RESUMEN
Background: Ophthalmological screening for cytomegalovirus retinitis (CMVR) for HIV/AIDS patients is important to prevent lifelong blindness. Previous studies have shown good properties of automated CMVR screening using digital fundus images. However, the application of a deep learning (DL) system to CMVR with ultra-wide-field (UWF) fundus images has not been studied, and the feasibility and efficiency of this method are uncertain. Methods: In this study, we developed, internally validated, externally validated, and prospectively validated a DL system to detect AIDS-related from UWF fundus images from different clinical datasets. We independently used the InceptionResnetV2 network to develop and internally validate a DL system for identifying active CMVR, inactive CMVR, and non-CMVR in 6960 UWF fundus images from 862 AIDS patients and validated the system in a prospective and an external validation data set using the area under the curve (AUC), accuracy, sensitivity, and specificity. A heat map identified the most important area (lesions) used by the DL system for differentiating CMVR. Results: The DL system showed AUCs of 0.945 (95 % confidence interval [CI]: 0.929, 0.962), 0.964 (95 % CI: 0.870, 0.999) and 0.968 (95 % CI: 0.860, 1.000) for detecting active CMVR from non-CMVR and 0.923 (95 % CI: 0.908, 0.938), 0.902 (0.857, 0.948) and 0.884 (0.851, 0.917) for detecting active CMVR from non-CMVR in the internal cross-validation, external validation, and prospective validation, respectively. Deep learning performed promisingly in screening CMVR. It also showed the ability to differentiate active CMVR from non-CMVR and inactive CMVR as well as to identify active CMVR and inactive CMVR from non-CMVR (all AUCs in the three independent data sets >0.900). The heat maps successfully highlighted lesion locations. Conclusions: Our UWF fundus image-based DL system showed reliable performance for screening AIDS-related CMVR showing its potential for screening CMVR in HIV/AIDS patients, especially in the absence of ophthalmic resources.
RESUMEN
Purpose: Retinal and optic nerve diseases have become the primary cause of irreversible vision loss and blindness. However, there is still a lack of thorough evaluation regarding their prevalence in China. Methods: This artificial intelligence-based national screening study applied a previously developed deep learning algorithm, named the Retinal Artificial Intelligence Diagnosis System (RAIDS). De-identified personal medical records from January 2019 to December 2021 were extracted from 65 examination centers in 19 provinces of China. Crude prevalence and age-sex-adjusted prevalence were calculated by mapping to the standard population in the seventh national census. Results: In 2021, adjusted referral possible glaucoma (63.29, 95% confidence interval [CI] = 57.12-68.90 cases per 1000), epiretinal macular membrane (21.84, 95% CI = 15.64-29.22), age-related macular degeneration (13.93, 95% CI = 11.09-17.17), and diabetic retinopathy (11.33, 95% CI = 8.89-13.77) ranked the highest among 10 diseases. Female participants had significantly higher adjusted prevalence of pathologic myopia, yet a lower adjusted prevalence of diabetic retinopathy, referral possible glaucoma, and hypertensive retinopathy than male participants. From 2019 to 2021, the adjusted prevalence of retinal vein occlusion (0.99, 95% CI = 0.73-1.26 to 1.88, 95% CI = 1.42-2.44), macular hole (0.59, 95% CI = 0.41-0.82 to 1.12, 95% CI = 0.76-1.51), and hypertensive retinopathy (0.53, 95% CI = 0.40-0.67 to 0.77, 95% CI = 0.60-0.95) significantly increased. The prevalence of diabetic retinopathy in participants under 50 years old significant increased. Conclusions: Retinal and optic nerve diseases are an important public health concern in China. Further well-conceived epidemiological studies are required to validate the observed increased prevalence of diabetic retinopathy, hypertensive retinopathy, retinal vein occlusion, and macular hole nationwide. Translational Relevance: This artificial intelligence system can be a potential tool to monitor the prevalence of major retinal and optic nerve diseases over a wide geographic area.
Asunto(s)
Inteligencia Artificial , Enfermedades del Nervio Óptico , Enfermedades de la Retina , Humanos , China/epidemiología , Prevalencia , Masculino , Femenino , Persona de Mediana Edad , Adulto , Anciano , Enfermedades de la Retina/epidemiología , Enfermedades de la Retina/diagnóstico , Enfermedades del Nervio Óptico/epidemiología , Enfermedades del Nervio Óptico/diagnóstico , Adulto Joven , Adolescente , Tamizaje Masivo/métodos , Anciano de 80 o más AñosRESUMEN
PURPOSE: Intraocular schwannoma is a rare tumour, which is often misdiagnosed. We presented the demographics and clinical characteristics of patients with intraocular schwannoma. METHODS: Retrospective case series were collected between May 2005 and July 2021 in Beijing Tongren Hospital. RESULTS: A total of 28 patients were diagnosed with intraocular schwannoma on histopathological examination of surgical specimen. The median age was 39 years (range: 12-64). Fourteen patients were female and 14 were male. Among the all subjects, 21/28 patients (75.0%) presented as visual loss, and 3/28 patients (10.7%) had visual field loss. Intraocular schwannoma presented as nonpigmented mass in the ciliary body in 12/28 cases (42.9%), in the choroid in 9/28 cases (32.1%), and in ciliochoroid in 7/28 cases (25.0%). Intraocular schwannoma was often clinically misdiagnosed as uveal melanoma, which occurred in 16/28 patients (57.1%). Tumour excision with pars plana vitrectomy was performed for all included patients. Endoresection with lens removal was performed for tumours in the choroid, while transscleral resection was performed for tumours located in ciliary body or ciliochoroid. Increased light transmission was detected in 12/28 cases (42.9%). In the consecutive follow-up (median: 73 months, range: 7-193 months), no cases of recurrence or metastatic disease were detected. CONCLUSIONS: Intraocular schwannoma is a rare benign tumour. It usually presents as nonpigmented mass, which can easily be misdiagnosed as nonpigmented uveal melanoma.
