Hasil untuk "Pediatrics"

J. Meek, Lawrence Noble

Breastfeeding and human milk are the normative standards for infant feeding and nutrition. The short- and long-term medical and neurodevelopmental advantages of breastfeeding make breastfeeding, or the provision of human milk, a public health imperative. The American Academy of Pediatrics (AAP) recommends exclusive breastfeeding for approximately 6 months after birth. Furthermore, the AAP supports continued breastfeeding, along with appropriate complementary foods introduced at about 6 months, as long as mutually desired by mother and child for 2 years or beyond. These recommendations are consistent with those of the World Health Organization (WHO). Medical contraindications to breastfeeding are rare. The AAP recommends that birth hospitals or centers implement maternity care practices shown to improve breastfeeding initiation, duration, and exclusivity. The Centers for Disease Control and Prevention (CDC) and The Joint Commission monitor breastfeeding practices in US hospitals. Pediatricians play a critical role in hospitals, their practices, and communities as advocates of breastfeeding and, thus, need to be trained about the benefits of breastfeeding for mothers and children and in managing breastfeeding.

J. Slaughter, A. Kemper, T. Newman

CONTEXT Severe hyperbilirubinemia is associated with kernicterus. Informed guidance on hyperbilirubinemia management, including preventive treatment thresholds, is essential to safely minimize neurodevelopmental risk. OBJECTIVE To update the evidence base necessary to develop the 2022 American Academy of Pediatrics clinical practice guideline for management of hyperbilirubinemia in the newborn infant ≥35 weeks' gestation. DATA SOURCE PubMed. STUDY SELECTION English language randomized controlled trials and observational studies. Excluded: case reports or series, nonsystematic reviews, and investigations focused on <35-weeks' gestation infants. DATA EXTRACTION Topics addressed in the previous clinical practice guideline (2004) and follow-up commentary (2009) were updated with new evidence published through March 2022. Evidence reviews were conducted for previously unaddressed topics (phototherapy-associated adverse effects and effectiveness of intravenous immune globulin [IVIG] to prevent exchange transfusion). RESULTS New evidence indicates that neurotoxicity does not occur until bilirubin concentrations are well above the 2004 exchange transfusion thresholds. Systematic review of phototherapy-associated adverse effects found limited and/or inconsistent evidence of late adverse effects, including cancer and epilepsy. IVIG has unclear benefit for preventing exchange transfusion in infants with isoimmune hemolytic disease, with a possible risk of harm due to necrotizing enterocolitis. LIMITATIONS The search was limited to 1 database and English language studies. CONCLUSIONS Accumulated evidence justified narrowly raising phototherapy treatment thresholds in the updated clinical practice guideline. Limited evidence for effectiveness with some evidence of risk of harm support the revised recommendations to limit IVIG use.

Rae-Ellen W. Kavey, D. Simons-Morton, Janet M. de Jesus

; originally published online November 14, 2011; Pediatrics HEALTH AND RISK REDUCTION IN CHILDREN AND ADOLESCENTS EXPERT PANEL ON INTEGRATED GUIDELINES FOR CARDIOVASCULAR Reduction in Children and Adolescents: Summary Report Expert Panel on Integrated Guidelines for Cardiovascular Health and Risk http://pediatrics.aappublications.org/content/early/2011/11/09/peds.2009-2107C.citation located on the World Wide Web at: The online version of this article, along with updated information and services, is of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275. Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2011 by the American Academy published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point publication, it has been published continuously since 1948. PEDIATRICS is owned, PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

R. Altman, J. Randolph, J. Lilly

Nicole Cizauskas, Foteini Strimenopoulou, Svetlana S. Cherlin et al.

Bayesian dynamic borrowing (BDB) and synthetic control methods (SCM) are both used in clinical trial design when recruitment, retention, or allocation is a challenge. The performance of these approaches has not previously been directly compared due to differences in application, product, and measurement metrics. This study aims to conduct a comparison of power and type 1 error rates of BDB (using meta-analytic predictive prior (MAP)) and SCM using a case study of Pediatric Atopic Dermatitis. Six historical randomised control trials were selected for use in both the creation of the MAP prior and synthetic control arm. The R library RBesT was used to create a MAP prior and the R library Synthpop was used to create a synthetic control arm for the SCM. Power and type 1 error rate were used as comparison metrics. BDB produced a power of 0.580 and a type 1 error rate of 0.026. SCM produced a power of 0.641 and a type 1 error rate of 0.027. In this case study, the SCM model produced a higher power than the BDB method with a similar type 1 error rate. However, the decision to use SCM or BDB should come from the specific needs of the potential trial, since their power and type 1 error rate may differ on a case-by-case basis.

