Ophthalmology and Therapy, Год журнала: 2024, Номер 13(7), С. 2009 - 2022
Опубликована: Июнь 1, 2024
Previous studies have explored the connections between various ocular biological parameters with myopia. Our previous study also found that pupil data can predict myopic progression during interventions for However, exploring association diameter and myopia in preschoolers were lacking. Hence this was aimed to investigate based on a real-world, large-scale dataset.
Язык: Английский
Taiwan Journal of Ophthalmology, Год журнала: 2023, Номер 13(2), С. 142 - 150
Опубликована: Апрель 1, 2023
Myopia as an uncorrected visual impairment is recognized a global public health issue with increasing burden on health-care systems. Moreover, high myopia increases one's risk of developing pathologic myopia, which can lead to irreversible impairment. Thus, increased resources are needed for the early identification complications, timely intervention prevent progression, and treatment complications. Emerging artificial intelligence (AI) digital technologies may have potential tackle these unmet needs through automated detection screening stratification, individualized prediction, prognostication progression. AI applications in children adults been developed detection, diagnosis, prediction Novel technologies, including multimodal AI, explainable federated learning, machine blockchain, further improve performance, safety, accessibility, also circumvent concerns explainability. Digital technology advancements include therapeutics, self-monitoring devices, virtual reality or augmented technology, wearable devices - provide possible avenues monitoring progression control. However, there challenges implementation requirements specific infrastructure resources, demonstrating clinically acceptable performance safety data management. Nonetheless, this remains evolving field address growing myopia.
Язык: Английский
Процитировано
13
Anales de Pediatría, Год журнала: 2024, Номер 100(3), С. 195 - 201
Опубликована: Март 1, 2024
Se examina el uso de la inteligencia artificial (IA) en campo atención a salud pediátrica dentro del marco «Medicina las 7P» (Predictiva, Preventiva, Personalizada, Precisa, Participativa, Periférica y Poliprofesional). destacan diversas aplicaciones IA diagnóstico, tratamiento control enfermedades pediátricas, así como su papel prevención gestión eficiente los recursos médicos con repercusión sostenibilidad sistemas públicos salud. presentan casos éxito aplicación ámbito pediátrico se hace un gran énfasis necesidad caminar hacia Medicina 7P. La está revolucionando sociedad general ofreciendo potencial para mejorar significativamente cuidado pediatría. This article examines the use of intelligence (AI) in field paediatric care within framework 7P medicine model (Predictive, Preventive, Personalized, Precise, Participatory, Peripheral and Polyprofessional). It highlights various applications AI diagnosis, treatment management diseases as well role prevention efficient health resources resulting impact on sustainability public systems. Successful cases application setting are presented, placing emphasis need to move towards model. Artificial is revolutionizing society at large has great potential for significantly improving care.
Процитировано
4
Graefe s Archive for Clinical and Experimental Ophthalmology, Год журнала: 2023, Номер 262(1), С. 3 - 17
Опубликована: Май 25, 2023
Язык: Английский
Процитировано
9
Frontiers in Medicine, Год журнала: 2024, Номер 11
Опубликована: Окт. 11, 2024
Purpose To develop a predictive model for orthokeratology (Ortho-K) lens decentration 1 month after wear. Methods This study included myopic children who were fitted with Ortho-K lenses at Fujian Provincial Hospital between December 2022 and May 2024. Corneal topography parameters other relevant metrics collected pre- post-treatment. Feature selection was conducted using univariate logistic regression Lasso analysis. A machine learning approach used to multiple models, including Decision Tree, Logistic Regression, Multilayer Perceptron, Random Forest, Support Vector Machine. Model performance evaluated accuracy, sensitivity, specificity, ROC curves, DCA calibration curves. SHAP values employed interpret the models. Results The Regression demonstrated best performance, an AUC of 0.82 (95% CI: 0.69–0.95), accuracy 77.59%, sensitivity 85%, specificity 61.11%. most significant predictors identified age, 8 mm sag height difference, 5 Kx1, 7 Kx2. analysis confirmed importance these features, particularly difference. Conclusions successfully predicted risk key corneal morphological age. provides valuable support clinicians in optimizing fitting strategies, potentially reducing adverse outcomes improving quality vision patients. Further validation clinical settings is recommended.
