Browsing by Keyword "computer vision"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item Automatic 3D model-based method for liver segmentation in MRI based on Active Contours and Total Variation minimization(2015-07-01) Bereciartua, Arantza; Picon, Artzai; Galdran, Adrian; Iriondo, Pedro M.; COMPUTER_VISION; Tecnalia Research & InnovationLiver cancer is one of the leading causes of cancer-related mortality worldwide. Non-invasive techniques of medical imaging such as Computerized Tomography (CT) and Magnetic Resonance Imaging (MRI) are often used by radiologists for diagnosis and surgery planning. With the aim of assuring the most reliable intervention planning to surgeons, new accurate methods and tools must be provided to locate and segment the regions of interest. Automated liver segmentation is a challenging problem for which promising results have been achieved mostly for CT. However, MRI is required by radiologists, since it offers better information for diagnosis purposes. MRI liver segmentation represents a challenge due to the presence of characteristic artifacts, such as partial volumes, noise, low contrast and poorly defined edges of the liver in relation to adjacent organs. In this paper, we present a method for MRI automatic 3D liver segmentation by means of an Active Contour model extended to 3D and minimized by Total Variation dual approach that has also been extended to 3D. A new approach to enhance the contrast in the input MRI image is proposed and it allows more accurate segmentation. The proposed methodology allows replacing the input image by a probability map obtained by means of a previously generated statistical model of the liver. An Accuracy of 98.89 and Dice Similarity Coefficient of 90.19 are in line with other state-of-the-art methodologies.Item A new tool to search tumor samples for research across Europe: BIOPOOL - Poster(Springer, 2015) Sáiz, Mónica; Belar, Oihana; de Jong, Bas; Kap, Marcel; Bereciartua, Arantza; Ruiz, Rebeca; Viguri, Maria Amparo; Rezola, Ricardo; de Miguel, Eduardo; Saiz, Alberto; Fernández, Sara; Gaafar, Ayman; Catón, Blanca; Aguirre, Javier; Doukas, Michael; Muñoz, Elena; Gandon, Fabienne; Riegman, Peter; Bilbao, RobertoObjective: Searching tissues for research across biobanks and pathology departments is complicated due to the geographically dispersed distribution, diagnosis heterogeneity, language diversity and lack of online robust samples catalogues. BIOPOOL consortium (www.biopoolproject.eu) was created to give a solution based on a new search approach similar to Google images but focused on histological images and associated clinical databases. The project is funded by the 7FP of the European Commission (GA296162). Method: Digital images from colon, breast and lung cancer were used to develop the software. Eleven pathologists worked closely with IT developers. They defined the technical and functional requirements of the final system, and the identification and validation of the key visual features and regions of interest on the histological images and their associated data. Relevant visual descriptors were identified, coded and extracted automatically from the images in order to be included in the searching tool and retrieved later in the case it matches the query. Results: A new web-based search portal was developed to find tumor samples across biobanks based on query by image and/or text. The legal and ethical issues were taken in consideration. Conclusion: BIOPOOL network, is now open to biobanks and pathology departments to bring together, in a European scale, a research infrastructure that could help to conduct research through the synergy of medical knowledge and technology.Item Services Associated to Digitalised Contents of Tissues in Biobanks Across Europe: A Proof of Concept – BIOPOOL(ISBER Annual Meeting, 2013-05) de Jong, Bas; Belar, Oihana; Bereciartua, Arantza; Picon, Artzai; Muñoz, Elena; Sevilla, D.; Moscone, F.; Gandon, Fabienne; Tosseti, E.; García, S.; Riegman, Peter; Bilbao, RobertoBackground: Pathology departments and biobanks are increasingly using Digital Pathology (DP) images for sharing of research results, ring trials, education, fast second-opinion diagnostics, pathology panels, digital back-up of slides, image analysis algorithms, and etcetera. To fully exploit the potential of DP, the BIOPOOL project develops software for extracting and gathering DP slides with well defined associated data from multiple biobanks and pathology archives to create pools of images, as biobanks networks, on which clinicians and researchers can search for reference, score for similarities with their own images using an innovative Content Based Image Retrieval system, and perform indepth image analyses. Methods: The BIOPOOL Proof-of-Concept (PoC) with minimal, critical functionality serves as the basis on which the system will be further developed. For this PoC we are studying existing DP image formats and systems that could be of use, designed both PoC and end-phase validation plans and end-phase functional requirements. Results: For the PoC, only colon DP slides with associated data (normal and high grade carcinoma), digitalised on Hamamatsu and Olympus scanners, are used. Pathologists have assigned morphological areas of interest for image searching development and creation of the basic DPpool, which were both validated. Functional requirements include a user-interface for searching on textual and morphology aspects, multiscanner format support, storage capacity, computational power for search processing and IT equipment and support. Conclusions: The PoC model is a template for expanding the BIOPOOL system to full functionality. After final validation BIOPOOL may then serve as a leading example for using the full potential of DP imaging.