toggle visibility Search & Display Options

Select All    Deselect All
 |   | 
Details
  Records Links
Author Debora Gil; Jaume Garcia; Ruth Aris; Guillaume Houzeaux; Manuel Vazquez edit   pdf
openurl 
  Title A Riemmanian approach to cardiac fiber architecture modelling Type Conference Article
  Year 2009 Publication 1st International Conference on Mathematical & Computational Biomedical Engineering Abbreviated Journal  
  Volume Issue Pages 59-62  
  Keywords cardiac fiber architecture; diffusion tensor magnetic resonance imaging; differential (Rie- mannian) geometry.  
  Abstract (down) There is general consensus that myocardial fiber architecture should be modelled in order to fully understand the electromechanical properties of the Left Ventricle (LV). Diffusion Tensor magnetic resonance Imaging (DTI) is the reference image modality for rapid measurement of fiber orientations by means of the tensor principal eigenvectors. In this work, we present a mathematical framework for across subject comparison of the local geometry of the LV anatomy including the fiber architecture from the statistical analysis of DTI studies. We use concepts of differential geometry for defining a parametric domain suitable for statistical analysis of a low number of samples. We use Riemannian metrics to define a consistent computation of DTI principal eigenvector modes of variation. Our framework has been applied to build an atlas of the LV fiber architecture from 7 DTI normal canine hearts.  
  Address  
  Corporate Author Thesis  
  Publisher Place of Publication Swansea (UK) Editor Nithiarasu, R.L.R.V.L.  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN ISBN Medium  
  Area Expedition Conference CMBE  
  Notes IAM Approved no  
  Call Number IAM @ iam @ FGA2009 Serial 1520  
Permanent link to this record
 

 
Author Esmitt Ramirez; Carles Sanchez; Debora Gil edit   pdf
url  doi
openurl 
  Title Localizing Pulmonary Lesions Using Fuzzy Deep Learning Type Conference Article
  Year 2019 Publication 21st International Symposium on Symbolic and Numeric Algorithms for Scientific Computing Abbreviated Journal  
  Volume Issue Pages 290-294  
  Keywords  
  Abstract (down) The usage of medical images is part of the clinical daily in several healthcare centers around the world. Particularly, Computer Tomography (CT) images are an important key in the early detection of suspicious lung lesions. The CT image exploration allows the detection of lung lesions before any invasive procedure (e.g. bronchoscopy, biopsy). The effective localization of lesions is performed using different image processing and computer vision techniques. Lately, the usage of deep learning models into medical imaging from detection to prediction shown that is a powerful tool for Computer-aided software. In this paper, we present an approach to localize pulmonary lung lesion using fuzzy deep learning. Our approach uses a simple convolutional neural network based using the LIDC-IDRI dataset. Each image is divided into patches associated a probability vector (fuzzy) according their belonging to anatomical structures on a CT. We showcase our approach as part of a full CAD system to exploration, planning, guiding and detection of pulmonary lesions.  
  Address Timisoara; Rumania; September 2019  
  Corporate Author Thesis  
  Publisher Place of Publication Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN ISBN Medium  
  Area Expedition Conference SYNASC  
  Notes IAM; 600.145; 600.140; 601.337; 601.323 Approved no  
  Call Number Admin @ si @ RSG2019 Serial 3531  
Permanent link to this record
 

 
Author Jose Elias Yauri; Aura Hernandez-Sabate; Pau Folch; Debora Gil edit  doi
openurl 
  Title Mental Workload Detection Based on EEG Analysis Type Conference Article
  Year 2021 Publication Artificial Intelligent Research and Development. Proceedings 23rd International Conference of the Catalan Association for Artificial Intelligence. Abbreviated Journal  
  Volume 339 Issue Pages 268-277  
  Keywords Cognitive states; Mental workload; EEG analysis; Neural Networks.  
  Abstract (down) The study of mental workload becomes essential for human work efficiency, health conditions and to avoid accidents, since workload compromises both performance and awareness. Although workload has been widely studied using several physiological measures, minimising the sensor network as much as possible remains both a challenge and a requirement.
Electroencephalogram (EEG) signals have shown a high correlation to specific cognitive and mental states like workload. However, there is not enough evidence in the literature to validate how well models generalize in case of new subjects performing tasks of a workload similar to the ones included during model’s training.
In this paper we propose a binary neural network to classify EEG features across different mental workloads. Two workloads, low and medium, are induced using two variants of the N-Back Test. The proposed model was validated in a dataset collected from 16 subjects and shown a high level of generalization capability: model reported an average recall of 81.81% in a leave-one-out subject evaluation.
 
