University of Dundee | Home

I'm a Marie Curie Early Stage Researcher (Marie Curie Doctoral Fellow) involved in the REVAMMAD project, part of the European Union's Marie Curie Initial Training Network programme (Marie Curie Actions, 7th framework).

My host institution is the School of Science and Engineering (Computing), University of Dundee (UK), where I'm advised by Prof. Emanuele Trucco.

My project is about leveraging modelling and machine learning for the analysis of curvilinear structures in medical images. This work is done in collaboration with clinical partners at Harvard Medical School and Tufts Medical Center (both in Boston, USA).

The technical challenge is to design and develop fully automated algorithms to quantify specific morphometric properties of the curvilinear structures in order to: (1) allow the analysis of large volumes of image data with cost-effective solutions; (2) provide objective measurements for potentially highly subjective ones (e.g., tortuosity).

Broadly speaking, my main research interests focus on novel Computer Vision and Image Processing techniques for Medical Image Analysis.


View Roberto Annunziata's profile on LinkedIn


NEWS

  • [21/09/2016]
    Excited to announce that at the beginning of October 2016 I will start a PostDoc in the Translational Imaging Group (TIG) led by Prof. Seb Ourselin at UCL (London). I will contribute to the success of a new ERC Starting Grant "Building Next-Generation Computational Tools for High Resolution Neuroimaging Studies" led by Dr Eugenio Iglesias.

  • [13/08/2016]
    Yesterday, I successfully defended my PhD thesis, subject to some minor revisions. It was great fun and an open, fair discussion. Thanks to Prof. Stephen J. McKenna (internal examiner), Prof. Giovanni Montana (external examiner from King's College of London), Dr Keith Edwards (convener) and Prof. Emanuele Trucco (my supervisor).
    This means that I managed to complete my PhD in Medical Image Analysis (Marie S.-Curie Doctoral Fellowship) in a bit less than 3 years (including viva and revised thesis, with about 7 months of placements at partner institutions). During these 3 years I have learned a lot, as a scientist and as a person (different cultures, different countries, different pizza!!!). I have had the chance to focus on the things I have true passion for (research, innovate, experiment) and I look forward to continuing on this path with the aim of giving tangible contributions to the humanity in the long-term.

  • [16/06/2016]
    As the French say..."Les jeux sont faits!"
    I have just sent my PhD thesis to the print office. Looking forward to the "viva voce" on the 12th of August.

  • [09/06/2016]
    Our journal paper (Annunziata et al., "Leveraging Multiscale Hessian-based Enhancement with a Novel Exudate Inpainting Technique for Retinal Vessel Segmentation") will appear on the next issue of the IEEE Journal of Biomedical and Health Informatics.

  • [14/05/2016]
    It is a great pleasure to share that IEEE Transactions on Medical Imaging accepted our journal paper:
    R. Annunziata and E. Trucco, "Accelerating Convolutional Sparse Coding for Curvilinear Structures Segmentation by Refining SCIRD-TS Filter Banks".
    I am grateful to Prof. Bennet Landman who handled the paper.

  • [21/04/2016]
    It is a great pleasure to announce that our Medical Image Analysis paper authored by R. Annunziata, A. Kheirkhah, S. Aggarwal, P. Hamrah and E. Trucco, titled "A Fully Automated Tortuosity Quantification System with Application to Corneal Nerve Fibres In Confocal Microscopy Images" has been accepted for publication.
    I am grateful to Prof. Nicholas Ayache who handled the paper.

  • [11/04/2016]
    I have modified the MICCAI paper "Scale and Curvature Invariant Ridge Detector for Tortuous and Fragmented Structures" and solved some incongruences between Eq. (2) and (3). You can find the ERRATA CORRIGE here.
    NOTICE THAT THIS MODIFICATION DOES NOT AFFECT THE EXPERIMENTAL RESULTS AND MATLAB IMPLEMENTATION.
    Thanks to the exceptional (anonymous) reviewer who spotted it!

  • [02/03/2016]
    An early version of our IOVS paper on tortuosity quantification and interpretation is available here.

  • [22/02/2016]
    Excited to start writing my PhD thesis after 2.5 years of research activities devoted to leveraging machine learning for biomarkers investigation (e.g. tortuosity) at the University of Dundee (UK, 2 years), Harvard Medical School (USA, 2 months) and Charite' - University of Medicine (DE, 4 months).

  • [17/12/2015]
    New journal paper:
    R. Annunziata et al., "Two-Dimensional Plane for Multi-Scale Quantification of Corneal Subbasal Nerve Tortuosity". Investigative Ophthalmology and Visual Science (ARVO).

  • [30/06/2015]
    IEEE EMBC paper accepted as oral presentation:
    R. Annunziata et al., "Combining Efficient Hand-Crafted Features with Learned Filters for Fast and Accurate Corneal Nerve Fibre Centreline Detection". IEEE Engineering in Medicine and Biology Society 2015, Milan.

  • [10/06/2015]
    Student Travel Award for MICCAI 2015.

  • [04/06/2015]
    Visiting Researcher (~ 2 months)
    MEEI, Dept of Ophthalmology, Harvard Medical School, Boston, MA, USA.

