University of Dundee

Emanuele Trucco

Emanuele Trucco, PhD, FRSA

University of Dundee, UK

VAMPIRE / CVIP group
School of Computing

Interests

Image processing / Automatic retinal image analysis / Vasculature detection and characterization / Summarization of angiographic sequences / Validation methodology

Techniques

Image and video processing / Machine learning / Statistical validation methods

Project: ESR 2.2 Dynamic vascular measurement in UWFV SLO fluorescein angiograms

Various instruments are used currently for clinical inspection of the fundus of the human eye, e.g., the slit camera, the fundus camera, and the scanning laser ophthalmoscopes (SLO). This project aims to build on the solid research platform of the VAMPIRE group at the University of Dundee to develop reliable algorithms for automatic analysis, quantification and summarization of the SLO fluorescein angiography (FA) exams./p>

SLO FA exams generate sequences of images of the eye fundus acquired over 10-15 minutes as a medium of contrast (fluorescein) perfuses the retinal vasculature and is then washed away by the blood stream. FA is an important means to identify anomalies in the retinal vasculature.

The VAMPIRE group has been working for 10 years with OPTOS plc, world-leading company for SLO devices, to develop software tool for SLO FA analysis. This project builds on a wealth of home-grown software for registration, vasculature detection and characterization, lesion detection and validation. In particular, the project addresses
– characterization and detection of lesions related to vein/artery occlusion and ischemia
– image feature selection for specific lesion classification
– validation on annotated and weakly annotated data sets.

The project includes a secondment to the University of Padova (Prof A Ruggeri) as part of the training and collaborative research, regular exposure to the research of project partners (especially to the research of Prof A Hunter from the University of Lincoln), and participation in the interdisciplinary, very rich cohort-building and training events of the project.

Major 5 publications

Yangfan Wang, Ping Lin, E Trucco: Retinal Vessel Segmentation Using Multiwavelet Kernels and Multiscale Hierarchical Decomposition. Pattern Recognition , in press, 2013. Published online 26 Feb 2013, DOI: 10.1016/j.patcog.2012.12.014

Emanuele Trucco, Alfredo Ruggeri, Thomas Karnowski, Luca Giancardo, Edward Chaum, Jean Pierre Hubschman, Bashir al-Diri, Carol Y Cheung, Damon Wong, Michael Abràmoff, Gilbert Lim, Dinesh Kumar, Philippe Burlina, Neil M Bressler, Herbert Jelinek, Fabrice Meriaudeau, Gwènolè Quellec, Tom MacGillivray, Bal Dhillon: Validating retinal fundus image analysis algorithms: issues and a proposal. Investigative Ophthalmology and Visual Science, in press, 2013.

AP Rovira, R Cabido, E Trucco, S McKenna and J P Hubschman: RERBEE: Robust Efficient Registration via Bifurcations and Elongated Elements applied to retinal fluorescein angiogram sequences. IEEE Trans on Medical Image Processing, vol 31 no 1, Jan 2012.

E. Trucco, H Azegrouz, B Dhillon: Modeling the tortuosity of retinal vessels: does calibre play a role?, IEEE Trans on Biomedical Engineering, vol 57 no 9, Sep 2010, pp 2239-2247.

C Lupascu, D Tegolo and E Trucco: FABC: Retinal Vessel Segmentation using AdaBoost. IEEE Transactions on Information Technology in Biomedicine, vol 14, no 5, Sept. 2010, pp 1267 – 1274. ISSN: 1089-7771.

 

Leave a Reply

Your email address will not be published. Required fields are marked *