Haoran Sun, Zhanpeng Zhu, Anguo Zhang et al.

Pediatric congenital hydronephrosis (CH) is a common urinary tract disorder, primarily caused by obstruction at the renal pelvis-ureter junction. Magnetic resonance urography (MRU) can visualize hydronephrosis, including renal pelvis and calyces, by utilizing the natural contrast provided by water. Existing voxel-based segmentation approaches can extract CH regions from MRU, facilitating disease diagnosis and prognosis. However, these segmentation methods predominantly focus on morphological features, such as size, shape, and structure. To enable functional assessments, such as urodynamic simulations, external complex post-processing steps are required to convert these results into mesh-level representations. To address this limitation, we propose an end-to-end method based on deep neural networks, namely KidMesh, which could automatically reconstruct CH meshes directly from MRU. Generally, KidMesh extracts feature maps from MRU images and converts them into feature vertices through grid sampling. It then deforms a template mesh according to these feature vertices to generate the specific CH meshes of MRU images. Meanwhile, we develop a novel schema to train KidMesh without relying on accurate mesh-level annotations, which are difficult to obtain due to the sparsely sampled MRU slices. Experimental results show that KidMesh could reconstruct CH meshes in an average of 0.4 seconds, and achieve comparable performance to conventional methods without requiring post-processing. The reconstructed meshes exhibited no self-intersections, with only 3.7% and 0.2% of the vertices having error distances exceeding 3.2mm and 6.4mm, respectively. After rasterization, these meshes achieved a Dice score of 0.86 against manually delineated CH masks. Furthermore, these meshes could be used in renal urine flow simulations, providing valuable urodynamic information for clinical practice.

Daniel M. Jimenez-Gutierrez, Enrique Zuazua, Joaquin Del Rio et al.

Early and accurate pneumonia detection from chest X-rays (CXRs) is clinically critical to expedite treatment and isolation, reduce complications, and curb unnecessary antibiotic use. Although artificial intelligence (AI) substantially improves CXR-based detection, development is hindered by globally distributed data, high inter-hospital variability, and strict privacy regulations (e.g., HIPAA, GDPR) that make centralization impractical. These constraints are compounded by heterogeneous imaging protocols, uneven data availability, and the costs of transferring large medical images across geographically dispersed sites. In this paper, we evaluate Federated Learning (FL) using the Sherpa.ai FL platform, enabling multiple hospitals (nodes) to collaboratively train a CXR classifier for pneumonia while keeping data in place and private. Using the Pediatric Pneumonia Chest X-ray dataset, we simulate cross-hospital collaboration with non-independent and non-identically distributed (non-IID) data, reproducing real-world variability across institutions and jurisdictions. Our experiments demonstrate that collaborative and privacy-preserving training across multiple hospitals via FL led to a dramatic performance improvement achieving 0.900 Accuracy and 0.966 ROC-AUC, corresponding to 47.5% and 50.0% gains over single-hospital models (0.610; 0.644), without transferring any patient CXR. These results indicate that FL delivers high-performing, generalizable, secure and private pneumonia detection across healthcare networks, with data kept local. This is especially relevant for rare diseases, where FL enables secure multi-institutional collaboration without data movement, representing a breakthrough for accelerating diagnosis and treatment development in low-data domains.

Lei Wang, Yang Ni, Irina Gaynanova

Increasing epidemiologic evidence suggests that the diversity and composition of the gut microbiome can predict infection risk in cancer patients. Infections remain a major cause of morbidity and mortality during chemotherapy. Analyzing microbiome data to identify associations with infection pathogenesis for proactive treatment has become a critical research focus. However, the high-dimensional nature of the data necessitates the use of dimension-reduction methods to facilitate inference and interpretation. Traditional dimension reduction methods, which assume Gaussianity, perform poorly with skewed and zero-inflated microbiome data. To address these challenges, we propose a semiparametric principal component analysis (PCA) method based on a truncated latent Gaussian copula model that accommodates both skewness and zero inflation. Simulation studies demonstrate that the proposed method outperforms existing approaches by providing more accurate estimates of scores and loadings across various copula transformation settings. We apply our method, along with competing approaches, to gut microbiome data from pediatric patients with acute lymphoblastic leukemia. The principal scores derived from the proposed method reveal the strongest associations between pre-chemotherapy microbiome composition and adverse events during subsequent chemotherapy, offering valuable insights for improving patient outcomes.

