Abstract
Purpose: :
Aim of the study is to increase quantitative informations on optical coherence tomography concerning tissues reflectivity, geometrical dimension of the retinal scanned area and its details.
Methods: :
We applied a number of image processing techniques (well known in industrial image processing) to selected OCT retinal scans. We have created a simple software tool, which implements different elaboration tasks: image enhancement on a logarithmic scale, binarization of the grey levels, detections and elimination of outlayers, application of median filtering, blob analysis to measure the areas of the low signal and edge detection to measure point to point distances.
Results: :
The experiments carried out highlighted the ability of the process to detect small retinal alterations like superficial epiretinal tractions and holes (5 pixels) and distances between very close points (3 pixels) to evaluate intraretinal edema or pathologic lesions. The elaboration of the images with a 1300 by 1030 pixels resolution permit to enhance intraretinal abnormalities of 6 pixels.
Conclusions: :
The results showed the possibility to obtain in a user friendly and interactive way geometrical dimension of any region of interest in the scan, to help the surgeon to most effectively perform quantitative analysis and patient follow up. The user can interactively chose the level of enhancement, perform the measurement of areas and perimeters within suitable regions and detect distances along suitable profiles down to few tens of pixel. The work performed so far demonstrate the flexibility and the accuracy of the overall process, to study clinical aspects at different stages, both pre and post operative.
Keywords: imaging/image analysis: clinical • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • macula/fovea