The training times of athletes could play a role in clinical sleep problems due to their associations with sleep difficulty scores.

OBJECTIVES: Sleep is a key component of athletic recovery, yet training times could influence the sleep of athletes. The aim of the current study was to compare sleep difficulties in athletes across different training time groups (early morning, daytime, late evening, early morning plus late evening) and to investigate whether training time can predict sleep difficulties. METHODS: Athletes from various sports who performed at a national-level (n = 273) answered the Athlete Sleep Screening Questionnaire (ASSQ) along with several other questionnaires related to demographics, exercise training, and mental health. From the ASSQ, a Sleep Difficulty Score (SDS) was calculated. Transformed SDS (tSDS) was compared across different training time categories using multiple one-way ANOVAs. A stepwise regression was then used to predict tSDS from various sleep-related factors. RESULTS: SDSs ranged from none (31%), mild (38%), moderate (22%), and severe (9%). However, the one-way ANOVAs revealed training earlier or later vs. training daytime shifted the tSDS in a negative direction, a trend toward increased sleep difficulty. In particular, athletes training in the late evening (>20:00 or >21:00) had a significantly higher tSDS when compared to daytime training (p = .03 and p < .01, respectively). The regression model (p < .001) explained 27% of variance in the tSDS using depression score, age, training time, and chronotype score. CONCLUSION: Among a heterogeneous sample of national-level athletes, 31% displayed moderate to severe SDSs regardless of their training time. However, when athletes trained outside daytime hours there was a tendency for the prevalence of sleep difficulties to increase.

Analysis of Movement and Actions of Wingers as Second-Line Players in Organized Attack in Handball.

In modern handball, one of the important performance indicators is the effectiveness of the attack, especially the running-in of wingers as line players which has not been explored adequately. The purpose of the study was to analyze the movements of wingers in the organized attack when they run in. Fifty-eight matches were analyzed from the 2022 EHF European Men's Handball Championship. A total of 491 attacks were recorded and 45 variables were analyzed in which wingers ran in as second pivots and the outcome was a throw. For the statistical analysis, descriptive and inductive statistics were used. The results showed that the average time of the running-in was 12.11 ± 9.28 s, the left wing ran in more often (60.1%) and wingers tended to move outside the defense formation (72.5%). Wingers ran in without the ball in possession (81.1%), moved toward defenders 2 and 5 and stood next to defenders, and occasionally blocked, slid or left their position. The defense's central zone was preferred by players to make a throw. Correspondence analysis showed that wingers finally returned to their initial position regardless of the evolution of the attack. Their role was to block a specific defender or disorganize the defense, for one defender to be isolated so that one attacking player could execute a shot from the central area. Conclusively, wingers play an important role, especially in run-in actions, at the completion of an attack, and in the final throw.

The role of physical activity on obstructive sleep apnea severity and hypoxic load, and the mismatch between subjective and objective physical activity assessments.

Obesity is the primary risk factor for the development of obstructive sleep apnea, and physical inactivity plays an important role. However, most studies have either only evaluated physical activity subjectively or objectively in obstructive sleep apnea. The objectives of this study were: (i) to assess the relationship between obstructive sleep apnea severity (both apnea-hypopnea index and desaturation parameters) and both objectively and subjectively measured physical activity after adjustment for anthropometry and body composition parameters; and (ii) to assess the relationship between objective and subjective physical activity parameters and whether obstructive sleep apnea severity has a modulatory effect on this relationship. Fifty-four subjects (age 47.7 ± 15.0 years, 46% males) were categorized into groups according to obstructive sleep apnea severity: no obstructive sleep apnea; mild obstructive sleep apnea; and moderate-to-severe obstructive sleep apnea. All subjects were evaluated with subjective and objective physical activity, anthropometric and body composition measurements, and 3-night self-applied polysomnography. A one-way ANOVA was used to evaluate the differences between the three obstructive sleep apnea severity groups and multiple linear regression to predict obstructive sleep apnea severity. Differences in subjectively reported sitting time (p ≤ 0.004) were found between participants with moderate-to-severe obstructive sleep apnea, and those with either mild or no obstructive sleep apnea (p = 0.004). Age, body mass index and neck circumference explained 63.3% of the variance in the apnea-hypopnea index, and age, body mass index and visceral adiposity explained 67.8% of the variance in desaturation parameters. The results showed that the person's physical activity does not affect obstructive sleep apnea severity. A weak correlation was found between objective and subjective physical activity measures, which could be relevant for healthcare staff encouraging patients with obstructive sleep apnea to increase their physical activity.

