No abstract

Early detection of microaneurysms (MAs), the first sign of Diabetic Retinopathy (DR), is an essential first step in automated detection of DR to prevent vision loss and blindness. This study presents a novel and different algorithm for automatic detection of MAs in fluorescein angiography (FA) fundus images, based on Radon transform (RT) and multi-overlapping windows. This project addresses a novel method, in detection of retinal land marks and lesions to diagnose the DR. At the first step, optic nerve head (ONH) was detected and masked. In preprocessing stage, top-hat transformation and averaging filter were applied to remove the background. In main processing section, firstly, we divided the whole preprocessed image into sub-images and then segmented and masked the vascular tree by applying RT in each sub-image. After detecting and masking retinal vessels and ONH, MAs were detected and numbered by using RT and appropriated thresholding. The results of the proposed method were evaluated reported on three different retinal images databases, the Mashhad Database with 120 FA fundus images, Second Local Database from Tehran with 50 FA retinal images and a part of Retinopathy Online Challenge (ROC) database with 22images. Automated DR detection demonstrated a sensitivity and specificity of 94% and 75% for Mashhad database and 100% and 70% for the Second Local Database respectively.

Ab stract-In medical images linear patterns such as blood vessels are important structures for computer-aided diagnosis and follow-up of many diseases. Moreover, image processing techniques are required to extract suitable information about vascular tree and its alteration. Analyzing of retinal blood vessel is critical work for the investigation of some diseases. In this study, we present an automated method for detecting retinal vasculatures based upon Radon transform. In preprocessing, we used top-hat transformation and averaging filter. Our main processing was included applying Radon transform, vessel certifying, and vessel refinement. Comparing the results of our method with gold standard showed that our results have more than 93% for true positive rate. In conclusion, it is possible to use Radon transform for vessel segmentation in fluorescein angiography fundus images, with acceptable sensitivity and specificity, as a necessary step in some diagnostic algorithm for retinal pathology.

No abstract

Purpose:To compare the distribution of different sized vessels using digital photographs of the ocular surface of diabetic and normal individuals.Methods:In this cross-sectional study, red-free conjunctival photographs of diabetic and normal individuals, aged 30-60 years, were taken under defined conditions and analyzed using a Radon transform-based algorithm for vascular segmentation. The image areas occupied by vessels (AOV) of different diameters were calculated. The main outcome measure was the distribution curve of mean AOV of different sized vessels. Secondary outcome measures included total AOV and standard deviation (SD) of AOV of different sized vessels.Results:Two hundred and sixty-eight diabetic patients and 297 normal (control) individuals were included, differing in age (45.50 ± 5.19 vs. 40.38 ± 6.19 years, P < 0.001), systolic (126.37 ± 20.25 vs. 119.21 ± 15.81 mmHg, P < 0.001) and diastolic (78.14 ± 14.21 vs. 67.54 ± 11.46 mmHg, P < 0.001) blood pressures. The distribution curves of mean AOV differed between patients and controls (smaller AOV for larger vessels in patients; P < 0.001) as well as between patients without retinopathy and those with non-proliferative diabetic retinopathy (NPDR); with larger AOV for smaller vessels in NPDR (P < 0.001). Controlling for the effect of confounders, patients had a smaller total AOV, larger total SD of AOV, and a more skewed distribution curve of vessels compared to controls.Conclusion:Presence of diabetes mellitus is associated with contraction of larger vessels in the conjunctiva. Smaller vessels dilate with diabetic retinopathy. These findings may be useful in the photographic screening of diabetes mellitus and retinopathy.

PurposeTo design software with a novel algorithm, which analyzes the tortuosity and vascular dilatation in fundal images of retinopathy of prematurity (ROP) patients with an acceptable accuracy for detecting plus disease.MethodsEighty-seven well-focused fundal images taken with RetCam were classified to three groups of plus, non-plus, and pre-plus by agreement between three ROP experts. Automated algorithms in this study were designed based on two methods: the curvature measure and distance transform for assessment of tortuosity and vascular dilatation, respectively as two major parameters of plus disease detection.ResultsThirty-eight plus, 12 pre-plus, and 37 non-plus images, which were classified by three experts, were tested by an automated algorithm and software evaluated the correct grouping of images in comparison to expert voting with three different classifiers, k-nearest neighbor, support vector machine and multilayer perceptron network. The plus, pre-plus, and non-plus images were analyzed with 72.3%, 83.7%, and 84.4% accuracy, respectively.ConclusionsThe new automated algorithm used in this pilot scheme for diagnosis and screening of patients with plus ROP has acceptable accuracy. With more improvements, it may become particularly useful, especially in centers without a skilled person in the ROP field.

No abstract

The identification of the optic nerve head (ONH) is necessary preprocessing step in retinal image analysis, for automated extraction of the anatomical components in retinal images. In this study, a new image processing method based on Radon transform (RT) and multi-overlapping windows was proposed for ONH detection in fluorescein angiography (FA) fundus images. At first, RT was applied to all fundus sub images to find candidates for the location of the ONH. Then, the accurate location was found using the minimum mean square error estimation. The results of our automated method for the ONH detection in the images showed sensitivity and specificity of 90.54%, 98.51% respectively for pixel based analysis, and according to manual ONH detection, our automated algorithm found 89 ONH out of 100 in true location for FA images. This study addresses a novel method in detection of retinal land marks. Sensitivity and specificity of this algorithm seems to be acceptable in comparison with other detection methods.