Asunto(s)
Neurilemoma , Humanos , Neurilemoma/diagnóstico , Neurilemoma/patología , Neurilemoma/cirugía , Persona de Mediana Edad , Adulto , Masculino , Femenino , Estudios Retrospectivos , Adolescente , Niño , Adulto Joven , Agudeza Visual/fisiología , Neoplasias del Ojo/diagnóstico , Neoplasias del Ojo/cirugía , Neoplasias del Ojo/patología , Cuerpo Ciliar/patología , Cuerpo Ciliar/cirugía , Neoplasias de la Úvea/diagnóstico , Neoplasias de la Úvea/patología , Neoplasias de la Úvea/cirugía , Vitrectomía , Neoplasias de la Coroides/diagnóstico , Neoplasias de la Coroides/patología , Neoplasias de la Coroides/cirugíaRESUMEN
BACKGROUND: Analyzing fundus images with deep learning techniques is promising for screening systematic diseases. However, the quality of the rapidly increasing number of studies was variable and lacked systematic evaluation. OBJECTIVE: To systematically review all the articles that aimed to predict systemic parameters and conditions using fundus image and deep learning, assessing their performance, and providing suggestions that would enable translation into clinical practice. METHODS: Two major electronic databases (MEDLINE and EMBASE) were searched until August 22, 2023, with keywords 'deep learning' and 'fundus'. Studies using deep learning and fundus images to predict systematic parameters were included, and assessed in four aspects: study characteristics, transparent reporting, risk of bias, and clinical availability. Transparent reporting was assessed by the TRIPOD statement, while the risk of bias was assessed by PROBAST. RESULTS: 4969 articles were identified through systematic research. Thirty-one articles were included in the review. A variety of vascular and non-vascular diseases can be predicted by fundus images, including diabetes and related diseases (19%), sex (22%) and age (19%). Most of the studies focused on developed countries. The models' reporting was insufficient in determining sample size and missing data treatment according to the TRIPOD. Full access to datasets and code was also under-reported. 1/31(3.2%) study was classified as having a low risk of bias overall, whereas 30/31(96.8%) were classified as having a high risk of bias according to the PROBAST. 5/31(16.1%) of studies used prospective external validation cohorts. Only two (6.4%) described the study's calibration. The number of publications by year increased significantly from 2018 to 2023. However, only two models (6.5%) were applied to the device, and no model has been applied in clinical. CONCLUSION: Deep learning fundus images have shown great potential in predicting systematic conditions in clinical situations. Further work needs to be done to improve the methodology and clinical application.
Asunto(s)
Aprendizaje Profundo , Fondo de Ojo , Humanos , Enfermedades de la Retina/diagnóstico por imagen , Enfermedades de la Retina/diagnósticoRESUMEN
The label transition matrix has emerged as a widely accepted method for mitigating label noise in machine learning. In recent years, numerous studies have centered on leveraging deep neural networks to estimate the label transition matrix for individual instances within the context of instance-dependent noise. However, these methods suffer from low search efficiency due to the large space of feasible solutions. Behind this drawback, we have explored that the real murderer lies in the invalid class transitions, that is, the actual transition probability between certain classes is zero but is estimated to have a certain value. To mask the invalid class transitions, we introduced a human-cognition-assisted method with structural information from human cognition. Specifically, we introduce a structured transition matrix network (STMN) designed with an adversarial learning process to balance instance features and prior information from human cognition. The proposed method offers two advantages: 1) better estimation effectiveness is obtained by sparing the transition matrix and 2) better estimation accuracy is obtained with the assistance of human cognition. By exploiting these two advantages, our method parametrically estimates a sparse label transition matrix, effectively converting noisy labels into true labels. The efficiency and superiority of our proposed method are substantiated through comprehensive comparisons with state-of-the-art methods on three synthetic datasets and a real-world dataset. Our code will be available at https://github.com/WheatCao/STMN-Pytorch.
RESUMEN
BACKGROUND: To investigate the relationship between body weight and Axial length in guinea pigs. METHODS: Forty pigmented guinea pigs were randomly divided into two groups, namely control group and negative lens-induced myopization (LIM) group. After measuring the baseline axial length and body weight (BW), guinea pigs of LIM group received bilateral negative lens-induced myopization using - 10.0 diopters lenses. One week later, the lenses were removed and biometric and ophthalmoscopic examinations were repeated. RESULTS: Two groups of guinea pigs showed no statistical difference in initial body weight and eye axis length. Compared to the control group, the lens-induced group had a lower weight (P = 0.02) and a longer axial length (P < 0.01) at the end of study Neither at baseline nor at week 1 did AL correlate with BW in both groups (Control Baseline: r = 0.306, P = 0.19; Control Week1: r = 0.333, P = 0.15; LIM Baseline: r=-0.142, P = 0.55; LIM Week 1: r = 0.189, P = 0.42). Lens-induction had a significant effect on axial elongation (P < 0.01) while body weight had no impact on such aspect (P > 0.05). CONCLUSION: In guinea pigs of the same age, axial length was not correlated with body weight. Also, baseline body weight had no impact on natural axial length growth or lens-induced myopia. Lens-induction caused a significant reduction in body weight gain.
