Isabel Guedes-Guedes

Isabel Guedes-Guedes


Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 December 2022 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
MGMarta Gonzalez-Hernandez
Marta Gonzalez-Hernandez
DGDaniel Gonzalez-Hernandez
Daniel Gonzalez-Hernandez
NCNisamar Betancor Caro
Nisamar Betancor Caro
IGIsabel Guedes-Guedes
Isabel Guedes-Guedes
MGMorten Kirk Guldager
Morten Kirk Guldager
MRManuel Gonzalez de la Rosa
Manuel Gonzalez de la Rosa
Background: Laguna ONhE provides a globin distribution function (GDF), in which a glaucoma discriminator based on deep learning plays an important role, and there is also an optimized globin individual pointer (GIP) for progression analysis. Methods: Signs of optic nerve glaucoma were identified in 1,124,885 fundus images from 203,115 diabetics obtained over 15 years and 117,813 control images. Results: A total of 743,696 images from 313,040 eyes of 173,661 diabetics were analysed. Some exclusions occurred due to excessive illumination, poor quality, or the absence of optic discs. Suspicion of glaucoma was reported in 6.70%, for an intended specificity of 99% (GDF < −15). More signs of glaucoma occur in diabetics as their years of disease increase, and after age 60, compared to controls. The GIP detected progression (p < 0.01) in 2.59% of cases with 4 controls and in 42.6% with 14 controls was higher in cases with lower GDF values. The GDF was corrected for the disc area and proved to be independent of it (r = 0.001925; p = 0.2814). Conclusions: The GDF index suggests a higher and increasing glaucoma probability in diabetics over time. Doubling the number of check-ups from four to eight increases the ability to detect GIP index progression by a factor of 5.

10 September 2023 in Journal of Clinical Medicine
Verified
Journal Article
Open access
PDF file
MGMarta Gonzalez-Hernandez
Marta Gonzalez-Hernandez
NCNisamar Betancor Caro
Nisamar Betancor Caro
FMFatima Mesa-Lugo
Fatima Mesa-Lugo
IRIvan Rodriguez-Talavera
Ivan Rodriguez-Talavera
APAlicia Pareja-Rios
Alicia Pareja-Rios
MRManuel Gonzalez de la Rosa
Manuel Gonzalez de la Rosa
Background: Previous retrospective results are evaluated prospectively and blinded. Methods: A total of 221 eyes previously classified as normal (G1), 279 as moderate risk of glaucoma (G2) and 217 as high risk (G3) according to the Globin Discriminant Function (GDF) Laguna-ONhE index were examined with OCT Spectralis- Results: In G1, the Bruch’s Membrane Opening Minimum Rim Width (BMO-MRW) was 332 ± 55 microns; in G2, it was 252 ± 47 (p < 0.0001); and in G3, 231 ± 44 (p < 0.0001). In G1, the 1% and 5% percentiles were 233 and 248, respectively; in G2, they were lower in 28.80% and 42.29% of cases, respectively; and in G3, in 50.23% and 63.59% of cases, respectively. Most of the cases were normal-tension glaucomas. Laguna-ONhE indices showed a curvilinear correlation with BMO-MRW results. The Retinal Nerve Fibre Layer (RNFL) showed a poor relationship with BMO. Assuming G1 to be truly normal, BMO-MRW would have a Receiver operating characteristic (ROC) curve area of 0.901 for G2 and G3 and 0.651 for RNFL. A significant reduction in pixels corresponding to vessels was found in G2 and G3 vs. G1 (p < 0.0001). Conclusions: In some cases, these defects appear to be mainly glaucomatous, and in others, they are associated with diabetic microangiopathy. In normal tension glaucoma, RNFL defects may be less severe than those inside the nerve.

31 January 2025 in Archivos de la Sociedad Española de Oftalmología (English Edition)
Verified
Journal Article
Closed access
CPC. Piña-Delgado
C. Piña-Delgado
MBM. Bolaños-Rivero
M. Bolaños-Rivero
IGIsabel Guedes-Guedes
Isabel Guedes-Guedes
LSL. Siguero-Martín
L. Siguero-Martín
IMI. de Miguel-Martínez
I. de Miguel-Martínez

19 November 2025 in International Journal of Retina and Vitreous
Verified
Journal Article
Open access
PDF file
IGIsabel Guedes-Guedes
Isabel Guedes-Guedes
PSPedro Saavedra Santana
Pedro Saavedra Santana
FLFrancisco Cabrera López
Francisco Cabrera López
ÁMÁngel Ramos Macías
Ángel Ramos Macías
ÁMÁngel Ramos de Miguel
Ángel Ramos de Miguel
AHAyoze González Hernández
Ayoze González Hernández
Objective or purpose: To evaluate the diagnostic performance and agreement of the EyeArt ® Artificial Intelligence (AI) system for detecting Diabetic Retinopathy (DR), comparing its results with ophthalmologists’ assessments in a regional screening program. Design: Cross-sectional observational study. Subjects, participants, and/or controls: A total of 498 diabetic patients aged 18 years or older were enrolled between June and September 2023 through the Retisalud DR screening program in the Canary Islands. No separate control group was included. Methods: All participants underwent non-mydriatic fundus photography using the TRC-NW400 camera. Retinal images were analyzed by the EyeArt ® AI system (version 2.1.0), and results were compared with assessments by ophthalmologists based on the International Clinical Diabetic Retinopathy scale (ICDR). Agreement was quantified using Cohen’s kappa coefficient. Additionally, mixed-effects logistic regression was used to explore associations between DR and clinical risk factors. Main outcome measures: Sensitivity, specificity, and agreement (Cohen’s kappa) of the AI system compared to clinical diagnosis; predictors of DR such as age, diabetes duration, presence of Diabetic Macular Edema (DME), and central retinal thickness (CRT-OCT). Results: The EyeArt® system achieved a binocular sensitivity of 100% (95% CI: 98.1–100) and a specificity of 93.5% (95% CI: 90.2–96.0). Agreement with ophthalmologist grading was excellent, with kappa values of 0.966 (right eye) and 0.978 (left eye). Younger age, longer diabetes duration, DME presence, and higher CRT were significantly associated with DR diagnosis. Conclusions: The EyeArt ® AI system showed excellent diagnostic accuracy and strong agreement with clinical evaluations in DR screening. Nonetheless, its tendency to overestimate DR severity indicates the need for further refinement of its grading algorithm. These findings support the potential integration of AI systems into large-scale DR screening programs, pending further validation.