    During this period I will have the chance to collaborate with worldwide leaders in corneal and transplantation immunology, corneal imaging and inflammatory corneal disorders.
    This visit will contribute significantly to my current research topic related to the automated analysis of the morphometric properties characterising corneal nerve fibres in an objective, systematic manner. Our aim is to discover biomarkers and early signs of disease conditions which could contribute to prevent further disorders or quantify objectively the severity of the pathology.

  • [31/05/2015]
    EMBS conference paper accepted:
    R. Annunziata et al., "Combining Efficient Hand-Crafted Features with Learned Filters for Fast and Accurate Corneal Nerve Fibre Centreline Detection". IEEE Engineering in Medicine and Biology Society 2015, Milan.

  • [28/05/2015]
    New journal paper accepted:
    R. Annunziata et al., "Leveraging Multiscale Hessian-based Enhancement with a Novel Exudate Inpainting Technique for Retinal Vessel Segmentation". IEEE Journal of Biomedical and Health Informatics.

  • [27/05/2015]
    Best paper award advertised on the Dundee School of Computing website.

  • [26/05/2015]
    "Best Paper Award for Outstanding Doctoral Student" during the 2015 PhD symposium at the School of Computing, University of Dundee.

  • [20/05/2015]
    Second paper accepted at MICCAI 2015.
    This makes my submission (2 first-author papers) 100% successful at the first attempt. Thanks to co-authors (clinical authors at MEEI, Harvard Medical School) for providing and annotating the dataset and to Prof. Trucco (my supervisor) for invaluable comments and suggestions.

  • [05/05/2015]
    Paper accepted at MICCAI 2015. Acceptance rate <30% (1020 early submissions).
    This is also the first time a paper from VAMPIRE group gets accepted since the beginning of the project (2002).

  • [12/04/2015]
    Participation in the poster VAMPIRE: Vessel Assessment and Measurement Platform for Images of the REtina presented at the SICSA workshop of Medical Image Analysis held at the School of Computing, University of Dundee, UK.

  • [27/02/2015]
    Next month I will take part in a short-term research project at OPTOS PLC. OPTOS is a world-leading company for Scanning Laser Ophthalmoscope devices based in Edinburgh, Scotland.

  • [Jan 2015]
    I will start lab tutoring activities with Programming Languages for Data Engineering. This laboratory aims at providing basics of data analysis using MATLAB.

  • [05/11/2014]
    Transfer of Ordinance viva passed successfully. Panel of examiners: Jano van Hemert (Optos plc), Marco Gaboardi (Dundee SoC fellow), Jens C. Brings (University of Copenhagen).

  • [24/09/2014]
    Brain Tumor Digital Pathology SEGMENTATION Challenge (MICCAI 2014): 2nd place.
    Preliminary version of our framework in "Brain Tumor Region Segmentation using Local Co-occurrence Features and Conditional Random Fields".

  • [08/09/2014]
    Our paper "Tortuosity Classification of Corneal Nerves Images Using a Multiple-Scale-Multiple-Window Approach" is NOW in the MICCAI, OMIA 2014 proceedings.

  • [03/09/2014]
    Our paper "An Experimental Assessment of Five Indices of Retinal Vessel Tortuosity with the RET-TORT Public Dataset" is NOW in the EMBC 2014 proceedings.

  • [03/09/2014]
    Proceedings of the SINAPSE conference are now available (posters PA23 and PD9) .

  • [11/08/2014]
    Invited review from Neurocomputing (Elsevier journal).

  • [29/07/2014]
    Poster presentation at MISS2014: you can download the booklet here (my poster is the number 3).

  • [06/07/2014]
    Conference paper acceptance:
    R. Annunziata, S. Aggarwal, A. Kheirkhah, B. M. Cavalcanti, P. Hamrah, E. Trucco,
    "Tortuosity Classification of Corneal Nerves Images Using a Multiple-Scale-Multiple-Window Approach".
    International Workshop on Ophthalmic Medical Image Analysis of MICCAI 2014 (Medical Image Computing and Computer Assisted Intervention) to be held in Boston, USA on September 14-18, 2014.

  • [03/07/2014]
    Our conference paper "An Experimental Assessment of Five Indices of Retinal Vessel Tortuosity with the RET-TORT Public Dataset" will be discussed at EMBC 2014 as oral presentation.

  • [18/06/2014]
    REVAMMAD (MARIE CURIE ITN project) Machine Learning Challenge on lesion detection WON during the training event held between 7-16 June in Crete. (13 Marie Curie fellows involved)

  • [02/06/2014]
    Conference paper acceptance:
    A. Lisowska, R. Annunziata, E. Trucco, D. Karl, G.K. Loh,
    "An Experimental Assessment of Five Indices of Retinal Vessel Tortuosity with the RET-TORT Public Dataset".
    36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC 2014) to be held in Sheraton Hotel & Towers, Chicago, Illinois, USA on August 26-30, 2014.

  • [22/05/2014]
    Acceptance to SINAPSE conference: poster presentation (first author) and co-author of another poster.
    16 Jun 2014, Edinburgh, UK.

  • [15/04/2014]
    Acceptance to Medical Image Analysis Summer School 2014 (MISS2014).
    The summer school is endorsed by the MICCAI society and Springer and it is sponsored by Toshiba and GIRPR.
    27 July - 2 Aug 2014 Favignana, Sicily (IT).