Ophthalmology. Retina · 2025
Foveal Avascular Zone Enlargement Significantly Improves Statistical Models for Detection of Diabetic Retinopathy Severity.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Busza Anna M, Duffy Brandon V, Kakihara Shinji, Fawzi Amani A
Аннотация
PURPOSE: To determine whether combining OCT angiography (OCTA) metrics and foveal avascular zone (FAZ) metrics can optimize the noninvasive detection of high-risk diabetic retinopathy (DR). We hypothesized that these statistical models could be improved by incorporating the FAZ enlargement metric.
DESIGN: Cross-sectional study.
SUBJECTS: Patients with diabetes mellitus and a wide range of DR severity.
METHODS: A total of 264 eyes from 197 subjects underwent averaged en face OCT and OCTA to measure structural and functional FAZ, vessel density (VD) and geometric perfusion deficit (GPD) in the superficial capillary plexus (SCP) and deep capillary plexus (DCP). Foveal avascular zone area measured by OCTA-the functional FAZ-identifies the perfused capillaries around the fovea, whereas en face OCT visualizes the capillary tubes surrounding the fovea regardless of their perfusion status-the structural FAZ. Foveal avascular zone enlargement was thus defined as the difference between functional and structural FAZ. One-way analysis of variance quantified relationships between OCTA parameters and FAZ enlargement with DR severity. The imaging variables were further assessed by model iterations of FAZ and OCTA parameters, as well as univariate and multivariable binary logistic regression with backward elimination to determine the simplest (least number of parameters), highest-performing model for detecting referable DR or proliferative DR (PDR) based on the quasi-likelihood information criterion (QIC) and area under the curve (AUC).
MAIN OUTCOME MEASURES: Optimized model performance for detecting referable DR and PDR.
RESULTS: The optimal model for detecting referable DR included only 2 parameters: DCP GPD and FAZ enlargement (AUC: 0.91, QIC: 207.8). For detecting PDR within referable DR, the best model included DCP VD, SCP GPD, and FAZ enlargement (AUC: 0.82, QIC: 184.7). Foveal avascular zone enlargement contributed to all top-performing models, whereas functional FAZ was not required.
CONCLUSIONS: Statistical models that included 2 to 3 OCTA parameters were able to detect referable DR and PDR. The addition of FAZ enlargement in these models consistently improved their performance, offering the potential for personalized risk stratification in patients with diabetes using OCTA.
FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
DESIGN: Cross-sectional study.
SUBJECTS: Patients with diabetes mellitus and a wide range of DR severity.
METHODS: A total of 264 eyes from 197 subjects underwent averaged en face OCT and OCTA to measure structural and functional FAZ, vessel density (VD) and geometric perfusion deficit (GPD) in the superficial capillary plexus (SCP) and deep capillary plexus (DCP). Foveal avascular zone area measured by OCTA-the functional FAZ-identifies the perfused capillaries around the fovea, whereas en face OCT visualizes the capillary tubes surrounding the fovea regardless of their perfusion status-the structural FAZ. Foveal avascular zone enlargement was thus defined as the difference between functional and structural FAZ. One-way analysis of variance quantified relationships between OCTA parameters and FAZ enlargement with DR severity. The imaging variables were further assessed by model iterations of FAZ and OCTA parameters, as well as univariate and multivariable binary logistic regression with backward elimination to determine the simplest (least number of parameters), highest-performing model for detecting referable DR or proliferative DR (PDR) based on the quasi-likelihood information criterion (QIC) and area under the curve (AUC).
MAIN OUTCOME MEASURES: Optimized model performance for detecting referable DR and PDR.
RESULTS: The optimal model for detecting referable DR included only 2 parameters: DCP GPD and FAZ enlargement (AUC: 0.91, QIC: 207.8). For detecting PDR within referable DR, the best model included DCP VD, SCP GPD, and FAZ enlargement (AUC: 0.82, QIC: 184.7). Foveal avascular zone enlargement contributed to all top-performing models, whereas functional FAZ was not required.
CONCLUSIONS: Statistical models that included 2 to 3 OCTA parameters were able to detect referable DR and PDR. The addition of FAZ enlargement in these models consistently improved their performance, offering the potential for personalized risk stratification in patients with diabetes using OCTA.
FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.