Asunto(s)
Cristalino , Miopía , Animales , Cobayas , Miopía/etiología , Longitud Axial del Ojo , Biometría , Modelos Animales de EnfermedadRESUMEN
This study aimed to examine the intraocular tolerability of the epidermal growth factor receptor antibody cetuximab, when applied intravitreally, and its effect on axial elongation. Guinea pigs aged 2-3 weeks were subjected to bilateral plano glasses and bilateral lens-induced myopization (LIM) as a single procedure for group I (n = 8) and group II (n = 8), respectively. In the animals of group III (n = 8), group IV (n = 8), and group V (n = 8), the right eyes of the animals, in addition to LIM, received four weekly intravitreal injections of cetuximab (Erbitux®) in doses of 6.25 µg, 12.5 µg, and 25 µg, respectively. As controls, the left eyes, in addition to LIM, received corresponding intraocular injections of phosphate-buffered saline. The animals underwent regular ophthalmoscopic examinations and biometry for axial length measurements. With increasing doses of cetuximab, the inter-eye difference in axial elongation (at study end, left eyes minus right eyes) were significantly the smallest in group I (0.00 ± 0.02 mm) and group II (-0.01 ± 0.02 mm), they were larger in group III (0.04 ± 0.04 mm) and group IV (0.10 ± 0.03 mm), and they were the largest in group V (0.11 ± 0.01 mm). The inter-eye difference in axial elongation enlarged (P < 0.001) with the number of injections applied. Retinal thickness at the posterior pole (right eyes) was significantly thicker in group V than in group II (P < 0.01). The density of apoptotic cells (visualized by TUNEL-staining) did not vary significantly between any of the groups (all P > 0.05). The results suggest that intravitreal injections of cetuximab in young guinea pigs with LIM resulted in a reduction in axial elongation in a dose-dependent and number of treatment-dependent manner. Intraocular toxic effects, such as intraocular inflammation, retinal thinning, or an increased density of apoptotic cells in the retina, were not observed in association with the intravitreally applied cetuximab.
Asunto(s)
Cristalino , Miopía , Cobayas , Animales , Miopía/metabolismo , Cetuximab/toxicidad , Cetuximab/metabolismo , Retina/metabolismo , Cristalino/metabolismo , Inyecciones Intraoculares , Modelos Animales de EnfermedadRESUMEN
Purpose: The mechanism underlying axial elongation during myopia progression remains unknown. Epidermal growth factor receptor (EGFR) signaling is associated with axial elongation. We explored whether mammalian target of rapamycin complex 1 (mTORC1) signaling acts as the downstream pathway of EGFR and participates in negative lens-induced axial elongation (NLIAE). Methods: Three-week-old male pigmented guinea pigs underwent binocular NLIAE. (1) To investigate whether EGFR is the upstream regulator of mTORC1, an EGFR inhibitor (20 µg erlotinib) was intravitreally injected once a week for three weeks. (2) To assess the effect of mTORC1 inhibition on NLIAE, an mTORC1 inhibitor (2 µg, 10 µg, and 20 µg everolimus) was intravitreally injected once a week for three weeks. (3) To explore the long-term effect of mTORC1 overactivation on axial elongation, an mTORC1 agonist (4 µg MHY1485) was intravitreally injected once a week for three months. Biometric measurements included axial length and choroidal thickness were performed. Results: Compared with the guinea pigs without NLIAE, NLIAE was associated with activation of mTORC1 signaling, which was suppressed by intravitreal erlotinib injection. Intravitreally injected everolimus suppressed NLIAE-induced axial elongation, mTORC1 activation, choroidal thinning, and hypoxia-inducible factor-1α expression in the sclera. Immunofluorescence revealed that the retinal pigment epithelium was the primary location of mTORC1 activation during NLIAE. Combining NLIAE and MHY1485 intravitreal injections significantly promoted axial elongation, choroidal thinning, and peripapillary choroidal atrophy. Conclusions: The mTORC1 signaling is associated with increased axial elongation, as in NLIAE, raising the possibility of inhibiting mTORC1 as a novel treatment for slowing myopia progression.
