R E T I N A		Glaucoma Detection from smartphone-based ophthalmoscopic images
		Development of Deep Learning algorithms for the analysis of fundus images acquired with smartphone-based ophthalmoscopes. The algorithms autoamtically select the region depicting the optic disc, segment the relevant structures and extract clinically useful indexex for Glaucoma diagnosis, e.g. Vertical Cup to Disc Ratio.
		Vessel Structure Tracking
		Identification and measurement of blood vessels in retinal images could allow quantitative evaluation of clinical features, which may allow early diagnosis and effective monitoring of therapies in retinopathy. A new system is proposed for the automatic extraction of the vascular structure in retinal images, based on a sparse tracking technique.
		Optic Disk Detection
		All retinal vessels originate from the optic disc and then follow a parabolic course towards retinal edges. Thus, a geometrical parametric model was proposed to describe the direction of these vessels and two of the model parameters are just the coordinates of the optic disc center.
		Non-Vascular Lesion Detection
		The most distinctive sign of diabetic retinopathy or severe hypertensive retinopathy are haemorrhages and microaneurysms (HM), hard exudates (HE) and cotton wool spots (CWS). Automatic detection of their presence in the retina is thus of paramount importance for assessing the presence of retinopathy.
		Vessel Analysis: Tortuosity and AVR
		Tortuosity is among the first alterations in retinal vessel network to appear in many retinopathies. Generalized arteriolar narrowing is an important sign of hypertension. It is expressed by AVR (Arteriolar-to-Venular diameter Ratio), which requires the long and subjective manual measurement of caliber in many arteries and veins.
		Dynamic Fluorangiography
		In order to assess blood circulation in the human retina, we studied a computerized system to automatically and accurately measure the transit time of fluorescein dye between the user-selected retinal locations.

C O R N E A		Automatic Nerve Tracing
		The recognition of nerve structures in the cornea appears to be an important clinical issue, e.g. to investigate about damages from surgical interventions (LASIK/PRK) or severity of diabetic neuropathy. We addressed the problem of recognizing and tracing corneal nerves in confocal microscopy images.
		3D Reconstruction of the Cornea
		The purpose is to get a 3D reconstruction of the cornea, starting from a confocal microscope sequence of images, from endothelium to epithelium. Confocal microscopy can provide sequences of images from all cornea layers in a rapid, in vivo and non invasive way. These images are useful to extract important clinical information on cornea state of health.
		Automatic Endothelial Cell Count
		The aim of the project is the development of an automatic system to perform estimation of the cellular density in images of the cornea endothelium, where the application of the conventional edge detection techniques or of the morphological operation techniques would fail because of the actual lack of visible edges.
		Identification of Endothelial Cells
		A new method is proposed for the automatic detection and analysis of cell field contours in images of corneal endothelium. The algorithm is based on a set of single-cell contour models (a cell field), individually described statistically in term of shape a-priori information and a-posteriori image representation.

M E L A N O M A		3D moles mapping
		The project aims at enhancing melanoma screening and diagnosis. Macroscopic images of the whole skin surface acquired from different viewpoints are used to build a total-body 3D moles mapping, through photogrammetry. The project is based on the Italian Patent n. 102016000132357.
		Morphological mole analysis
		Automatic segmentation and the consequent quantitative estimation of clinical indexes (Asimmetry, Border irregularity, Color variations, Dimension, etc.) is a fast and objective tool able to assist clinicians in the diagnostic process of melanoma. The proposed analysis take advantage of the 3D model obtained by photogrammetry.
		Thermographic mole analysis
		the temperature variation due to the greater vascularization of melanoma compared to normal nevi is a parameter of great clinical interest. We are developing methods able to accurately estimate it from thermography.

		Chromosomes Segmentation
		The work is aimed at the development and realization of algorithms for the segmentation, disentangling and subsequent classification of chromosomes from Q-band prometaphase images.
		Chromosomes Classification
		The manual analysis of the karyogram is a complex and time-consuming operation, as it requires meticulous attention to details and well-trained personnel. Routine Q-Band laboratory images show chromosomes which are randomly rotated, blurred or corrupted by overlapping and by dye stains. We address here the problem of robust automatic classification system, still an open issue.

return to top