OBJECTIVES: Glaucoma is an incurable eye disease and the second leading cause of blindness in the world. Until 2020, the number of patients of this disease is estimated to increase. This paper proposes a glaucoma detection method using statistical features and the k-nearest neighbor algorithm as the classifier. METHODS: We propose three statistical features, namely, the mean, smoothness and 3rd moment, which are extracted from images of the optic nerve head. These three features are obtained through feature extraction followed by feature selection using the correlation feature selection method. To classify those features, we apply the k-nearest neighbor algorithm as a classifier to perform glaucoma detection on fundus images. RESULTS: To evaluate the performance of the proposed method, 84 fundus images were used as experimental data consisting of 41 glaucoma image and 43 normal images. The performance of our proposed method was measured in terms of accuracy, and the overall result achieved in this work was 95.24%, respectively. CONCLUSIONS: This research showed that the proposed method using three statistics features achieves good performance for glaucoma detection.
Blindness ; Classification ; Eye Diseases ; Glaucoma ; Humans ; Methods ; Optic Disk ; Optic Nerve Diseases ; Retinal Degeneration

PURPOSE: To seek for mechanisms to prevent fixed dilated pupil including Urrets-Zavalia syndrome after intraocular surgery by analyzing and classifying the causes of such cases. METHODS: Medical records and anterior segment photographic images of patients with fixed dilated pupil who underwent penetrating keratoplasty, lamellar keratoplasty, or cataract surgery were analyzed in a retrospective manner from April, 1984 to February, 2014. RESULTS: Among 15 cases of postoperative fixed dilated pupil, 8 eyes of keratoconus eyes had received penetrating keratoplasty done and 7 eyes with ocular disorders other than keratoconus underwent intraocular surgeries. In cases 1 and case 2, which received penetrating keratoplasty for keratoconus, dilated pupil with regular pupil border, iris atrophy, and secondary glaucoma occurred; these cases were classified as group 1 and diagnosed as Urrets-Zavalia syndrome. Cases from 3 to 8 which also received penetrating keratoplasties due to keratoconus, irregularly dilated pupil, severe iris atrophy, posterior synechiae after moderate to severe inflammation in the anterior chamber, and fibrotic membrane on the anterior capsule occurred; these cases were classified as group 2. Finally, cases 9 to 15, which had mild inflammation, no fibrotic membrane, and regularly fixed dilated pupil after receiving other intraocular surgeries were classified as group 3. CONCLUSIONS: Differences exist between definite Urrets-Zavalia syndrome and postoperative fixed dilated pupil with regards to regularity of pupillary margin, degree of iris atrophy, posterior synechiae, fibrotic membrane, and posterior subcapsular opacity. Therefore, a new classification of fixed dilated pupil after intraocular surgery which addresses these characteristics is required and various trials to prevent the adverse postoperative complications of fixed dilated pupil should be performed. Preventive measures may include careful control of intraocular pressure, restricting atropine use, completely removing of viscoelastics, and minimal air or gas injection.
Anterior Chamber ; Atrophy ; Atropine ; Cataract ; Classification* ; Corneal Transplantation ; Glaucoma ; Humans ; Inflammation ; Intraocular Pressure ; Iris ; Keratoconus ; Keratoplasty, Penetrating ; Medical Records ; Membranes ; Postoperative Complications ; Pupil* ; Retrospective Studies

PURPOSE: To evaluate retinal nerve fiber layer (RNFL) defect by a new scoring system for RNFL photography using the Heidelberg Retina Angiograph 1 (HRA1). METHODS: This retrospective study included 128 healthy eyes and 836 primary open-angle glaucoma eyes. The RNFL photography using HRA1 was interpreted using a new scoring system, and correlated with visual field indices of standard automated perimetry (SAP). Using the presence of RNFL defect, darkness, width, and location, we established the new scoring system of RNFL photos. RESULTS: The mean RNFL defect score I in the early, moderate, severe, and control groups were 7.3, 9.2, 10.4, and 3.6, respectively. The mean RNFL defect score II in the early, moderate, severe, and control groups were 14.5, 28.5, 43.4, and 3.4, respectively. Correlations between the RNFL defect score II and the mean deviation of SAP was the strongest of the various combinations (r=-0.675, P<.001). CONCLUSIONS: Using a new scoring system, we propose a method for semi-quantitative interpretation of RNFL photographs. This scoring system may be helpful to distinguish between normal and glaucomatous eyes, and the score is associated with the severity of visual field loss.
Equipment Design ; Female ; Fluorescein Angiography/*instrumentation ; Fundus Oculi ; Glaucoma, Open-Angle/*classification/pathology/physiopathology ; Humans ; Male ; Middle Aged ; Nerve Fibers/*pathology ; Perimetry ; Photography/*instrumentation ; Reproducibility of Results ; Research Design/*statistics & numerical data ; Retinal Ganglion Cells/*pathology ; Retrospective Studies ; Severity of Illness Index ; Visual Fields