Review and perspective on sleep-disordered breathing research and translation to clinics.

Sleep-disordered breathing, ranging from habitual snoring to severe obstructive sleep apnea, is a prevalent public health issue. Despite rising interest in sleep and awareness of sleep disorders, sleep research and diagnostic practices still rely on outdated metrics and laborious methods reducing the diagnostic capacity and preventing timely diagnosis and treatment. Consequently, a significant portion of individuals affected by sleep-disordered breathing remain undiagnosed or are misdiagnosed. Taking advantage of state-of-the-art scientific, technological, and computational advances could be an effective way to optimize the diagnostic and treatment pathways. We discuss state-of-the-art multidisciplinary research, review the shortcomings in the current practices of SDB diagnosis and management in adult populations, and provide possible future directions. We critically review the opportunities for modern data analysis methods and machine learning to combine multimodal information, provide a perspective on the pitfalls of big data analysis, and discuss approaches for developing analysis strategies that overcome current limitations. We argue that large-scale and multidisciplinary collaborative efforts based on clinical, scientific, and technical knowledge and rigorous clinical validation and implementation of the outcomes in practice are needed to move the research of sleep-disordered breathing forward, thus increasing the quality of diagnostics and treatment.

Are they all born to score? The relationship between throwing arm and scoring from the 7-meter line in semi-professional handball.

ABSTRACTIndications of laterall biases favouring left-handers have been found in various sports; especially interactive sports where the athletes have limited time to react to incoming objects. The aim of this study was therefore to explore whether any lateral biases exist in handball by examining 7-meter shots. A total of 6846 7-meter throws from 240 7-meter shooters across four seasons in the semi-professional Icelandic elite handball division (male and female) were analyzed. Out of the 240 7-meter shooters, of which 151 were male and 89 were female, 22% were left-handed (22% of the males and 20% of the females). The left-handed 7-meter shooters took a disproportionate number of the 7-meter shots, with left-handed shooters performing 29% of the 7-meter shots (27% in the male league and 33% in the female league). The results of a Bayesian two-level analysis indicated that left-handedness is not associated with greater success from the 7-meter line at the semi-professional level.

Analyzing the Relationship between Participation in Sports and Harmful Alcohol Drinking in Early Adolescence: Two-Year Prospective Analysis.

Alcohol drinking is an important health-related problem and one of the major risk factors for a wide array of non-communicable diseases, while there is a lack of studies investigating environment-specific associations between sports participation and alcohol drinking in adolescence. This study prospectively investigated the relationship between sports factors (i.e., participation in sports and competitive achievement), with the prevalence of harmful alcohol drinking (HD), and HD initiation in 14-to-16 years old adolescents from Bosnia and Herzegovina (n = 641, 337 females, 43% living in rural community). Participants were tested over 4-time points divided by approximately 6 months, from the beginning of high school to the end of the second grade. Variables included gender, factors related to sport participation, a community of residence (urban or rural), and outcome: alcohol consumption was assessed by the AUDIT questionnaire. Results evidenced that the prevalence of HD increased over the study period from 6 to 19%, with no significant differences between urban and rural youth. Logistic regression for HD as criterion evidenced adolescents who participated in sports and then quit as being at particular risk for drinking alcohol at the study baseline. Sports factors were not correlated with HD initiation in the period between 14 and 16 years of age. It seems that the problem of alcohol drinking should be preventively targeted in all youth, irrespective of living environment. Although sports participation was not evidenced as being a factor of influence on HD initiation, results highlight the necessity of developing targeted preventive campaigns against alcohol drinking for adolescents who quit sports.

Effectiveness of Growing-Up Milk Containing Only A2 ß-Casein on Digestive Comfort in Toddlers: A Randomized Controlled Trial in China.