Asunto(s)
Everolimus , Miopía , Masculino , Animales , Cobayas , Clorhidrato de Erlotinib/uso terapéutico , Everolimus/farmacología , Everolimus/uso terapéutico , Miopía/tratamiento farmacológico , Inyecciones Intravítreas , Receptores ErbB , MamíferosRESUMEN
BACKGROUND: Retinoblastoma is the most common intraocular malignancy in childhood. With the advanced management strategy, the globe salvage and overall survival have significantly improved, which proposes subsequent challenges regarding long-term surveillance and offspring screening. This study aimed to apply a deep learning algorithm to reduce the burden of follow-up and offspring screening. METHODS: This cohort study includes retinoblastoma patients who visited Beijing Tongren Hospital from March 2018 to January 2022 for deep learning algorism development. Clinical-suspected and treated retinoblastoma patients from February 2022 to June 2022 were prospectively collected for prospective validation. Images from the posterior pole and peripheral retina were collected, and reference standards were made according to the consensus of the multidisciplinary management team. A deep learning algorithm was trained to identify "normal fundus", "stable retinoblastoma" in which specific treatment is not required, and "active retinoblastoma" in which specific treatment is required. The performance of each classifier included sensitivity, specificity, accuracy, and cost-utility. RESULTS: A total of 36,623 images were included for developing the Deep Learning Assistant for Retinoblastoma Monitoring (DLA-RB) algorithm. In internal fivefold cross-validation, DLA-RB achieved an area under curve (AUC) of 0.998 (95% confidence interval [CI] 0.986-1.000) in distinguishing normal fundus and active retinoblastoma, and 0.940 (95% CI 0.851-0.996) in distinguishing stable and active retinoblastoma. From February 2022 to June 2022, 139 eyes of 103 patients were prospectively collected. In identifying active retinoblastoma tumours from all clinical-suspected patients and active retinoblastoma from all treated retinoblastoma patients, the AUC of DLA-RB reached 0.991 (95% CI 0.970-1.000), and 0.962 (95% CI 0.915-1.000), respectively. The combination between ophthalmologists and DLA-RB significantly improved the accuracy of competent ophthalmologists and residents regarding both binary tasks. Cost-utility analysis revealed DLA-RB-based diagnosis mode is cost-effective in both retinoblastoma diagnosis and active retinoblastoma identification. CONCLUSIONS: DLA-RB achieved high accuracy and sensitivity in identifying active retinoblastoma from the normal and stable retinoblastoma fundus. It can be used to surveil the activity of retinoblastoma during follow-up and screen high-risk offspring. Compared with referral procedures to ophthalmologic centres, DLA-RB-based screening and surveillance is cost-effective and can be incorporated within telemedicine programs. CLINICAL TRIAL REGISTRATION: This study was registered on ClinicalTrials.gov (NCT05308043).
Asunto(s)
Aprendizaje Profundo , Neoplasias de la Retina , Retinoblastoma , Humanos , Retinoblastoma/diagnóstico , Estudios de Cohortes , Algoritmos , Estudios Retrospectivos , Neoplasias de la Retina/diagnósticoRESUMEN
PURPOSE: Eyelid sebaceous gland carcinoma (SGC) is a malignancy with fatal risk, high recurrence rate, and pagetoid spread. Thus, recurrence risk prediction and prompt treatment are extremely important. This study aimed to develop a nomogram to predict SGC recurrence based on potential risk factors. METHODS: We conducted a retrospective study to train and test a nomogram based on the clinical data of 391 patients across our hospital (304) and other grass-roots hospitals (87). After Cox regression, predictors included in the nomogram were selected, and sensitivity, specificity, concordance index (C-index), etc., were calculated to test their discrimination ability. RESULTS: After a median follow-up period of 4.12 years, SGC recurred in 52 (17.11%) patients. The 1-, 2-, and 5-year recurrence-free survival rates were 88.3%, 85.4%, and 81.6%, respectively. We examined five risk factors, such as lymph node metastasis at initial diagnosis (hazard ratio [HR], 2.260; 95% confidence interval [CI], 1.021-5.007), Ki67 (HR, 1.036; 95% CI, 1.020-1.052), histology differentiation degree (HR, 2.274; 95% CI, 1.063-4.865), conjunctival pagetoid infiltration (HR, 2.100; 95% CI, 1.0058-4.167), and orbital involvement (HR, 4.764; 95% CI, 1.436-15.803). The model had good discrimination in both internal and external test sets. The model had good discrimination in both internal and external test sets. The sensitivity of the internal test and external test set were 0.722 and 0.806, respectively, and specificity of the internal test and external test set were 0.886 and 0.893, respectively. CONCLUSION: We examined the potential risk factors for eyelid SGC recurrence and constructed a nomogram, which complements the TNM system in terms of prediction, indicating that our nomogram has the potential to reach clinical significance. This nomogram has the potential to assist healthcare practitioners in promptly detecting patients who are at an elevated risk and in tailoring clinical interventions to meet their individualized needs.