PURPOSE: This study investigated whether the combined use of Frequency Doubling Technology and Heidelberg Retina Tomography II increases the usefulness of glaucoma detection. METHODS: The study subjects were, 59 patients with POAG and 34 normal persons were selected. The authors defined a total of six discriminative criteria: F1, in which the conventional analysis of FDT was used; F2, in which the FDT scoring system was used; F3, in which the Moorfield classification of HRT was used; F4, in which the new HRT discrimination function `Z' was used; F5, in which F1 and F3 were combined, and F6, in which F2 and F4 were combined; The usefulness of each criterion was investigated. RESULTS: The sensitivity, specificity, and area under the ROC curve of the six discriminative criteria were: 86.4%, 58.8%, and 0.740 in F1, 4.5%, 80.0% (cutoff: 12.50), and 0.882 in F2, 79.7%, 61.8%, and 0.679 in F3, 77.6%, 74.3% (cutoff: 0), and 0.857 in F4, 77.6%, 71.4% (cutoff: 1.0805), and 0.767 in F5, and 86.2%, 91.4% (cutoff: 0.495) and 0.931 in F6. CONCLUSIONS: Combined use of functional (FDT) and morphological (HRT) aspects, which are early parameters of glaucoma damage, gave more useful results than a single analysis of each. Thererfore, the combined use will help the early detection of glaucoma.
Classification ; Discrimination (Psychology)* ; Glaucoma* ; Humans ; Retina* ; ROC Curve ; Sensitivity and Specificity

PURPOSE: To find the optimal parameter of the optic nerve head (ONH) analysis using Optical Coherence Tomography (OCT) and to form a diagnostic classification for the detection of glaucoma in Koreans. METHODS: The subjects of this study were 799 eyes of 400 patients at Kangnam St. Mary's Hospital from June 2003 to March 2004. Complete ophthalmic examinations containing the visual field test and OCT ONH analysis (StratusOCT, software ver. 3.0, Carl Zeiss Meditec, Inc, Dublin, CA, USA) were performed and the results were analyzed in the normal, glaucoma suspect and glaucoma groups. RESULTS: In the 799 eyes, there were 378 glaucomatous eyes. There was a statistically significant difference between glaucoma and non-glaucoma patients in all parameters of OCT ONH analysis except for the disc area. The horizontal integrated rim area (HIRA) was most significantly correlated with the visual field indices (MD and CPSD) and was the most useful parameter in glaucoma detection. Based on age and HIRA, the new discriminant formula = -1.858 - 0.028 x Age + 2.518 x HIRA was obtained and its accuracy, sensitivity and specificity were 72.8%, 71.5% and 73.8% respectively. CONCLUSIONS: HIRA was the most useful parameter in the detection of glaucoma. The new discriminant formula from OCT ONH analysis is expected to be useful in the diagnosis and follow-up of glaucoma in Koreans.
Classification ; Diagnosis* ; Follow-Up Studies ; Glaucoma* ; Humans ; Optic Disk* ; Optic Nerve* ; Sensitivity and Specificity ; Tomography, Optical Coherence* ; Visual Field Tests ; Visual Fields

PURPOSE: To evaluate the changes of electroretinographic responses according to the lens opacity type and severity in cataract patients with normal retina. METHODS: Retrospective study was performed on 79 eyes of 79 patients who had undergone electroretinograms (ERG) before cataract surgery. The patients with high myopia, diabetes mellitus, hypertension or glaucoma were excluded. Cataracts were classified using Lens Opacities Classification System III (LOCS III). Whether the ERG results are different according to the location and severity of lens opacity was analyzed. RESULTS: Marked reduction in amplitude and delay in implicit time were observed in mature cataracts. Amplitudes of a- and b-wave were reduced significantly, especially a-wave in maximal combined response and b-wave in rod response (P<0.01). Among the 4 groups, excluding the mature opacity, amplitude was reduced in posterior subcapsular opacity. Implicit times were also delayed, but not significant. CONCLUSIONS: Lens opacity, especially posterior subcapsular opacity affects the ERG results significantly. These findings should be considered in evaluating the retinal abnormalities in patients with cataract.
Cataract* ; Classification ; Diabetes Mellitus ; Glaucoma ; Humans ; Hypertension ; Myopia ; Retina ; Retinaldehyde ; Retrospective Studies