BACKGROUND: Emerging clinical evidence indicates the potential gastrointestinal (GI) benefits of milk containing only A2 ß-casein, but data from randomized controlled trials is sparse among pediatric populations. We aimed to evaluate the effectiveness of growing-up milk (GUM) containing only A2 ß-casein on GI tolerance in toddlers. METHODS: A total of 387 toddlers aged 12-36 months were recruited in Beijing, China, and randomized in a 1:1:1 ratio to consume one of two commercially available A2 GUMs (combined in the analysis as A2 GUM) or continue their current feeding regimen of conventional milk for 14 days. The primary outcome was the total Gut Comfort Score (GCS) (range: 10-60; higher values indicate greater GI distress) derived from a 10-item (score range: 1-6 per item) parent-reported questionnaire, reflecting GI tolerance. RESULTS: The GCS (mean ± SD) was comparable between the A2 GUM and conventional milk groups on day 7 (14.7 ± 5.0 vs. 15.0 ± 6.1, p = 0.54) and day 14 (14.0 ± 4.5 vs. 14.3 ± 5.5, p = 0.51). Parents reported less constipation in those consuming A2 GUM vs. conventional milk on day 14 (1.3 ± 0.6 vs. 1.4 ± 0.9, p = 0.020). Among 124 participants with minor GI distress at baseline (GCS ≥ 17, top tertile range 17-35), GCS was significantly lower in those consuming A2 GUM on day 7 (18.2 ± 5.1 vs. 21.2 ± 6.8, p = 0.004) and day 14 (17.1 ± 5.3 vs. 19.6 ± 6.3, p = 0.026), as were individual GI symptoms (all p < 0.05). In the toddlers without GI issues at baseline (GCS < 17), a low GCS was maintained throughout the study period after switching to A2 GUM (mean values range 10-13). CONCLUSIONS: Growing-up milk containing only A2 ß-casein were well-tolerated and associated with lower parent-reported constipation scores after two weeks when compared to conventional milks. In healthy toddlers with minor GI distress, A2 GUM improved overall digestive comfort and GI-related symptoms within one week.

Peripheral venous catheter-related phlebitis: A meta-analysis of topical treatment.

AIM: To systematically evaluate the efficacy of different topical treatments for PVC-related phlebitis in hospital in-patients. DESIGN: A systematic review and meta-analysis. METHODS: A selection was made of experimental and quasi-experimental studies published in English or Spanish. These should provide data on the degree of phlebitis, pain and infiltration (means and standard deviations, mainly) of hospitalized patients with phlebitis secondary to peripheral venous catheter. All those studies that reflected systemic or exclusive prevention treatments were excluded. Searches were from inception to April 2020. The date of data collection was from December 2020 to May 2021. The selection criteria were based on the PICOS model. Risk of bias was assessed using the Cochrane Collaboration tool. RESULTS: Twelve studies (726 patients) met the inclusion criteria. With respect to the decrease in the degree of phlebitis, was found ichthammol glycerine, followed by heparinoids. As for degree of pain, sesame oil obtained the most marked reduction. In terms of degree of infiltration, heparinoids and ichthammol glycerine were the only products to achieve a statistically significant reduction. The most important limitations are the low quantity and quality of the trials included. Insufficient data are available to draw valid conclusions about the efficacy of any treatment.

Nutrition in school-age children: a rationale for revisiting priorities.

Middle childhood and early adolescence have received disproportionately low levels of scientific attention relative to other life stages, especially as related to nutrition and health. This is partly due to the justified emphasis on the first 1000 days of life, and the idea that early deficits and consequences may not be fully reversible. In addition, these stages of life may superficially appear less "eventful" than infancy or late adolescence. Finally, there has been historical ambiguity and inconsistency in terminology, depending on whether viewing "childhood" through physiologic, social, legal, or other lenses. Nevertheless, this age bracket, which encompasses most of the primary education and basic schooling years for most individuals, is marked by significant changes, inflection points, and sexually driven divergence in somatic and brain growth and development trajectories. These constitute transformative changes, and thus middle childhood and early adolescence represents a major and last opportunity to influence long-term health and productivity. This review highlights the specificities of growth and development in school age, with a focus on middle childhood and early adolescence (5 years-15 years of age, for the purposes of this review), the role of nutrition, the short- and long-term consequences of inadequate nutrition, and the current global status of nutrition in this age group. Adequate attention and emphasis on nutrition in the school-age years is critical: (a) for maintaining an adequate course of somatic and cognitive development, (b) for taking advantage of this last major opportunity to correct deficits of undernutrition and "catch-up" to normal life course development, and (c) for addressing the nutritional inadequacies and mitigating the longer-term consequences of overnutrition. This review summarizes and provides a rationale for prioritizing nutrition in school-age children, and for the need to revisit priorities and focus on this part of the life cycle to maximize individuals' potential and their contribution to society.