Asunto(s)
Carcinoma , Neoplasias de los Párpados , Humanos , Nomogramas , Estudios Retrospectivos , Glándulas Sebáceas/patología , Neoplasias de los Párpados/diagnóstico , Neoplasias de los Párpados/patología , Factores de Riesgo , Carcinoma/patología , Párpados/patologíaRESUMEN
Background: Epidermal growth factor (EGF) and its family members have been reported to be involved in myopic axial elongation. We examined whether short hairpin RNA attenuated adeno-associated virus (shRNA-AAV)-induced knockdown of amphiregulin, an EGF family member, has an influence on axial elongation. Methods: Three-week-old pigmented guinea pigs underwent lens-induced myopization (LIM) without additional intervention (LIM group; n = 10 animals) or additionally received into their right eyes at baseline an intravitreal injection of scramble shRNA-AAV (5 × 1010 vector genome [vg]) (LIM + Scr-shRNA group; n = 10) or of amphiregulin (AR)-shRNA-AAV (5 × 1010 vg/5 µL) (LIM + AR-shRNA-AAV group; n = 10), or they received an injection of AR-shRNA-AAV at baseline and three weekly amphiregulin injections (20 ng/5 µL) (LIM + AR-shRNA-AAV + AR group; n = 10). The left eyes received equivalent intravitreal injections of phosphate-buffered saline. Four weeks after baseline, the animals were sacrificed. Results: At study end, interocular axial length difference was higher (P < 0.001), choroid and retina were thicker (P < 0.05), and relative expression of amphiregulin and p-PI3K, p-p70S6K, and p-ERK1/2 was lower (P < 0.05) in the LIM + AR-shRNA-AAV group than in any other group. The other groups did not differ significantly when compared with each other. In the LIM + AR-shRNA-AAV group, the interocular axial length difference increased with longer study duration. TUNEL assay did not reveal significant differences among all groups in retinal apoptotic cell density. In vitro retinal pigment epithelium cell proliferation and migration were the lowest (P < 0.05) in the LIM + AR-shRNA-AAV group, followed by the LIM + AR-shRNA-AAV + AR group. Conclusions: shRNA-AAV-induced knockdown of amphiregulin expression, in association with suppression of epidermal growth factor receptor signaling, attenuated axial elongation in guinea pigs with LIM. The finding supports the notion of EGF playing a role in axial elongation.
Asunto(s)
Dependovirus , Miopía , Animales , Cobayas , Dependovirus/genética , Anfirregulina/metabolismo , Factor de Crecimiento Epidérmico , ARN Interferente Pequeño/genética , Miopía/metabolismo , Retina/metabolismoRESUMEN
PURPOSE: Studies have indicated that the observed association between vitamin D and myopia was confounded by time spent outdoors. This study aimed to elucidate this association using a national cross-sectional dataset. METHODS: Participants with 12 to 25 years who participated in non-cycloplegic vision exam from National Health and Nutrition Examination Survey (NHANES) 2001 to 2008 were included in the present study. Myopia was defined as spherical equivalent of any eyes ≤ -0.5 diopters (D). RESULTS: 7,657 participants were included. The weighted proportion of emmetropes, mild myopia, moderate myopia, and high myopia were 45.5%, 39.1%, 11.6%, and 3.8%, respectively. After adjusting for age, gender, ethnicity, TV/computer usage, and stratified by education attainment, every 10â nmol/L increment of serum 25(OH)D concentration was associated with a reduced risk of myopia (odds ratio [OR] = 0.96, 95% confidence interval [95%CI] 0.93-0.99 for any myopia; OR = 0.96, 95%CI 0.93-1.00 for mild myopia; OR = 0.99, 95%CI 0.97-1.01 for moderate myopia; OR = 0.89, 95%CI 0.84-0.95 for high myopia). Serum 25(OH)D level was closely correlated with time spent outdoors. After categorizing time spent outdoors into quarters (low, low-medium, medium-high, and high), every 1 quarter increment of time spent outdoors was associated with 2.49â nmol/L higher serum 25(OH)D concentration. After adjusting for time spent outdoors, serum 25(OH)D level did not show significant association with myopia (OR = 1.01, 95%CI 0.94-1.06 for 10â nmol/L increment). CONCLUSIONS: The association between high serum vitamin D and reduced risk of myopia is confounded by longer time spent outdoors. Evidence from the present study does not support that there is a direct association between serum vitamin D level with myopia.
Asunto(s)
Miopía , Vitamina D , Humanos , Adolescente , Adulto Joven , Estudios Transversales , Encuestas Nutricionales , Vitaminas , Miopía/epidemiologíaRESUMEN
OBJECTIVES: Major ocular diseases share common risk factors and pathogeneses with stroke. This study aimed to evaluate the relation between stroke and ocular diseases including visual impairment (VI). METHODS: The cross-sectional study investigated the prevalence and associations of VI and major eye diseases with stroke among 4570 participants in the 2005-2008 National Health and Nutrition Examination Survey (NHANES). The association of VI and major ocular diseases with stroke were estimated using univariate and multivariate logistic regression crude models and models adjusted for demographics and clinical factors. We also conducted stratified analyses by diabetes and hypertension status. RESULTS: VI was associated with stroke, and the odds ratios (ORs) for mild and moderate and severe visual impairment (MSVI) were 6.79 (95% confidence interval (CI): 2.44-18.88) and 9.46 (95% CI: 2.19-40.94) after adjusting for age and gender (all P < 0.05). Ocular disease was associated with stroke with OR reaching 5.54 (95% CI: 1.83-16.74), and the OR was 9.61 (95% CI: 3.05-30.23) for stroke patients suffering DR after adjusting for age and gender (all P < 0.05). After multivariable adjustment, the associations were limited to mild VI (OR = 10.00, 95% CI: 3.16-30.58), MSVI (OR = 8.57, 95% CI: 1.58-43.36), and any ocular disease (OR = 5.18, 95% CI: 1.46-18.42) (all P < 0.05). Significant associations between stroke and any ocular disease and DR were observed among diabetic participants, and significant relation between stroke and MSVI was found among hypertension patients. CONCLUSIONS: The sample of the US population demonstrates significant associations between VI and major ocular disease with stroke.