Язык: Английский
Процитировано
3
Clinical Optometry, Год журнала: 2025, Номер Volume 17, С. 83 - 114
Опубликована: Март 1, 2025
Abstract: With the global shortage of eye care professionals and increasing burden vision impairment, particularly in low- middle-income countries, there is an urgent need for innovative solutions to bridge gaps services. Advances artificial intelligence (AI) over recent decades have significantly impacted healthcare, including field optometry. When integrated into optometric workflows, AI has potential streamline decision-making processes enhance system efficiency. To realize this potential, it essential develop models that can improve each stage patient workflow, screening, detection, diagnosis, management. This review explores application optometry, focusing on its optimize various aspects care. We examined across key areas Our analysis considered crucial parameters, model selection, sample sizes training validation, evaluation metrics, explainability models. comprehensive identified both strengths weaknesses existing The majority image-based studies utilized CNN or transfer learning models, while clinical data-based primarily employed RF, SVM, XGBoost. In general, trained large datasets achieved higher accuracy. However, many optometry-focused faced limitations due insufficient sizes— 28% were fewer than 500 samples, 18% used 200 half validated their with 38% validating 200. Additionally, some same data validation experienced overfitting, leading reduced Notably, 20% included reported accuracy below 80%, limiting practical applicability settings. provides optometrists valuable insights field, aiding informed implementation Keywords: intelligence,
Язык: Английский
Процитировано
0
Pediatric Investigation, Год журнала: 2025, Номер unknown
Опубликована: Март 18, 2025
ABSTRACT The global incidence of myopia is increasing, and high increases the risk pathological myopia, which can lead to irreversible visual impairment, posing a significant health concern. Artificial intelligence (AI) may be solution pandemic, with potential applications in early identification, stratification, progression prediction, timely intervention address unmet needs. AI has been developed detect, diagnose, predict both children adults. In this review, current state technology field comprehensively reviewed, challenges, development status, future directions have also discussed, hold great significance for further application management.
Язык: Английский
Процитировано
0
European Journal of Ophthalmology, Год журнала: 2025, Номер unknown
Опубликована: Апрель 13, 2025
Refractive error is among the leading causes of visual impairment globally. The diagnosis and management refractive has traditionally relied on comprehensive eye examinations by care professionals, but access to these specialized services remained limited in many areas world. Given this, artificial intelligence (AI) shown immense potential transforming error. We review AI applications across various aspects – from axial length prediction using fundus images risk stratification for myopia progression. algorithms can be trained analyze clinical data detect as well predict associated risks For treatments such implantable collamer orthokeratology lenses, models facilitate vault size optimal lens fitting with high accuracy. Furthermore, demonstrated promise optimizing surgical planning outcomes procedures. Emerging digital technologies telehealth, smartphone applications, virtual reality integrated present novel avenues screening. discuss key challenges, including validation datasets, lack standardization, image quality issues, population heterogeneity, practical deployment, ethical considerations regarding patient privacy that need addressed before widespread implementation.
Язык: Английский
Процитировано
0
Frontiers in Cell and Developmental Biology, Год журнала: 2023, Номер 11
Опубликована: Апрель 28, 2023
The rapid development of computer science over the past few decades has led to unprecedented progress in field artificial intelligence (AI). Its wide application ophthalmology, especially image processing and data analysis, is particularly extensive its performance excellent. In recent years, AI been increasingly applied optometry with remarkable results. This review a summary different models algorithms used (for problems such as myopia, strabismus, amblyopia, keratoconus, intraocular lens) includes discussion limitations challenges associated this field.
Язык: Английский
Процитировано
6
International Journal of Ophthalmology, Год журнала: 2023, Номер 16(9), С. 1406 - 1416
Опубликована: Авг. 22, 2023
With the rapid development of computer technology, application artificial intelligence (AI) to ophthalmology has gained prominence in modern medicine. As optometry is closely related ophthalmology, AI research on also increased. This review summarizes current and technologies used for diagnosis optometry, myopia, strabismus, amblyopia, optical glasses, contact lenses, other aspects. The aim identify mature models that are suitable potential algorithms may be future clinical practice.
Язык: Английский
Процитировано
5
Acta Ophthalmologica, Год журнала: 2024, Номер unknown
Опубликована: Март 22, 2024
Abstract Purpose To develop and validate an effective nomogram for predicting poor response to orthokeratology. Methods Myopic children (aged 8–15 years) treated with orthokeratology between February 2018 January 2022 were screened in four hospitals of different tiers (i.e. municipal provincial) China. Potential predictors included 32 baseline clinical variables. Nomogram the outcome (1‐year axial elongation ≥0.20 mm: response; <0.20 good response) was computed from a logistic regression model least absolute shrinkage selection operator. The data First Affiliated Hospital Chengdu Medical College randomly assigned (7:3) training validation cohorts. An external cohort three independent multicentre used test. Model performance assessed by discrimination (the area under curve, AUC), calibration (calibration plots) utility (decision curve analysis). Results Between March 2023, 1183 eligible subjects College, then divided into ( n = 831) 352) A total 405 cohort. Predictors age, spherical equivalent, length, pupil diameter, surface asymmetry index parental myopia p < 0.05). This demonstrated excellent calibration, net benefit discrimination, AUC 0.871 (95% CI 0.847–0.894), 0.863 (0.826–0.901) 0.817 (0.777–0.857) training, cohorts, respectively. online calculator generated free access http://39.96.75.172:8182/#/nomogram ). Conclusion provides accurate individual prediction overnight Chinese myopic children.
Язык: Английский
Процитировано
1