Decline in Running Performance in Highest-Level Soccer: Analysis of the UEFA Champions League Matches.

It is widely recognized that there is a decline in match running performance (MRP) towards the end of matches. To clarify whether it is primarily a consequence of fatigue, pacing or situational influences, this study aimed to examine MRP across 15-min match periods for players on different playing positions. Players' MRP (n = 244) were examined from the UEFA Champions League matches (n = 20) using a semiautomatic optical tracking system. Linear mixed models for repeated measures were adjusted to analyze MRP over the six 15-min match periods while controlling the influence of situational factors. No effects of match outcome, match location, team, and opponent quality on total distance (TD) and high-intensity running (HIR) for players in all playing positions were found (F = 0.03−2.75; all p > 0.05). Significant differences in TD (F = 17.57−53.01; η2 = 0.39−0.52, all large effect sizes) and HIR (F = 3.67−7.64; η2 = 0.05−0.19, small to medium effect sizes) among six 15-minute match periods were found for players in all playing positions. In addition, players in all playing positions covered less TD (d = 1.41−2.15, large to very large effect sizes) and HIR (d = 0.16−0.6, trivial to medium effect sizes) in the last compared to the first 15-min match period. No differences in TD and HIR between the last two match periods in the second half were observed. This study confirmed that soccer players reduce MRP towards the end of matches, and suggest that the decline of MPR in highest-level soccer may be a consequence of pacing strategies.

The Sleep Revolution project: the concept and objectives.

Obstructive sleep apnea is linked to severe health consequences such as hypertension, daytime sleepiness, and cardiovascular disease. Nearly a billion people are estimated to have obstructive sleep apnea with a substantial economic burden. However, the current diagnostic parameter of obstructive sleep apnea, the apnea-hypopnea index, correlates poorly with related comorbidities and symptoms. Obstructive sleep apnea severity is measured by counting respiratory events, while other physiologically relevant consequences are ignored. Furthermore, as the clinical methods for analysing polysomnographic signals are outdated, laborious, and expensive, most patients with obstructive sleep apnea remain undiagnosed. Therefore, more personalised diagnostic approaches are urgently needed. The Sleep Revolution, funded by the European Union's Horizon 2020 Research and Innovation Programme, aims to tackle these shortcomings by developing machine learning tools to better estimate obstructive sleep apnea severity and phenotypes. This allows for improved personalised treatment options, including increased patient participation. Also, implementing these tools will alleviate the costs and increase the availability of sleep studies by decreasing manual scoring labour. Finally, the project aims to design a digital platform that functions as a bridge between researchers, patients, and clinicians, with an electronic sleep diary, objective cognitive tests, and questionnaires in a mobile application. These ambitious goals will be achieved through extensive collaboration between 39 centres, including expertise from sleep medicine, computer science, and industry and by utilising tens of thousands of retrospectively and prospectively collected sleep recordings. With the commitment of the European Sleep Research Society and Assembly of National Sleep Societies, the Sleep Revolution has the unique possibility to create new standardised guidelines for sleep medicine.

The Changing Landscape of Children's Diet and Nutrition: New Threats, New Opportunities.

BACKGROUND: Over the last 30-40 years, we have seen an improvement in global child undernutrition, with major reductions in wasting and stunting. Meanwhile, childhood obesity has dramatically increased, initially in high-income populations and subsequently in the more economically vulnerable. These trends are related to significant changes in diet and external factors, including new environmental threats. SUMMARY: Obesity rates first increased in older children, then gradually in infants. And in the next couple of years, there will be more overweight and obese than moderately or severely underweight children in the world. The changes in childhood nutritional landscape are a result of poor diets. Today, almost 50% of the world's population consumes either too many or too few calories. Dietary disparities between countries result in disparities of under- and overnutrition and impact the global health landscape. Most children with obesity, wasting, and micronutrient deficiencies live in lower income countries and in lower income families within any country. High energy-low nutrient diets are contributing to the increase in non-communicable diseases, which will manifest later in this generation of children. In 1990, child wasting was the #1 leading risk factor for mortality for all ages, and high BMI was #16; today, they are #11 and #5, respectively. COVID-19 and climate change are new major threats to global nutrition. Current and future efforts to improve the state of child nutrition require multisectoral approaches to reprioritize actions which address current trends and emerging threats.