Asunto(s)
Hipertensión , Baja Visión , Humanos , Encuestas Nutricionales , Estudios Transversales , Trastornos de la Visión/etiología , Baja Visión/epidemiología , Factores de Riesgo , Hipertensión/complicaciones , Hipertensión/epidemiología , PrevalenciaRESUMEN
PURPOSES: Many factors were reported to be associated with diabetic retinopathy (DR); however, their contributions remained unclear. We aimed to evaluate the prognostic and diagnostic accuracy of logistic regression and three machine learning models based on various medical records. METHODS: This was a cross-sectional study. We investigated the prevalence and associations of DR among 757 participants aged 40 years or older in the 2005-2006 National Health and Nutrition Examination Survey (NHANES). We trained the models to predict if the participants had DR with 15 predictor variables. Area under the receiver operating characteristic (AUROC) and mean squared error (MSE) of each algorithm were compared in the external validation dataset using a replicate cohort from NHANES 2007-2008. RESULTS: Among the 757 participants, 53 (7.00%) subjects had DR, the mean (standard deviation, SD) age was 57.7 (13.04), and 78.0% were male (n = 42). Logistic regression revealed that female gender (OR = 4.130, 95% CI: 1.820-9.380; P < 0.05), HbA1c (OR = 1.665, 95% CI: 1.197-2.317; P < 0.05), serum creatine level (OR = 2.952, 95% CI: 1.274-6.851; P < 0.05), and eGFR level (OR = 1.009, 95% CI: 1.000-1.014, P < 0.05) increased the risk of DR. The average performance obtained from internal validation was similar in all models (AUROC ≥ 0.945), and k-nearest neighbors (KNN) had the highest value with an AUROC of 0.984. In external validation, they remained robust or with modest reductions in discrimination with AUROC still ≥ 0.902, and KNN also performed the best with an AUROC of 0.982. Both logistic regression and machine learning models had good performance in the clinical diagnosis of DR. CONCLUSIONS: This study highlights the utility of comparing traditional logistic regression to machine learning models. We found that logistic regression performed as well as optimized machine learning methods when classifying DR patients.
Asunto(s)
Diabetes Mellitus , Retinopatía Diabética , Humanos , Masculino , Femenino , Retinopatía Diabética/diagnóstico , Retinopatía Diabética/epidemiología , Encuestas Nutricionales , Modelos Logísticos , Estudios Transversales , Aprendizaje Automático , Registros MédicosRESUMEN
BACKGROUND: High myopia-related complications have become a major cause of irreversible vision loss. Evaluating the association between potential factors and high myopia can provide insights into pathophysiologic mechanisms and further intervention targets for myopia progression. METHOD: Participants aged 12-25 years from National Health and Nutrition Examination Survey 2001-2006 were selected for the analysis. Myopia was defined as spherical equivalent (sum of spherical error and half of the cylindrical error) of any eyes ≤-0.5 diopters. High myopia was defined as the spherical equivalent of any eye ≤ - 5.00 diopters. Essential variables were selected by Random Forest algorithm and verified by multivariable logistic regression. RESULTS: A total of 7,033 participants and 74 potential factors, including demographic (4 factors), physical examination (6 factors), nutritional and serological (45 factors), immunological (9 variables), and past medical history factors (10 factors), were included into the analysis. Random Forest algorithm found that several anthropometric, nutritional, and serological factors were associated with high myopia. Combined with multivariable logistic regression, high levels of serum vitamin A was significantly associated with an increased prevalence of high myopia (adjusted odd ratio = 1.46 for 1 µmol/L increment, 95% confidence interval [CI] 1.01-2.10). Furthermore, we found that neither C-reactive protein nor asthma increased the risk and severity of myopia. CONCLUSION: High levels of serum vitamin A was seemingly associated with an increased prevalence of high myopia. This borderline significant association should be interpreted with caution because the potential increased type I error after the multiple testing. It still needs further investigation regarding the mechanism underlying this association. Neither C-reactive protein nor asthma increased the risk and severity of myopia.