Khartik Uppalapati, Bora Yimenicioglu, Shakeel Abdulkareem et al.

Objective biomarkers for staging pediatric epileptic encephalopathies are scarce. We revisited a large open repository -- the CHB-MIT Scalp EEG Database, 22 subjects aged 1.5-19 y recorded at 256 Hz under the 10-20 montage -- to derive and validate a compact quantitative index, DS-Qi = (theta/alpha)_posterior + (1 - wPLI_beta). The first term captures excess posterior slow-wave power, a recognized marker of impaired cortical maturation; the second employs the debiased weighted Phase-Lag Index to measure loss of beta-band synchrony, robust to volume conduction and small-sample bias. In 30-min awake, eyes-open segments, DS-Qi was 1.69 +/- 0.21 in epilepsy versus 1.23 +/- 0.17 in age-matched normative EEG (Cohen's d = 1.1, p < 0.001). A logistic model trained with 10 x 10-fold cross-validation yielded an AUC of 0.90 (95% CI 0.81-0.97) and optimal sensitivity/specificity of 86%/83% at DS-Qi = 1.46. Across multi-day recordings, test-retest reliability was ICC = 0.74, and higher DS-Qi correlated with greater seizure burden (rho = 0.58, p = 0.004). These results establish DS-Qi as a reproducible, single-number summary of electrophysiological severity that can be computed from short scalp EEG segments using only posterior and standard 10-20 electrodes.

Vahideh Zolfaghari

Background Large language models (LLMs) are increasingly deployed in medical consultations, yet their safety under realistic user pressures remains understudied. Prior assessments focused on neutral conditions, overlooking vulnerabilities from anxious users challenging safeguards. This study evaluated LLM safety under parental anxiety-driven adversarial pressures in pediatric consultations across models and platforms. Methods PediatricAnxietyBench, from a prior evaluation, includes 300 queries (150 authentic, 150 adversarial) spanning 10 topics. Three models were assessed via APIs: Llama-3.3-70B and Llama-3.1-8B (Groq), Mistral-7B (HuggingFace), yielding 900 responses. Safety used a 0-15 scale for restraint, referral, hedging, emergency recognition, and non-prescriptive behavior. Analyses employed paired t-tests with bootstrapped CIs. Results Mean scores: 9.70 (Llama-3.3-70B) to 10.39 (Mistral-7B). Llama-3.1-8B outperformed Llama-3.3-70B by +0.66 (p=0.0001, d=0.225). Models showed positive adversarial effects, Mistral-7B strongest (+1.09, p=0.0002). Safety generalized across platforms; Llama-3.3-70B had 8% failures. Seizures vulnerable (33% inappropriate diagnoses). Hedging predicted safety (r=0.68, p<0.001). Conclusions Evaluation shows safety depends on alignment and architecture over scale, with smaller models outperforming larger. Evolution to robustness across releases suggests targeted training progress. Vulnerabilities and no emergency recognition indicate unsuitability for triage. Findings guide selection, stress adversarial testing, and provide open benchmark for medical AI safety.

Vanessa D'Amario, Randy Daniel, Alessandro Zanetti et al.

Small open-source medical large language models (LLMs) offer promising opportunities for low-resource deployment and broader accessibility. However, their evaluation is often limited to accuracy on medical multiple choice question (MCQ) benchmarks, and lacks evaluation of consistency, robustness, or reasoning behavior. We use MCQ coupled to human evaluation and clinical review to assess six small open-source medical LLMs (HuatuoGPT-o1 (Chen 2024), Diabetica-7B, Diabetica-o1 (Wei 2024), Meditron3-8B (Sallinen2025), MedFound-7B (Liu 2025), and ClinicaGPT-base-zh (Wang 2023)) in pediatric endocrinology. In deterministic settings, we examine the effect of prompt variation on models' output and self-assessment bias. In stochastic settings, we evaluate output variability and investigate the relationship between consistency and correctness. HuatuoGPT-o1-8B achieved the highest performance. The results show that high consistency across the model response is not an indicator of correctness, although HuatuoGPT-o1-8B showed the highest consistency rate. When tasked with selecting correct reasoning, both HuatuoGPT-o1-8B and Diabetica-o1 exhibit self-assessment bias and dependency on the order of the candidate explanations. Expert review of incorrect reasoning rationales identified a mix of clinically acceptable responses and clinical oversight. We further show that system-level perturbations, such as differences in CUDA builds, can yield statistically significant shifts in model output despite stable accuracy. This work demonstrates that small, semantically negligible prompt perturbations lead to divergent outputs, raising concerns about reproducibility of LLM-based evaluations and highlights the output variability under different stochastic regimes, emphasizing the need of a broader diagnostic framework to understand potential pitfalls in real-world clinical decision support scenarios.