Complete Lipopolysaccharide of Piscirickettsia salmonis Is Required for Full Virulence in the Intraperitoneally Challenged Atlantic Salmon, Salmo salar, Model.

Bacterial cell envelopes play a critical role in host-pathogen interactions. Macromolecular components of these structures have been closely linked to the virulence of pathogens. Piscirickettsia salmonis is a relevant salmonid pathogen with a worldwide distribution. This bacterium is the etiological agent of piscirickettsiosis, a septicemic disease that causes a high economic burden, especially for the Chilean salmon farming industry. Although P. salmonis has been discovered long ago, its pathogenicity and virulence mechanisms are not completely understood. In this work, we present a genetic approach for producing in-frame deletion mutants on genes related to the biosynthesis of membrane-associated polysaccharides. We provide a detailed in vitro phenotype description of knock-out mutants on wzx and wcaJ genes, which encode predicted lipopolysaccharide (LPS) flippase and undecaprenyl-phosphate glucose phosphotransferase enzymes, respectively. We exhibit evidence that the wzx mutant strain carries a defect in the probably most external LPS moiety, while the wcaJ mutant proved to be highly susceptible to the bactericidal action of serum but retained the ability of biofilm production. Beyond that, we demonstrate that the deletion of wzx, but not wcaJ, impairs the virulence of P. salmonis in an intraperitoneally infected Atlantic salmon, Salmo salar, model of piscirickettsiosis. Our findings support a role for LPS in the virulence of P. salmonis during the onset of piscirickettsiosis.

Un instrumento de evaluación del aprendizaje para el manejo de catéteres venosos periféricos: adaptación, extensión y validación en español / A learning assessment instrument for the management of peripheral venous catheters: adaptation, extension and validation in Spanish

La práctica de enfermería debe basarse en la práctica basada en la evidencia disponible debido a su impacto en la atención segura del paciente. Su uso no solo debe fomentarse en los profesionales, sino que debe iniciarse en la formación en enfermería. El objetivo de este estudio se basa en traducir, adaptar y validar un instrumento de evaluación del conocimiento basado en la evidencia científica disponible en relación al manejo de los catéteres venosos periféricos. Se siguió un proceso en dos etapas: 1) traducción, adaptación y ampliación del instrumento; y 2) pruebas psicométricas. El estudio incluyó 675 estudiantes de enfermería. El coeficiente de consistencia interna alfa de Cronbach fue 0,703 y el coeficiente de correlación intraclase de 0.91. El instrumento se presentó bastante equilibrado en cuanto a dificultad: 46,6% de ítems fáciles, 13.3% de ítems de dificultad media y 53.3% de alta dificultad. En conclusión, es un instrumento simple de utilizar y de puntuar. La versión española tiene unas buenas propiedades psicométricas y proporciona un instrumento válido y fiable para la valoración de los conocimientos basados en la evidencia para un manejo óptimo de catéteres venosos periféricos en personas portadoras. (AU)

Nursing practice should be based on available evidence-based practice because of its impact on patient safety in care. Its use should not only be encouraged in professionals but should begin in nursing education. The objective of this study is to translate, adapt, extend and validate a learning assessment instrument in relation to knowledge of peripheral venous catheter management. A two-stage process was followed: 1) translation, adaptation and extension of the instrument; and 2) psychometric testing. The study included 675 nursing students. Cronbach's alpha internal consistency coefficient was 0.703 and the intraclass correlation coefficient was 0.91. The instrument was fairly balanced in terms of difficulty: 46.6% easy items, 13.3% items of medium difficulty and 53.3% of high difficulty. In conclusion, it is a simple instrument to use and to score. The Spanish version has good psychometric properties and provides a valid and reliable instrument for the assessment of knowledge for the management of catheters. (AU)

Real-World Evidence of Epidemiology and Clinical Outcomes in Multiple Myeloma, Findings from the Registry of Hemato-Oncologic Malignancies in Colombia, Observational Study.

