ABSTRACT
Purpose@#To report the availability of optical coherence tomography angiography (OCTA) when diagnosing type 1 and type 2 choroidal neovascularization (CNV) associated with age-related macular degeneration. @*Methods@#We conducted a retrospective chart review of 63 eyes of 59 patients who visited the outpatient clinic from April 2018 to January 2020 with treatment-naïve type 1/2 CNV associated with age-related macular degeneration. The CNV was diagnosed through fluorescein angiography and indocyanine green angiography on the appearance of hyperfluorescence and leakage at the late phase. The CNV type was classified by location using OCT. These results were compared with those of automatic and manual segmentations via OCTA. @*Results@#Of the 63 eyes, 45 eyes showed type 1 CNV and 18 eyes indicated type 2 CNV. Using the automatic segmented display of OCTA, type 1 CNV was identified in 29 of 45 eyes and type 2 CNV was confirmed in 17 of 18 eyes. In addition, when manual segmentation was performed, type 1 CNV was found in 41 of 45 eyes and type 2 CNV in 18 of 18 eyes. The sensitivity of diagnosis of type 1 CNV using only automatic segmentation in OCTA was 64.44%; the sensitivity increased to 91.11% when additional manual segmentation was performed (p < 0.001). For type 2 CNV, the sensitivity was 94.44% using automatic segmentation in OCTA and 100% with additional manual segmentation (p = 1.000). @*Conclusions@#Automatic segmentation of OCTA and OCT analyses enabled diagnosis of type 1 and type 2 CNV associated with age-related macular degeneration. The diagnosis accuracy or type 1 CNV improved significantly when manual segmentation was added.
ABSTRACT
Purpose@#To report the availability of optical coherence tomography angiography (OCTA) when diagnosing type 1 and type 2 choroidal neovascularization (CNV) associated with age-related macular degeneration. @*Methods@#We conducted a retrospective chart review of 63 eyes of 59 patients who visited the outpatient clinic from April 2018 to January 2020 with treatment-naïve type 1/2 CNV associated with age-related macular degeneration. The CNV was diagnosed through fluorescein angiography and indocyanine green angiography on the appearance of hyperfluorescence and leakage at the late phase. The CNV type was classified by location using OCT. These results were compared with those of automatic and manual segmentations via OCTA. @*Results@#Of the 63 eyes, 45 eyes showed type 1 CNV and 18 eyes indicated type 2 CNV. Using the automatic segmented display of OCTA, type 1 CNV was identified in 29 of 45 eyes and type 2 CNV was confirmed in 17 of 18 eyes. In addition, when manual segmentation was performed, type 1 CNV was found in 41 of 45 eyes and type 2 CNV in 18 of 18 eyes. The sensitivity of diagnosis of type 1 CNV using only automatic segmentation in OCTA was 64.44%; the sensitivity increased to 91.11% when additional manual segmentation was performed (p < 0.001). For type 2 CNV, the sensitivity was 94.44% using automatic segmentation in OCTA and 100% with additional manual segmentation (p = 1.000). @*Conclusions@#Automatic segmentation of OCTA and OCT analyses enabled diagnosis of type 1 and type 2 CNV associated with age-related macular degeneration. The diagnosis accuracy or type 1 CNV improved significantly when manual segmentation was added.
ABSTRACT
PURPOSE: To compare corneal astigmatism, keratometry and corneal higher order aberrations between the light emitting diode corneal topography analyzer and Scheimpflug Imager. METHODS: This prospective study involved 45 patients (45 eyes) who visited Seoul St. Mary's hospital before cataract surgery from June 7, 2017, to August 2, 2017. For each eye, keratometry, astigmatism and its axis of cornea, higher-order aberrations were evaluated with a Scheimpflug Imager (Pentacam HR®, Oculus, Wetzlar, Germany) and a color-LED corneal topographer (Cassini®, i-Optics, Den Haag, The Netherlands). RESULTS: Astigmatism magnitude measured using Cassini® and Pentacam® showed no statistically differences but anterior and total astigmatic axes were significantly different, as measured by the two devices (p < 0.05). Anterior and total mean keratometry were statistically significantly different, as measured by the two devices (p < 0.05). J0 and J45 vectors of anterior and total cornea were statistically different (p < 0.05). In addition, Cassini® and Pentacam® showed discrepancies between total corneal astigmatism, total J0 and J45 vectors. Corneal anterior spherical aberration, vertical and horizontal coma, and oblique and horizontal trefoil aberrations were not statistically different between the two devices. CONCLUSIONS: Astigmatic axes obtained from the two devices based on different principles showed statistically significant differences. Astigmatism magnitude was not statistically different but showed a discrepancy between the two devices.
Subject(s)
Humans , Astigmatism , Cataract , Coma , Cornea , Corneal Topography , Corneal Wavefront Aberration , Lotus , Prospective Studies , SeoulABSTRACT
PURPOSE: In this study, we compared the risk of cardiovascular complications between diabetic retinopathy (DR) groups based on the Framingham Risk Score. METHODS: Subjects 40 years of age or older were enrolled in the present study from September 2008 to September 2009. Five scales were used to evaluate the severity of DR for the most severely affected eye in each patient: no DR, mild nonproliferative DR (NPDR), moderate NPDR, severe NPDR and proliferative DR (PDR). The first group included no DR patients and the second group included mild NPDR patients. The third group included moderate NPDR, severe NPDR and PDR patients. The Framingham Risk Score was analyzed among the groups to determine whether a difference in score existed. RESULTS: DR was found in 126 subjects. The first group included 403 subjects, the second group 22 subjects and the third group 104 subjects. The average Framingham Risk Score that appeared with increasing severity of DR was 10.89 ± 7.98 in the first group, 12.25 ± 10.84 in the second group and 13.25 ± 10.66 in the third group. The average difference between the first and third groups was statistically significant (p = 0.014). CONCLUSIONS: In our study, the Framingham Risk Score increased significantly with more severe DR (p = 0.042). Our results are useful when considering the easily accessible identification of DR and that cardiovascular disease is one of the main causes of death worldwide. However, confirming the exact causal relationship based on our results is difficult because the study design was cross-sectional and additional studies are necessary.