Asunto(s)
Asma , Miopía , Asma/diagnóstico , Asma/epidemiología , Proteína C-Reactiva , Humanos , Aprendizaje Automático , Miopía/diagnóstico , Miopía/epidemiología , Encuestas Nutricionales , Prevalencia , Factores de Riesgo , Vitamina ARESUMEN
Background: Diabetes significantly increases the risk of postoperative macular edema (PME) after cataract surgery, leading to potential worst post-operative outcomes. This study aims to compare the effect of different prophylactic interventions in improving postoperative anatomic and visual acuity outcomes of diabetes patients who underwent cataract surgery. Methods: We searched MEDLINE, Embase, Web of Science databases from inception until February 2nd, 2022, for studies including studies reporting PME events and/or best-corrected visual acuity (BCVA) outcomes. Random-effects Bayesian network meta-analysis was performed to compare the efficiency of intravitreal anti-vascular endothelial growth factor injections (anti-VEGF), nonsteroidal anti-inflammatory drugs (NSAIDs) and topical steroids eye drop at 1 week, 1 month, 3 months, 6 months after cataract surgery. Findings: The total of 2566 participants from 17 randomized controlled trials were included in the network meta-analysis, with moderate risk of bias and no evidence of publication of bias. Compared to placebo/steroid eye drop alone, patients received additional topical NSAIDs or intravitreal anti-VEGF injections had lower risk of PME at 1 month (NSAIDs: OR=0·221, 95% Confidence interval [CI], 0·044-0·755, I2 =0·0%, 5 studies; anti-VEGF: OR=0·151, 95%CI, 0·037-0·413, I2 =0·0%, 5 studies) and 3 month (NSAIDs: OR=0·370, 95%CI, 0·140-0·875, I2 =0·0%, 8 studies; anti-VEGF: OR=0·203, 95%CI, 0·101-0·353, I2 =0·0%, 4 studies) after cataract surgery. Further, additional anti-VEGF exhibited better BCVA outcome at 1 month (mean difference of LogMAR: -0·083, 95%CI, -0·17 to -0·014, I2 =62·0%, 5 studies), and 3 months (mean difference of LogMAR: -0·061, 95%CI, -0·11 to -0·011, I2 =0·0%, 5 studies) after cataract surgery. Such additional benefits did not reach statistic significant at 6 months after surgery. Interpretation: Our data suggests that compared to placebo/steroid eye drop alone, additional prophylactic anti-VEGF intervention could be considered for preventing the occurrence of PME after cataract surgery in patients with diabetes. Funding: Research and Development of Special (2020-1-2052); Science & Technology Project of Beijing Municipal Science & Technology Commission (Z201100005520045, Z181100001818003).
RESUMEN
Importance: The lack of experienced ophthalmologists limits the early diagnosis of retinal diseases. Artificial intelligence can be an efficient real-time way for screening retinal diseases. Objective: To develop and prospectively validate a deep learning (DL) algorithm that, based on ocular fundus images, recognizes numerous retinal diseases simultaneously in clinical practice. Design, Setting, and Participants: This multicenter, diagnostic study at 65 public medical screening centers and hospitals in 19 Chinese provinces included individuals attending annual routine medical examinations and participants of population-based and community-based studies. Exposures: Based on 120â¯002 ocular fundus photographs, the Retinal Artificial Intelligence Diagnosis System (RAIDS) was developed to identify 10 retinal diseases. RAIDS was validated in a prospective collected data set, and the performance between RAIDS and ophthalmologists was compared in the data sets of the population-based Beijing Eye Study and the community-based Kailuan Eye Study. Main Outcomes and Measures: The performance of each classifier included sensitivity, specificity, accuracy, F1 score, and Cohen κ score. Results: In the prospective validation data set of 208â¯758 images collected from 110â¯784 individuals (median [range] age, 42 [8-87] years; 115â¯443 [55.3%] female), RAIDS achieved a sensitivity of 89.8% (95% CI, 89.5%-90.1%) to detect any of 10 retinal diseases. RAIDS differentiated 10 retinal diseases with accuracies ranging from 95.3% to 99.9%, without marked differences between medical screening centers and geographical regions in China. Compared with retinal specialists, RAIDS achieved a higher sensitivity for detection of any retinal abnormality (RAIDS, 91.7% [95% CI, 90.6%-92.8%]; certified ophthalmologists, 83.7% [95% CI, 82.1%-85.1%]; junior retinal specialists, 86.4% [95% CI, 84.9%-87.7%]; and senior retinal specialists, 88.5% [95% CI, 87.1%-89.8%]). RAIDS reached a superior or similar diagnostic sensitivity compared with senior retinal specialists in the detection of 7 of 10 retinal diseases (ie, referral diabetic retinopathy, referral possible glaucoma, macular hole, epiretinal macular membrane, hypertensive retinopathy, myelinated fibers, and retinitis pigmentosa). It achieved a performance comparable with the performance by certified ophthalmologists in 2 diseases (ie, age-related macular degeneration and retinal vein occlusion). Compared with ophthalmologists, RAIDS needed 96% to 97% less time for the image assessment. Conclusions and Relevance: In this diagnostic study, the DL system was associated with accurately distinguishing 10 retinal diseases in real time. This technology may help overcome the lack of experienced ophthalmologists in underdeveloped areas.
Asunto(s)
Retinopatía Diabética , Enfermedades del Nervio Óptico , Enfermedades de la Retina , Adulto , Inteligencia Artificial , Retinopatía Diabética/diagnóstico , Femenino , Humanos , Masculino , Retina/diagnóstico por imagen , Enfermedades de la Retina/diagnóstico por imagenRESUMEN
For uveal melanoma (UM) patients, it is significant to establish diagnosis and prognosis evaluation systems through imaging techniques. However, imaging examinations are short of quantitative biomarkers and it is difficult to finish early diagnosis of UM. In order to discover new molecular biomarkers for the diagnosis and prognostic evaluation of UM, six circulating miRNAs (mir-132-3p, mir-21-5p, mir-34a-5p, mir-126-3p, mir-199a-3p, mir-214-3p) were chosen as candidates for independent validation. Validation of these miRNAs was performed in a cohort of 20 patients, including 10 spindle-shaped melanoma and 10 epithelioid cell melanoma, and 10 healthy donors. Then 5 patients with metastatic UM were included to validate the performance of miRNAs in advanced UM. Serum levels of miRNAs were determined using quantitative real-time PCR. We confirmed significantly higher levels of three miRNAs in serum of UM patients in comparison to healthy controls, and miR-199a-3p had the best performance (p < 0.0001; AUC = 0.985). MiR-214-3p and miR-21-5p were significantly upregulated in serum of epithelioid cell melanoma patients compared to spindle-shaped melanoma patients and miR-132-3p and, conversely, were significantly downregulated in serum of epithelioid cell melanoma patients. MiR-21-5p shows their best performance (p < 0.0001; AUC = 0.980). Both miR-199a-3p and miR-21-5p showed great performance in advanced UM. Significantly higher levels of miR-21-5p (p < 0.001) were found in serum of metastatic UM patients compared to patients with localized spindle-shaped melanoma, and significantly higher levels of miR-199a-3p (p < 0.001) were detected in serum of metastatic UM patients compared to healthy controls. Our preliminary data indicate promising diagnostic utility of circulating miR-199a-3p and promising prognostic utility of circulating miR-21-5p in both early and advanced UM patients.