INTRODUCTION/BACKGROUND: Multiple Myeloma (MM) is a plasma cell derived clonal disorder that represents around 1% of all newly diagnosed neoplasms. Limited data regarding MM treatment in Latin America is available, and access to novel agents for a substantial portion of the population is limited by their high costs. MATERIALS (OR PATIENTS) AND METHODS: RENEHOC is a bidirectional (retrospective and prospective) multicenter observational registry of hematological malignancies in Colombia. MM patients included up to July 2020 were analyzed on this report. RESULTS: 890 are reported with a median follow-up of 18 months (IQR: 7-42 months). Patients were classified by age group (≤ or > 65 years). Median age at diagnosis was 67 years (IQR: 59-75 years) and 47.1% of patients were women. 709 patients (79.6%) received Bortezomib-based schemes as part of the first line. Two hundred and fifty-two patients (28.3%) were consolidated with Autologous Stem Cell Transplantation (ASCT) in first-line. ASCT consolidation and age were the main independent factors influencing outcomes; in the non-ASCT cohort, 5-year overall survival was 48.7% (CI 41.8-55.2) compared to 80.7% (CI 73-86.4) in ASCT patients. CONCLUSION: This data depicts the reality of MM in Colombia, which likely reflects other Latin American countries, where access barriers to diagnosis and treatment are echoed in advanced stage diagnosis and a low rate of transplants. These seem to negatively impact survival despite the availability of most novel drugs approved for this disease. Thus, emphasizing the paradox that prevails in most of the region: availability without equitable access.

Urinary fluoride and micronutrients intake in children from San Luis Potosi, Mexico.

To assess the relationship between urinary fluoride and micronutrients intake we recruited 121 schoolchildren from San Luis Potosí. We evaluated fluoride concentrations in drinking water and urine with the ion-selective electrode method and estimated calcium, magnesium, and phosphorus intakes with a validated consumption frequency questionnaire. About 72% of the population used tap water for drinking and cooking. Fluoride concentrations were 2.36 ± 0.02 mg/L in tap water, 0.14 ± 0.04 mg/L in bottled water and 2.05 ± 0.62 mg/L in urine. Urinary fluoride was higher in children with a consumption <50% of the Recommended Daily Intake (RDI) of calcium (2.20 v/s 1.96 mg/L; p < 0.05), <50% RDI of magnesium (2.58 v/s 1.96 and 1.90 mg/L; p < 0.05) and >150% RDI of phosphorus (2.43 v/s 1.82 mg/L; p < 0.05). These data indicated that fluoride concentration in tap water is higher than bottled water and estimated micronutrients intake is related to urinary fluoride.

Estimated Energy Requirements of Infants and Young Children up to 24 Months of Age.

BACKGROUND: Establishing energy requirements in infants and young children is important in developing age-appropriate diet recommendations but most published guidelines for energy requirements have 1 or more limitations related to the data underlying the calculations. OBJECTIVE: To develop a comprehensive set of daily energy requirements for infants and young children aged 0-24 mo meeting the ideals of worldwide applicability to all healthy children based on the use of the doubly labeled water (DLW) technique to measure total energy expenditure (TEE), the use of recent, international growth charts, and calculation of values across a wide range of body weight. METHODS: Daily estimated energy requirements (EERs) were calculated in 1-mo increments from 0 to 24 mo for boys, girls, and combined, using as inputs the following: 1) TEE measured using the DLW technique, 2) energy deposition estimates from the Institute of Medicine, and 3) body weight values from the 25th to 75th percentiles from the 2006 WHO growth charts. EERs were combined for age groups 0 to <6, 6-8, 9-11, and 12-24 mo by averaging EERs from individual months. The EER calculations were supported by a systematic literature review and a meta-regression of existing studies. RESULTS: Energy requirements naturally increase with age and are slightly higher in boys than in girls. The EERs derived in this study are similar to those in other recent international efforts. CONCLUSIONS: This updated set of EERs for infants and young children expand and improve upon the methodology used to establish previous published guidelines. These estimates have multiple potential uses including planning age-appropriate menus for the complementary feeding period, the development of foods that are more precisely targeted to the needs of infants and children at particular ages, and establishing macronutrient requirements within specific age groups based on a percentage of energy, such as dietary fat.

