplification refractory mutation system for A744S, M694l; and PCR-electrophoresis assay for l692del. RESULTS: Of the 139 unrelated patients investigated, 61 (44%) had 1 or 2 mutations. In 78 (56%) probands no mutation was identified: 28 patients were homozygous; 16 were compound-heterozygous; 2 had complex alleles; and 17 had only 1 identifiable mutation. Of the mutations, M680l, M694V, M694l, V726A, A744S, R761H, l692DEL, and E148Q accounted for 32, 27, 13, 5, 3, 1, 1, and 18%, respectively. CONCLUSION: The profile of the MEFV gene mutations in the Tunisian population is concordant with other Arab populations but with some differences. M680l is the most common mutation, while V726A, the commonest mutation among Arabs, is rare in our population.

RRH, encoding the RPE-expressed opsin-like peropsin, is not mutated in retinitis pigmentosa and allied diseases.

Many genes from retinoid metabolism cause retinitis pigmentosa. Peropsin, an opsin-like protein with unknown function, is specifically expressed in apical retinal pigment epithelium microvilli. Since rhodopsin and RGR, another opsin-like protein, cause retinitis pigmentosa, we used D-HPLC to screen for the peropsin gene RRH in 331 patients (288 with retinitis pigmentosa and 82 with other retinal dystrophies). We found 13 nonpathogenic variants only, among which a c.730_731delATinsG that truncates the last two transmembrane-spanning fragments and the Lys284 required for retinol binding, but does not segregate with the disease phenotype. We conclude that RRH is not a frequent gene in retinitis pigmentosa.

Homozygous mutation in MERTK causes severe autosomal recessive retinitis pigmentosa.

PURPOSE: Gene identification in retinitis pigmentosa is a prerequisite to future therapies. Accordingly, autosomal recessive retinitis pigmentosa families were genotyped to search for causative mutations. METHODS: Members of a consanguineous Moroccan family had standard ophthalmologic examination, optical coherence tomography-3 scan, autofluorescence testing, and electroretinogram. Their DNA was genotyped with the 250K SNP microchip (Affymetrix) and homozygosity mapping was done. MERTK exons were polymerase chain reaction amplified and sequenced. RESULTS: Two sisters and one brother out of 6 siblings had rod cone dystrophy type of retinitis pigmentosa. Salient features were night blindness starting in early infancy, dot-like whitish deposits in fovea and macula with corresponding autofluorescent dots in youngest patients, decreased visual acuity, and cone responses higher than rod responses at electroretinogram. The patients were homozygous in regions from chromosomes 2 and 8, but only that of chromosome 2 was inherited from a common ancestor. Sequencing of the MERTK gene belonging to the chromosome 2 region showed that the 3 affected patients carried a novel homozygous mutation in exon 17, c.2323C>T, leading to p.Arg775X, while their unaffected brothers and sister, parents, and paternal grandfather were heterozygous. CONCLUSIONS: MERTK mutations lead to severe retinitis pigmentosa with discrete dot-like autofluorescent deposits at early stages, which are a hallmark of this MERTK-specific dystrophy.

Screening genes of the visual cycle RGR, RBP1 and RBP3 identifies rare sequence variations.

The visual cycle is essential for vision and several genes encoding proteins of the cycle have been found mutated in various forms of inherited retinal dystrophy. We screened 3 genes of the visual cycle. RGR, encoding the retinal pigment epithelium (RPE) G protein-coupled receptor acting in vitro as a photoisomerase; RBP1, encoding the ubiquitous cellular retinol binding protein carrying intracellular all-trans retinoids; RBP3, encoding the interphotoreceptor retinoid binding protein, a retinal-specific protein which shuttles all-trans retinol from photoreceptors to RPE and 11-cis retinal from RPE to photoreceptors. We used denaturing high performance liquid chromatography (D-HPLC) and direct sequencing to screen 216 patients (134 with autosomal recessive or sporadic retinitis pigmentosa (RP) and 82 with other retinal dystrophies) for RBP1 and RBP3, and 331 patients for RGR (79 cases with autosomal dominant RP and 36 RP cases with undetermined inheritance were added to the 216 previous patients). Several variants were found in the 3 genes, including unique amino acid changes, but none of them showed evidence of pathogenicity. It is likely that mutations in RGR, RBP3, and possibly RBP1 occur rarely in inherited retinal dystrophies.