REVAMMAD Project (2013 - 2017)

The Retinal Vascular Modelling, Measurement and Diagnosis (REVAMMAD) project is an international research project from the European Commission. It involves researchers and institutions from 12 different countries. It is aimed at combating some of the EU's most prevalent chronic medical conditions using retinal imaging.

My involvement in this project mainly consists on employing diverse computer vision techniques, such as image registration or super resolution, to improve image quality as well as facilitating the analysis of retinal images by clinicians, allowing them to perform faster and more accurate diagnoses.

  1. Retinal Image Registration through 3D Eye Modelling and Pose Estimation
    C. Hernandez-Matas
    PhD thesis, University of Crete (Greece), 2017
  2. REMPE: Registration through Eye Modelling and Pose Estimation
    Novel solution to the problem of retinal image registration. We solve it via simultaneously estimating the relative pose of the cameras that acquired the images as well as the shape and the pose of the eye. The method utilizes an ellipsoidal model for the eye, and the pose of the cameras is estimated utilizing RANSAC, followed by a variant of Particle Swarm Optimization (PSO).
    URL: http://www.ics.forth.gr/cvrl/rempe/
  3. An Experimental Evaluation of the Accuracy of Keypoints-based Retinal Image Registration
    C. Hernandez-Matas, X. Zabulis, A.A. Argyros
    39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 377-381, Jeju Island, July 11-15, 2017
  4. FIRE: Fundus Image Registration Dataset
    C. Hernandez-Matas, X. Zabulis, A. Triantafyllou, P. Anyfanti, S. Douma, A.A. Argyros
    Journal for Modeling in Ophthalmology, vol. 1, no. 4, pp. 16-28, Jul. 2017.
  5. FIRE: Fundus Image Registration Dataset
    The dataset consists of 129 retinal images forming 134 image pairs. These image pairs are split into 3 different categories depending on their characteristics. The images were acquired with a Nidek AFC-210 fundus camera, which acquires images with a resolution of 2912x2912 pixels and a FOV of 45° both in the x and y dimensions. Images were acquired at the Papageorgiou Hospital, Aristotle University of Thessaloniki, Thessaloniki from 39 patients.
    URL: http://www.ics.forth.gr/cvrl/fire/
  6. Retinal Image Registration Through Simultaneous Camera Pose and Eye Shape Estimation
    C. Hernandez-Matas, X. Zabulis, A.A. Argyros
    38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 3247-3251, Orlando, August 16-20, 2016
    DOI: 10.1109/EMBC.2016.7591421
  7. Retinal Image Registration under the Assumption of a Spherical Eye
    C. Hernandez-Matas, X. Zabulis, A. Triantafyllou, P. Anyfanti, A.A. Argyros
    Computerized Medical Imaging and Graphics, Volume 55, January 2017, Pages 95-105
    DOI: 10.1016/j.compmedimag.2016.06.006
  8. In Vivo Analysis of the Time and Spatial Activation Pattern of Microglia in the Retina Following Laser-Induced Choroidal Neovascularization
    S. Crespo-Garcia, N. Reichhart, C. Hernandez-Matas, X. Zabulis, N. Kociok, C. Brockmann, A.M. Joussen, O. Strauß
    Experimental Eye Research, Volume 139, October 2015, Pages 13-21
    DOI: 10.1016/j.exer.2015.07.012
  9. Retinal Image Registration Based on Keypoint Correspondences, Spherical Eye Modeling and Camera Pose Estimation
    C. Hernandez-Matas, X. Zabulis, A.A. Argyros
    37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 5650-5654, Milan, August 25-29, 2015
    DOI: 10.1109/EMBC.2015.7319674
  10. Image Processing Allows Reliable In Vivo Detection and Quantification of Retinal Microglia Following Laser-Induced Choroidal Neovascularization
    S. Crespo-Garcia, N. Reichart, C. Hernandez-Matas, X. Zabulis, N. Kociok, A.M. Joussen, O. Strauss
    Investigative Ophthalmology & Visual Science, vol. 56, issue 7, pp. 425, 2015
  11. Super Resolution for Fundoscopy Based on 3D Image Registration
    C. Hernandez-Matas, X. Zabulis
    36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 6332-6338, Chicago, August 26-30, 2014
    DOI: 10.1109/EMBC.2014.6945077

Back to top

3D Image Segmentation (2011)

A study about the capabilities of applying the depth maps generated by Kinect to image segmentation. Specifically, it focus on the provision of an alternative attention mechanism to select fixations points for the segmentation, as well as two sets of cues for the object-background separation.

Developed in C++.

This project was developed as a Master Thesis for the Royal Institute of Technology.

  1. Active 3D Scene Segmentation using Kinect
    C. Hernandez-Matas
    MSc thesis, Royal Institute of Technology (Sweden), 2011

Back to top

Sound Ray Tracer (2008)

Given the size of an orthogonal room, the reflection coefficient of its walls, the location of a source of sound and the location of a listening device, this program calculates the normalized impulse response between those 2 points by taking into account all the reflections of the sound in the walls.

Developed in MATLAB.

This project was developed in collaboration with Diana Gómez Olmedilla as a Bachelor Thesis for the Blekinge Tekniska Högskolan and the Technical University of Madrid.

  1. Image Theory Applied to Virtual Microphones
    C. Hernandez-Matas, D. Gomez Olmedilla
    BSc thesis, Blekinge Institute of Technology (Sweden) / Technical University of Madrid (Spain), 2008

Back to top