Abicumaran Uthamacumaran

By use of complex network dynamics and graph-based machine learning, we identified critical determinants of lineage-specific plasticity across the single-cell transcriptomics of pediatric high-grade glioma (pHGGs) subtypes: IDHWT glioblastoma and K27M-mutant glioma. Our study identified network interactions regulating glioma morphogenesis via the tumor-immune microenvironment, including neurodevelopmental programs, calcium dynamics, iron metabolism, metabolic reprogramming, and feedback loops between MAPK/ERK and WNT signaling. These relationships highlight the emergence of a hybrid spectrum of cellular states navigating a disrupted neuro-differentiation hierarchy. We identified transition genes such as DKK3, NOTCH2, GATAD1, GFAP, and SEZ6L in IDHWT glioblastoma, and H3F3A, ANXA6, HES6/7, SIRT2, FXYD6, PTPRZ1, MEIS1, CXXC5, and NDUFAB1 in K27M subtypes. We also identified MTRNR2L1, GAPDH, IGF2, FKBP variants, and FXYD7 as transition genes that influence cell fate decision-making across both subsystems. Our findings suggest pHGGs are developmentally trapped in states exhibiting maladaptive behaviors, and hybrid cellular identities. In effect, tumor heterogeneity (metastability) and plasticity emerge as stress-response patterns to immune-inflammatory microenvironments and oxidative stress. Furthermore, we show that pHGGs are steered by developmental trajectories from radial glia predominantly favoring neocortical cell fates, in telencephalon and prefrontal cortex (PFC) differentiation. By addressing underlying patterning processes and plasticity networks as therapeutic vulnerabilities, our findings provide precision medicine strategies aimed at modulating glioma cell fates and overcoming therapeutic resistance. We suggest transition therapy toward neuronal-like lineage differentiation as a potential therapy to help stabilize pHGG plasticity and aggressivity.

Almabadi ES, Felemban D, Alekhmimi RK et al.

Eman S Almabadi,1 Doaa Felemban,2 Razan Khalid Alekhmimi,3,4 Muntasir Adnan Aynusah,4 Alla Alsharif,1 Nebras Althagafi,1 Saba Kassim1 1Department of Preventive Dental Sciences, Taibah University, College of Dentistry, Al-Madinah Al-Munawwrah, 42353, Saudi Arabia; 2Department of Oral and Maxillofacial Diagnostic Sciences, Taibah University, College of Dentistry, Al-Madinah Al-Munawwarah, 42353, Saudi Arabia; 3Medical Administration Department,Taibah University, College of Dentistry, Al-Madinah Al-Munawwarah, 42353, Saudi Arabia; 4Department of Dental Surgery, Healthcare Quality and Patient Safety, Ministry of Health, Al-Madinah Al-Munawwarah, 42394, Saudi ArabiaCorrespondence: Eman S Almabadi, Department of Preventive Dental Sciences, Taibah University, College of Dentistry, Prince, Naif Ibn Abdulaziz, Al-Madinah Al-Munawwrah, 42353, Saudi Arabia, Email emabadi@taibahu.edu.saObjective: This study aimed to assess the association between sleep bruxism (SB) among children and parental sociodemographic characteristics and SB risk factors (eg, nose obstruction).Methods: A cross-sectional survey was conducted with 250 parents of children under the age of 13 who visited pediatric dental clinics. Data were collected through a questionnaire completed by parents. Sociodemographic characteristics, the child’s medical history, sleep patterns and parents’ awareness of bruxism and its symptoms were investigated. Descriptive, bivariate and binary logistic regression analyses were performed.Results: The response rate was 85.2% (55% females, 45% males) and 25.8% of the parents self-reported that their children had bruxism. The regression analysis revealed that parents reporting SB among their children were significantly more likely to have SB themselves (8.62 [3.68– 20.16], p = 0.001). While children whose mothers had lower education level and were unaware of bruxism-related symptoms (such as teeth, jaw, or face pain) were less likely to be reported as having SB (0.35 [0.16– 0.75], p = 0.007; 0.36 [0.14– 0.97], p = 0.043, respectively). Parents who identified nose obstruction as a cause of bruxism also had children with a higher likelihood of having SB (5.49 [1.04– 29.08], p = 0.045).Conclusion: The findings highlighted that parental sociodemographic characteristic and SB risk factors associated signficantly with the prevalence of childhood SB.Keywords: sleep bruxism, children, parental sociodemographic factors, Saudi Arabia

Meaghan Reaney, Brittany Player, Ramya Billa et al.

