METHODS: Firstly, color fundus images from the publicly available database DRIVE were converted from RGB to grayscale. To enhance the contrast of the dark objects (blood vessels) against the background, the dot product of the grayscale image with itself was generated. To rectify the variation in contrast, we used a 5 × 5 window filter on each pixel. Based on 5 regional features, 1 intensity feature and 2 Hessian features per scale using 9 scales, we extracted a total of 24 features. A linear minimum squared error (LMSE) classifier was trained to classify each pixel into a vessel or non-vessel pixel.
RESULTS: The DRIVE dataset provided 20 training and 20 test color fundus images. The proposed algorithm achieves a sensitivity of 72.05% with 94.79% accuracy.
CONCLUSIONS: Our proposed algorithm achieved higher accuracy (0.9206) at the peripapillary region, where the ocular manifestations in the microvasculature due to glaucoma, central retinal vein occlusion, etc. are most obvious. This supports the proposed algorithm as a strong candidate for automated vessel segmentation.
DESIGN: Prospective, population cohort study.
PARTICIPANTS: The Singapore Malay Eye Study baseline participants (age, ≥40 years; 2006-2008) were followed up in 2011 through 2013, and 1901 of 3280 of eligible participants (72.1%) took part.
METHODS: Fundus photographs were graded using the Wisconsin AMD grading system.
MAIN OUTCOME MEASURES: Incidence of early and late AMD.
RESULTS: Gradable fundus photographs were available for 1809 participants who attended both baseline and 6-year follow-up examinations. The age-standardized incidences of early and late AMD were 5.89% (95% confidence interval [CI], 4.81-7.16) and 0.76% (95% CI, 0.42-1.29), respectively. The 5-year age-standardized incidence of early AMD (calculated based on the 6-year incidence) was lower in our population (5.58%; 95% CI, 4.43-7.01) compared with the Beaver Dam Eye Study population (8.19%). The incidence of late AMD in our population was similar to that of the Beaver Dam Eye Study population (0.98% [95% CI, 0.49-1.86] vs. 0.91%), the Blue Mountains Eye Study population (1.10% [95% CI, 0.52-9.56] vs. 1.10%), and the Hisayama Study population (1.09% [95% CI, 0.54-4.25] vs. 0.84%). The incidence of late AMD increased markedly with increasing baseline AREDS score (step 0, 0.23%; step 4, 9.09%).
CONCLUSIONS: This study documented the incidence of early and late AMD in a Malay population. The AREDS simplified severity scale is useful in predicting the risk of late AMD development in Asians.