Análisis de la capa de células ganglionares con deep learning en el diagnóstico de glaucoma / Ganglion cell layer analysis with deep learning in glaucoma diagnosis

Objetivo Determinar y comparar la precisión diagnóstica en glaucoma de dos modelos de aprendizaje profundo, usando imágenes en infrarrojo del nervio óptico, del fondo de ojo y de la capa de células ganglionares (CCG). Métodos Hemos seleccionado una muestra de pacientes normales y con glaucoma. Se recogieron tres imágenes en infrarrojo con un tomógrafo de coherencia óptica de tipo spectral-domain (SD-OCT). La primera corresponde a la imagen de barrido confocal del fondo de ojo, la segunda es un recorte de la primera centrada en el nervio óptico, y la tercera fue la imagen del corte SD-OCT de la CCG. Nuestros modelos de aprendizaje profundo se desarrollaron en la plataforma MATLAB con las redes neuronales preentrenadas ResNet50 y VGG19. Resultados Se recogieron 498 ojos de 298 pacientes. De los 498 ojos, 312 son glaucomatosos y 186 son normales. En la prueba, la precisión de los modelos fue de 96% (ResNet50) y 96% (VGG19) para las imágenes de la CCG, de 90% (ResNet50) y 90% (VGG19) para las imágenes del nervio óptico y de 82% (ResNet50) y 84% (VGG19) para las de fondo de ojo. El área ROC en la prueba fue de 0,96 (ResNet50) y 0,97 (VGG19) para las imágenes de la CCG, de 0,87 (ResNet50) y 0,88 (VGG19) para las imágenes del nervio óptico, y de 0,79 (ResNet50) y 0,81 (VGG19) para las imágenes de fondo de ojo. Conclusiones Los dos modelos de aprendizaje profundo analizados, aplicados sobre las imágenes de la CCG, ofrecen una alta precisión diagnóstica, sensibilidad y especificidad en el diagnóstico de glaucoma (AU)

Objective To determine and compare the diagnostic precision in glaucoma of two deep learning models using infrared images of the optic nerve, eye fundus, and the ganglion cell layer (GCL). Methods We have selected a sample of normal and glaucoma patients. Three infrared images were registered with a spectral-domain optical coherence tomography (SD-OCT). The first corresponds to the confocal scan image of the fundus, the second is a cut-out of the first centered on the optic nerve, and the third was the SD-OCT image of the GCL. Our deep learning models are developed on the MatLab platform with the ResNet50 and VGG19 pre-trained neural networks. Results 498 eyes of 298 patients were collected. Of the 498 eyes, 312 are glaucoma and 186 are normal. In the test, the precision of the models was 96% (ResNet50) and 96% (VGG19) for the GCL images, 90% (ResNet50) and 90% (VGG19) for the optic nerve images and 82% (ResNet50) and 84% (VGG19) for the fundus images. The ROC area in the test was 0.96 (ResNet50) and 0.97 (VGG19) for the GCL images, 0.87 (ResNet50) and 0.88 (VGG19) for the optic nerve images, and 0.79 (ResNet50) and 0.81 (VGG19) for the fundus images. Conclusions Both deep learning models, applied to the GCL images, achieve high diagnostic precision, sensitivity and specificity in the diagnosis of glaucoma (AU)

Ganglion cell layer analysis with deep learning in glaucoma diagnosis. / Análisis de la capa de células ganglionares con deep learning en el diagnóstico de glaucoma.

OBJECTIVE: To determine and compare the diagnostic precision in glaucoma of two deep learning models using infrared images of the optic nerve, eye fundus, and the ganglion cell layer (GCL). METHODS: We have selected a sample of normal and glaucoma patients. Three infrared images were registered with a spectral-domain optical coherence tomography (SD-OCT). The first corresponds to the confocal scan image of the fundus, the second is a cut-out of the first centered on the optic nerve, and the third was the SD-OCT image of the GCL. Our deep learning models are developed on the MatLab platform with the ResNet50 and VGG19 pre-trained neural networks. RESULTS: 498 eyes of 298 patients were collected. Of the 498 eyes, 312 are glaucoma and 186 are normal. In the test, the precision of the models was 96% (ResNet50) and 96% (VGG19) for the GCL images, 90% (ResNet50) and 90% (VGG19) for the optic nerve images and 82% (ResNet50) and 84% (VGG19) for the fundus images. The ROC area in the test was 0.96 (ResNet50) and 0.97 (VGG19) for the GCL images, 0.87 (ResNet50) and 0.88 (VGG19) for the optic nerve images, and 0.79 (ResNet50) and 0.81 (VGG19) for the fundus images. CONCLUSIONS: Both deep learning models, applied to the GCL images, achieve high diagnostic precision, sensitivity and specificity in the diagnosis of glaucoma.