PURPOSE: We assessed whether Heidelberg retinal tomograph(HRT)is helpful in predicting the progression of glaucomatous field defect. METHODS: The periodic HRT and visual field results were obtained from 103 patients with primary open angle glaucoma (POAG) or glaucoma suspect who had been in our hospital from 1998 to 2000. Except 45 patients who had undergone ocular operation or had a follow-up of less than 1 year, 58 patients (48 eyes with POAG and 29 eyes with glaucoma) were selected. RESULTS: For glaucoma patients, the sensitivity and specificity to HRT on progression of glaucomatous field defect showed 35.7% and 33.3%, respectively. Their positive predictive value was 78.9%, and negative predictive value 6.9%. According to the results obtained after adjusting the cutoff value of classification score in HRT to -0.32 through the ROC curve, the sensitivity and specificity of progression in the glaucomatous field for glaucoma patients were changed to 57.1% and to 33.3%. Furthermore, the positive predictive value was 86% and the negative predictive value was 10% after the adjustment above. CONCLUSIONS: Our results showed that there was no significant correlation between HRT and progression of glaucomatous field defect in POAG patients or glaucoma suspect patients.
Classification ; Follow-Up Studies ; Glaucoma ; Glaucoma, Open-Angle ; Humans ; Retinaldehyde* ; ROC Curve ; Sensitivity and Specificity ; Visual Fields*

PURPOSE: To evaluate the relationship between results obtained by Heidelberg Retina Tomography (HRT) and frequency-doubling technology (FDT) in glaucoma suspect. METHODS: A total of 24 patients with glaucoma suspect were included in this study. Each subject underwent 2 visual field tests (Humphrey central 30-2 and FDT N-30), a comprehensive ophthalmic examination, and HRT upon each eye. RESULTS: A comparison of FDT global indices with the parameters of HRT, MD and PSD of FDT showed no correlation with HRT parameters. However, there was a significant correlation between the number of abnormal points in FDT and the following HRT parameters cup-disc area ratio, rim volume, mean nerve fiber layer thickness, retinal nerve fiber layer cross sectional area, and classification (p<0.05). When comparing the group having two or less abnormal points by FDT to the group having three or more abnormal points by FDT, differences were found in the following HRT parameters disc area, cup-disc area ratio, rim volume, cup shape measure, mean nerve fiber layer thickness, nerve fiber layer cross sectional area, and classification (p<0.05). CONCLUSIONS: Significant correlations were found between the number of abnormal points in FDT and HRT parameters. We believe that the correlations between the results of FDT and HRT parameters in preperimetric glaucoma with normal achromatic visual field suggest that FDT may be used for early diagnosis in patients with glaucoma.
Classification ; Early Diagnosis ; Glaucoma* ; Humans ; Nerve Fibers ; Retina ; Retinaldehyde ; Visual Field Tests ; Visual Fields

PURPOSE: It is difficult to estimate definite intraocular pressure of neonates. We intended to evaluate neonatal normal intraocular pressure using Tono-pen, and compared mean intraocular pressures between groups divided by sex, birth weight, and gestational age. METHOD: We measured the intraocular pressure with Tono-pen in 504 eyes of 252 newborn, who admitted in our hospital from June 2000 to January 2001 at their the second day of life. One observer measured the intraocular pressure with Tono-pen five to ten times after proparacain HCL 0.5% instillation on regular time. Only 5% of Tono-pen's coefficient of variance was considered valuable. Neonates suspected having congenital glaucoma or retinopathy were excluded. Conclusively, 462 eyes of 231 newborns were included. Mean gestational age was 38 weeks and 6 days. Mean birth weight was 3.23 kg. We achieved average intraocular pressures of neonates having gestational age between 38 weeks and 42 weeks, and birth weight between 3.0 kg and 3.5 kg. We achieved also average intraocular pressure of neonates divided by sex, gestational age, and birth weight, compared them with average intraocular pressure of normal neonates by 3-way ANOVA. RESULT: Average of intraocular pressure of normal neonates was 13.61+/-2.02 mmHg. Intraocular pressure of normal neonates is not statistically significant in the classification by sex, gestational age, and birth weight.(p-value>0.05) CONCLUSION: These value will aid the ophthalmologists in the diagnosis and screening of congenital glaucoma.
Birth Weight ; Classification ; Diagnosis ; Gestational Age ; Glaucoma ; Humans ; Infant, Newborn* ; Intraocular Pressure* ; Mass Screening ; Parturition

Objective: To determine the occurrence of the various types of glaucoma seen at the Ophthalmology Department of a tertiary training hospital. Methodology: Descriptive study based on the medical records available at the glaucoma clinic of a tertiary training hospital from January 1999 to August 2000 Results: 310 medical records were reviewed of which 34.84 percent were of males and 65.16 percent females. 84 percent of the patients were older than 40 years old. 54.83 percent of the patients had primary glaucomas of which primary angle closure (34.83 percent) was the most common followed by the open angle type (20 percent). 12.58 percent of the patients had secondary glaucomas Conclusion: The predominant form of glaucoma in the region remains to be primary angle closure and this has been demonstrated in this present study.
Human ; Male ; Female ; Aged 80 and over ; Aged ; Middle Aged ; Adult ; Young Adult ; Adolescent ; Child ; EY DISEASES ; GLAUCOMA ; GLAUCOMA/CLASSIFICATION ; GLAUCOMA/EPIDEMIOLOGY ; GLAUCOMA, ANGLE-CLOSURE ; GLAUCOMA, OPEN-ANGLE ; HOSPITALS ; HOSPITAL/CLASSIFICATION