RESUMEN
Background: Cause-specific prevalence data of vision loss and blindness is fundamental for making public health policies and is essential for prioritizing scientific advances and industry research. Methods: Cause-specific vision loss data from the Global Health Data Exchange was used. The burden of vision loss was measured by prevalence and years lived with disability (YLDs). Findings: In 2019, uncorrected refractory error and cataract were the most common causes for vision loss and blindness globally. Women have higher rates of cataract, age-related macular degeneration (AMD), and diabetic retinopathy (DR) than men. In the past 30 years, the prevalence of moderate/severe vision loss and blindness due to neonatal disorders has increased by 13.73 and 33.53%, respectively. Retinopathy of prematurity (ROP) is the major cause of neonatal disorders related vision loss. In 2019, ROP caused 101.6 thousand [95% uncertainty intervals (UI) 77.5-128.2] cases of vision impairment, including 49.1 thousand (95% UI 28.1-75.1) moderate vision loss, 27.5 thousand (95% UI 19.3-36.60) severe vision loss and, 25.0 thousand (95% UI 14.6-35.8) blindness. The prevalence of new-onset ROP in Africa and East Asia was significantly higher than other regions. Variation of preterm birth prevalence can explain 49.8% geometry variation of ROP-related vision loss burden among 204 countries and territories. After adjusting for preterm prevalence, government health spending per total health spending (%), rather than total health spending per person, was associated with a reduced burden of ROP-related vision loss in 2019 (-0.19 YLDs for 10% increment). By 2050, prevalence of moderate, severe vision loss and blindness due to ROP is expected to reach 43.6 (95% UI 35.1-52.0), 23.2 (95% UI 19.4-27.1), 31.9 (95% UI 29.7-34.1) per 100,000 population. Conclusion: The global burden of vision loss and blindness highlights the prevalent of ROP, a major and avoidable cause for childhood vision loss. Advanced screening techniques and treatments have shown to be effective in preventing ROP-related vision loss and are urgently needed in regions with high ROP-related blindness rates, including Africa and East Asia.
RESUMEN
BACKGROUND: Uveal melanoma (UM) is the most common primary intraocular malignancy in adults. Plaque brachytherapy (PRT) is widely accepted as an effective globe-conserving treatment modality for UM. However, local treatment failure and complications lead to the enucleation of irradiated eyes. We conducted this study to explore the causes and long-term prognosis for UM patients who accepted secondary enucleation after plaque radiotherapy. METHODS: This was a retrospective cohort study. Data of patients who underwent secondary enucleation for UM after plaque radiotherapy, from July 2007 to July 2019, at Beijing Tongren Hospital were analyzed. Kaplan-Meier analysis was performed to assess the probability of indications, metastasis, and metastasis-related death. Cox regression analysis was used to analyze associations of the prognostic factors. RESULTS: Eight hundred and eighty patients were clinically diagnosed with uveal melanoma and initially treated by iodine-125 plaque radiotherapy, 132 of whom underwent secondary enucleation and pathological examination in the same hospital. Fifty-two (39.4%) eyes were enucleated simply because of uncontrollable neovascular glaucoma (NVG). Forty-four (33.3%) patients suffered from tumor recurrence. Tumor non-response occurred in 18 (13.6%) cases. Ten (7.6%) eyes received enucleation entirely due to other types of glaucoma. Failure to preserve the eyes for other reasons occurred in eight (6.1%) patients. At a median follow-up of 58.1 [IQR: 40.9-90.5] months, the systemic spread was detected in 45 (34.1%) patients, and 38 of them died. On multivariate analysis, tumor largest basal diameter (HR 1.15 [95% CI: 1.01, 1.31]), tumor non-response (HR 7.22 [95% CI: 2.63, 19.82]), and recurrence (HR 3.29 [95% CI: 1.54, 7.07]) were risk factors for metastasis. Increased age (HR 1.54 [95% CI: 1.07, 2.23]), tumor non-response (HR 7.91 [95% CI: 2.79, 22.48]), and recurrence (HR 3.08 [95% CI: 1.13, 7.23]) were risk factors for metastasis-related death. CONCLUSIONS: NVG was the major reason for secondary enucleation for Chinese UM patients after PRT. Tumor non-response and recurrence were associated with a significantly higher risk of long-term metastasis and metastasis-related death.