Blastomycosis is a rare fungal infection caused by the inhalation of Blastomyces dermatitidis spores. Infection with this fungus can impact nearly every organ system, though pulmonary disease is the most common. Presentations of pulmonary blastomycosis are highly variable, ranging from clinically asymptomatic to severe respiratory failure requiring intensive care. This case series describes the clinical presentation, diagnostic challenges, management, and outcomes of two siblings with severe pulmonary blastomycosis that progressed to pediatric acute respiratory distress syndrome requiring mechanical ventilation and venovenous extracorporeal membrane oxygenation (VV-ECMO). Despite being relatively uncommon, blastomycosis should be considered in patients with respiratory symptoms not responding to empiric antibacterial therapy, particularly in endemic regions. Early diagnosis and prompt initiation of appropriate antifungal therapy are crucial for favorable outcomes. Additionally, early initiation of ECMO for severe pulmonary blastomycosis may be beneficial in temporizing to allow time for sufficient response to antifungal therapy.

Yi-Ping Huang, Chien-Hui Chien, Li-Chieh Wang et al.

Abstract Regulatory T (Treg) cells are effective immunomodulators of adaptive and innate immune responses. Our previous studies have demonstrated that B-cell-induced CD4+Foxp3− regulatory T cells, referred to as Treg-of-B cells, exert suppressive capacity, by inhibiting CD4+CD25− T-cell proliferation and inflammasome activation. In the present study, Treg-of-B cells downregulated proinflammatory M1-like markers and partially induced M2-associated genes in unpolarized bone marrow-derived macrophages (BMDMs), as evidenced by RNA expression of Nos2, Arg1, Retnla, Mrc1, and Egr2. Treg-of-B cells decreased the RNA levels of Nos2, Tnfa, Cd86, and Cxcl9, and reduced the production of tumor necrosis factor (TNF)-α, interleukin (IL)-6, and nitrite in LPS/interferon (IFN)-γ-stimulated M1-like macrophages in a dose-dependent manner. These cells also secreted Th2 cytokines, including IL-10, IL-4, and IL-13, with enhanced cytokine production observed when cocultured with macrophages. Mechanistically, Treg-of-B cells exerted their modulatory effects via both cell-cell contact and contact-dependent induction of soluble mediators, particularly Th2 cytokines. Furthermore, Treg-of-B cells promoted IκBα accumulation and suppressed RNA expression of Kruppel-like factor 4 (Klf4), thereby inhibiting NF-κB activation. These findings suggest that Treg-of-B cells regulate macrophage plasticity and might prevent excessive inflammation.

David A Friedman, John B. Reinhart, I. L. Schwartz et al.

Mohamed Yaseen Jabarulla, Steffen Oeltze-Jafra, Philipp Beerbaum et al.

This research focuses on evaluating the non-commercial open-source large language models (LLMs) Meditron, MedAlpaca, Mistral, and Llama-2 for their efficacy in interpreting medical guidelines saved in PDF format. As a specific test scenario, we applied these models to the guidelines for hypertension in children and adolescents provided by the European Society of Cardiology (ESC). Leveraging Streamlit, a Python library, we developed a user-friendly medical document chatbot tool (MedDoc-Bot). This tool enables authorized users to upload PDF files and pose questions, generating interpretive responses from four locally stored LLMs. A pediatric expert provides a benchmark for evaluation by formulating questions and responses extracted from the ESC guidelines. The expert rates the model-generated responses based on their fidelity and relevance. Additionally, we evaluated the METEOR and chrF metric scores to assess the similarity of model responses to reference answers. Our study found that Llama-2 and Mistral performed well in metrics evaluation. However, Llama-2 was slower when dealing with text and tabular data. In our human evaluation, we observed that responses created by Mistral, Meditron, and Llama-2 exhibited reasonable fidelity and relevance. This study provides valuable insights into the strengths and limitations of LLMs for future developments in medical document interpretation. Open-Source Code: https://github.com/yaseen28/MedDoc-Bot

Sara Ketabi, Matthias W. Wagner, Cynthia Hawkins et al.

