Content-Based Image Retrieval
ParaDISE (Parallel Distributed Image Search Engine) is an image retrieval engine developed by the medGIFT research group as a replacement of the GIFT search engine in the context of the KHRESMOI project. The main concepts behind its design are scalability, flexibility, expandability and interoperability, allowing it to be used in standalone applications, integrated systems and for research purposes.
ParaDISE can be found at http://paradise.khresmoi.eu/
The research stay of Antonio Foncubierta at Martinos Center of Harvard Medical School finishes this week.
He’ll return to medGIFT offices in Sierre for his Thesis Defense in September.
Three papers submitted by the medGIFT group were accepted at the Swiss Society for Medical Informatics 2014 (SGMI 2014). The presentations will be held on the 8th and 9th September 2014 at Bern, Switzerland.
The accepted papers are:
– “Crowdsourcing for Medical Image Classification” by Alba García Seco de Herrera, Antonio Foncubierta-Rodríguez, Dimitrios Markonis, Roger Schaer, Henning Müller.
– “A Modern Web Interface for Medical Image Retrieval” by Roger Schaer and Henning Müller.
– ” Using Google Glass to enhance pre-hospital care” by Antoine Widmer and Henning Müller.
The medGIFT group had a paper presentation at the 37th annual international ACM Special Interest Group On Informaition Retrieval 2014 (ACM SIGIR 2014) conference. The conference took place at Queensland, Australia from the 6-11 July 2014.
The paper “Multi-modal relevance feedback for medical image retrieval” by Dimitrios Markonis, Roger Schaer and Henning Müller was presented as part of the Medical Information Retrieval (MedIR) workshop.
MedGIFT group won the 1st place at the International Conference on Biomedical Health Informatics (BHI) 2014 : Student Best Rapid Fire Session. The paper ‘Multi Atlas-Based Segmentation with an Intensity Feature Refinement’ by Oscar A. Jiménez del Toro and Henning Müller was presented during the session ‘Big Data in Healthcare’ at the BHI 2014.
The Student Best Rapid Fire Session is sponsored by the IEEE EMBS Biomedical and Health Informatics Technical Committee and IEEE EMBS Journal of Biomedical and Health Informatics.
The paper ‘Multi Atlas–Based Segmentation with an Intensity Feature Refinement’ by Jiménez del Toro et. al. will be presented at the 2014 IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI).
The presentation will take place during the Special Session ‘Big Data in Healthcare’, June 3, 2014.
The medGIFT group submitted three papers to the Swiss Society for Medical Informatics Annual Meeting 2014.
The presentations ‘Anatomical correlations for a hierarchical multi-atlas segmentation of the pancreas in CT images’ and ‘Anatomical correlations for a hierarchical multi-atlas segmentation of CT images’ by Oscar Jiménez and Henning Müller were presented at the IEEE International Symposium on Biomedical Imaging (ISBI) 2014.
The presentations were part of the ISBI challenges ‘Pancreas segmentation in Abdominal CT’ and VISCERAL ISBI challenge. The symposium took place from 29 April to 2 May in Beijing, China.
A draft version of the submitted paper can be accessed through our publications website.
Digital health, or e-health, is one of the major issues of the 21st century. The second E-health Day on 6 June 2014 will be a chance to discover both Swiss and international expertise in this field. This expertise is found in applied research as well as in young companies and well-established SMEs.
The event, drawing above all on real-life case studies, will give all those attending a chance to gain a relatively exhaustive overview of the issues and the main players in the field, as well as the challenges that await them.
This day is also intended to be an exchange and networking opportunity.
We look forward to welcoming you!
The book ‘Medical Computer Vision. Large Data in Medical Imaging’ is now available in Springer. You can find the online version at http://link.springer.com/openurl.asp?genre=issue&issn=0302-9743&volume=8331
This book constitutes the thoroughly refereed post-workshop proceedings of the Third International Workshop on Medical Computer Vision, MCV 2013, held in Nagoya, Japan, in September 2013 in conjunction with the 16th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2013. The 7 revised full papers and 12 poster papers presented were selected from 25 submissions. They have been organized in topical sections on registration and visualization, segmentation, detection and localization, and features and retrieval. In addition, the volume contains two invited papers describing segmentation task and data set of the VISCERAL benchmark challenge.
The results of the Khresmoi project were demonstrated to Commissioner Neelie Kroes, Vice-President of the European Commision, at the European Data Forum in Athens on the 19th of March 2014.
The latest video for the project KHRESMOI is now available.
Go to the link in YouTube to watch it: https://www.youtube.com/watch?v=7Ve2VUGH9Tk&feature=youtu.be
The KHRESMOI project was presented at the CeBIT 2014 in Hannover, Germany.
More information about the project and the presentation at CeBIT, the world’s largest fair for information technology, is also mentioned in the MUW news website 
Yashin Dicente Cid just joined us as a PhD candidate on lung hypertension image analysis. Welcome to him!
The new Shambala interface is now publicly available at http://faster.hevs.ch:8080/shambala/
It includes image manipulation, modality classification and compound figure separation. Feel free to go ahead and try it.
The poster ‘Finding seed points for organ segmentation using example annotations’ by Ranveer Joyseeree and Henning Müller was presented at SPIE 2014, California, USA.
The matlab code used in the TIP 2013 paper ‘Rotation-Covariant Texture Learning Using Steerable Riesz Wavelets’ is now available. The code and the paper can be found in http://publications.hevs.ch/index.php/publications/show/1373
The paper is written by Adrien Depeursinge, Antonio Foncubierta-Rodriguez, Dimitri Van De Ville and Henning Müller.
Abstract—We propose a texture learning approach that exploits
local organizations of scales and directions. First, linear
combinations of Riesz wavelets are learned using kernel support
vector machines. The resulting texture “signatures” are modeling
optimal class–wise discriminatory properties. The visualization
of the obtained signatures allows verifying the visual relevance
of the learned concepts. Second, the local orientations of the
signatures are optimized to maximize their responses, which
is carried out analytically and can still be expressed as a
linear combination of the initial steerable Riesz templates. The
global process is iteratively repeated to obtain final rotation–
covariant texture signatures. Rapid convergence of class–wise
signatures is observed, which demonstrates that the instances
are projected into a feature space that leverages the local
organizations of scales and directions. Experimental evaluation
reveals an average classification accuracies in the range of 97%
to 98% for the Outex TC 00010, the Outex TC 00012, and the
Contrib TC 00000 suite for even orders of the Riesz transform,
and suggests high robustness to changes in images orientation
and illumination. The proposed framework requires no arbitrary
choices of scales and directions and is expected to perform well
in a large range of computer vision applications.
The chapter ‘Fusion Techniques in Biomedical Information Retrieval’ by Alba García Seco de Herrera and Henning Müller was included in the book ‘Fusion in Computer Vision -Understanding Complex Visual Content in Springer Advances in Computer Vision and Pattern Recognition’ Eds. Bogdan Ionescu, Jenny Benois-Pineau, Tomas Piatrik, Georges Quénot.
The book will be published by Springer.
The Year One review of the project VISCERAL took place in Luxembourg, January 22nd.
The VISCERAL project passed its first year EU review as excellent. For more info about the project go to http://www.visceral.eu