Automated analysis and interpretation of retinal images has become an incontournable diagnostic step in ophthalmology. Retinal blood vessels morphology can be an important indicator for diseases such as diabetic retinopathy; and their detection also serves for image registration. This paper presents a method based on mathematical morphology for extraction of vascular tree in color retinal image with low contrast. It consists in contrast enhancement and application of watershed transformation in order to segment blood vessels in digital fundus images.

No abstract

No abstract

Keratoconus is an eye disease causing progressive corneal thinning. At an early stage (fruste keratoconus), the symptoms can overlap with those of other eye disorders, making the diagnosis difficult. This led us to propose a new image processing pipeline to automatically calculate the main descriptors used by Ophthalmologists to detect keratoconus, and then classify these data in order to propose an intelligent system able to help specialists in the early recognition of this pathology. To accomplish this, we elaborated a new benchmark database from a corneal topographic Orbscan II device. For keratoconus classification, five different machine learning methods are tested on our new locally collected database, which are: K-Nearest Neighbors (K-NN), Support Vector Machines (SVM), Decision Trees (DT), and two neural networks classifiers (the Radial Basis Function (RBF) and the Multi-Layer Perceptron (MLP)). Experimental results indicate that all classifiers achieved good precision when using all descriptors (the numerical parameters given by the ORBSCAN II topographer and the descriptors obtained after an image processing of the topography map). Furthermore, the SVM outperformed the other classifiers with an accuracy rate of 95.04%. These results were confirmed and validated by a group of experts in ophthalmology and prove the efficiency of our system and the coherence of our new database.

The presence of microcalcifications (MCs) in X-ray mammograms provides an important early sign of women breast cancer. However, their detection still remains very complex due to the diversity in shape, size, their distributions and to the low contrast between the cancerous areas and surrounding bright structures in mammograms. This paper presents an effective approach based on mathematical morphology for detection of MCs in digitised mammograms. The developed approach performs an initial step in order to extract the breast area and removing unwanted artefacts out of the mammogram. Subsequently, an enhancement process is applied to improve appearance and increase the contrast of images and to eliminate noise. Once the breast region has been found, a segmentation phase through morphological watershed is performed in order to detect MCs. The performance of our approach is evaluated using a total of 22 mammograms extracted from the MIAS mammographic database, showing the presence of MCs. The obtained results were compared with manual detection, marked by an expert mammographic radiologist. These results show that the system is very effective, especially in terms of sensitivity.