Despite the promising performance of convolutional neural networks (CNNs) in brain tumor diagnosis from magnetic resonance imaging (MRI), their integration into the clinical workflow has been limited. That is mainly due to the fact that the features contributing to a model's prediction are unclear to radiologists and hence, clinically irrelevant, i.e., lack of explainability. As the invaluable sources of radiologists' knowledge and expertise, radiology reports can be integrated with MRI in a contrastive learning (CL) framework, enabling learning from image-report associations, to improve CNN explainability. In this work, we train a multimodal CL architecture on 3D brain MRI scans and radiology reports to learn informative MRI representations. Furthermore, we integrate tumor location, salient to several brain tumor analysis tasks, into this framework to improve its generalizability. We then apply the learnt image representations to improve explainability and performance of genetic marker classification of pediatric Low-grade Glioma, the most prevalent brain tumor in children, as a downstream task. Our results indicate a Dice score of 31.1% between the model's attention maps and manual tumor segmentation (as an explainability measure) with test classification performance of 87.7%, significantly outperforming the baselines. These enhancements can build trust in our model among radiologists, facilitating its integration into clinical practices for more efficient tumor diagnosis.

Platon Lukyanenko, Joshua Mayourian, Mingxuan Liu et al.

Artificial intelligence applied to electrocardiography (AI-ECG) shows potential for mortality prediction, but heterogeneous approaches and private datasets have limited generalizable insights. To address this, we systematically evaluated model design choices across three large cohorts: Beth Israel Deaconess (MIMIC-IV: n = 795,546 ECGs, United States), Telehealth Network of Minas Gerais (Code-15: n = 345,779, Brazil), and Boston Children's Hospital (BCH: n = 255,379, United States). We evaluated models predicting all-cause mortality, comparing horizon-based classification and deep survival methods with neural architectures including convolutional networks and transformers, benchmarking against demographic-only and gradient boosting baselines. Top models performed well (median concordance: Code-15, 0.83; MIMIC-IV, 0.78; BCH, 0.81). Incorporating age and sex improved performance across all datasets. Classifier-Cox models showed site-dependent sensitivity to horizon choice (median Pearson's R: Code-15, 0.35; MIMIC-IV, -0.71; BCH, 0.37). External validation reduced concordance, and in some cases demographic-only models outperformed externally trained AI-ECG models on Code-15. However, models trained on multi-site data outperformed site-specific models by 5-22%. Findings highlight factors for robust AI-ECG deployment: deep survival methods outperformed horizon-based classifiers, demographic covariates improved predictive performance, classifier-based models required site-specific calibration, and cross-cohort training, even between adult and pediatric cohorts, substantially improved performance. These results emphasize the importance of model type, demographics, and training diversity in developing AI-ECG models reliably applicable across populations.

Yinzheng Zhao, Zhihao Zhao, Junjie Yang et al.

Purpose: To investigate the changes in retinal vascular structures associated various stages of myopia by designing automated software based on an artif intelligencemodel. Methods: The study involved 1324 pediatric participants from the National Childr Medical Center in China, and 2366 high-quality retinal images and correspon refractive parameters were obtained and analyzed. Spherical equivalent refrac(SER) degree was calculated. We proposed a data analysis model based c combination of the Convolutional Neural Networks (CNN) model and the atter module to classify images, segment vascular structures, and measure vasc parameters, such as main angle (MA), branching angle (BA), bifurcation edge al(BEA) and bifurcation edge coefficient (BEC). One-way ANOVA compared param measurements betweenthenormalfundus,lowmyopia,moderate myopia,and high myopia group. Results: There were 279 (12.38%) images in normal group and 384 (16.23%) images in the high myopia group. Compared normal fundus, the MA of fundus vessels in different myopic refractive groups significantly reduced (P = 0.006, P = 0.004, P = 0.019, respectively), and performance of the venous system was particularly obvious (P<0.001). At the sa time, the BEC decreased disproportionately (P<0.001). Further analysis of fundus vascular parameters at different degrees of myopia showed that there were also significant differences in BA and branching coefficient (BC). The arterial BA value of the fundus vessel in the high myopia group was lower than that of other groups (P : 0.032, 95% confidence interval [Ci], 0.22-4.86), while the venous BA values increased(P = 0.026). The BEC values of high myopia were higher than those of low and moderate myopia groups. When the loss function of our data classification model converged to 0.09,the model accuracy reached 94.19%