  Address Virtual; October 20-22 2021  
  Corporate Author Thesis  
  Publisher Place of Publication Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN ISBN Medium  
  Area Expedition Conference CCIA  
  Notes IAM; 600.139; 600.118; 600.145 Approved no  
  Call Number Admin @ si @ Serial 3723  
Permanent link to this record
 

 
Author Gemma Sanchez; Josep Llados; Enric Marti edit  url
openurl 
  Title Segmentation and analysis of linial texture in plans Type Conference Article
  Year 1997 Publication Intelligence Artificielle et Complexité. Abbreviated Journal  
  Volume Issue Pages  
  Keywords Structural Texture, Voronoi, Hierarchical Clustering, String Matching.  
  Abstract (down) The problem of texture segmentation and interpretation is one of the main concerns in the field of document analysis. Graphical documents often contain areas characterized by a structural texture whose recognition allows both the document understanding, and its storage in a more compact way. In this work, we focus on structural linial textures of regular repetition contained in plan documents. Starting from an atributed graph which represents the vectorized input image, we develop a method to segment textured areas and recognize their placement rules. We wish to emphasize that the searched textures do not follow a predefined pattern. Minimal closed loops of the input graph are computed, and then hierarchically clustered. In this hierarchical clustering, a distance function between two closed loops is defined in terms of their areas difference and boundary resemblance computed by a string matching procedure. Finally it is noted that, when the texture consists of isolated primitive elements, the same method can be used after computing a Voronoi Tesselation of the input graph.  
  Address Paris, France  
  Corporate Author Thesis  
  Publisher Place of Publication Paris Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN ISBN Medium  
  Area Expedition Conference AERFAI  
  Notes DAG;IAM; Approved no  
  Call Number IAM @ iam @ SLM1997 Serial 1649  
Permanent link to this record
 

 
Author David Roche; Debora Gil; Jesus Giraldo edit   pdf
url  openurl
  Title Assessing agonist efficacy in an uncertain Em world Type Conference Article
  Year 2012 Publication 40th Keystone Symposia on mollecular and celular biology Abbreviated Journal  
  Volume Issue Pages 79  
  Keywords  
  Abstract (down) The operational model of agonism has been widely used for the analysis of agonist action since its formulation in 1983. The model includes the Em parameter, which is defined as the maximum response of the system. The methods for Em estimation provide Em values not significantly higher than the maximum responses achieved by full agonists. However, it has been found that that some classes of compounds as, for instance, superagonists and positive allosteric modulators can increase the full agonist maximum response, implying upper limits for Em and thereby posing doubts on the validity of Em estimates. Because of the correlation between Em and operational efficacy, τ, wrong Em estimates will yield wrong τ estimates.
In this presentation, the operational model of agonism and various methods for the simulation of allosteric modulation will be analyzed. Alternatives for curve fitting will be presented and discussed.
 
  Address Fairmont Banff Springs, Banff, Alberta, Canada  
  Corporate Author Keystone Symposia Thesis  
  Publisher Keystone Symposia Place of Publication Editor A. Christopoulus and M. Bouvier  
  Language english Summary Language english Original Title  
  Series Editor Keystone Symposia Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN ISBN Medium  
  Area Expedition Conference KSMCB  
  Notes IAM Approved no  
  Call Number IAM @ iam @ RGG2012 Serial 1855  
Permanent link to this record
 

 
Author Jaume Garcia; David Rotger; Francesc Carreras; R.Leta; Petia Radeva edit   pdf
doi  isbn
openurl 
  Title Contrast echography segmentation and tracking by trained deformable models Type Conference Article
  Year 2003 Publication Proc. Computers in Cardiology Abbreviated Journal  
  Volume 30 Issue Pages 173-176  
  Keywords  
  Abstract (down) The objective of this work is to segment the human left ventricle myocardium (LVM) in contrast echocardiography imaging and thus track it along a cardiac cycle in order to extract quantitative data about heart function. Ultrasound images are hard to work with due to their speckle appearance. To overcome this we report the combination of active contour models (ACM) or snakes and active shape models (ASM). The ability of ACM in giving closed and smooth curves in addition to the power of the ASM in producing shapes similar to the ones learned, evoke to a robust algorithm. Meanwhile the snake is attracted towards image main features, ASM acts as a correction factor. The algorithm was tested independently on 180 frames and satisfying results were obtained: in 95% the maximum difference between automatic and experts segmentation was less than 12 pixels.  
  Address  
  Corporate Author Thesis  
  Publisher Place of Publication Centre de Visió per Computador – Dept. Informàtica, UAB Edifici O – Campus UAB, 08193 Bellater Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN 0276-6547 ISBN 0-7803-8170-X Medium  
  Area Expedition Conference  
  Notes IAM;MILAB Approved no  
  Call Number IAM @ iam @ GRC2003 Serial 1512  
Permanent link to this record
 

 
Author Petia Radeva; Enric Marti edit   pdf
doi  openurl
  Title An improved model of snakes for model-based segmentation Type Conference Article
  Year 1995 Publication Proceedings of Computer Analysis of Images and Patterns Abbreviated Journal  
  Volume Issue Pages 515-520  
  Keywords  
  Abstract (down) The main advantage of segmentation by snakes consists in its ability to incorporate smoothness constraints on the detected shapes that can occur. Likewise, we propose to model snakes with other properties that reflect the information provided about the object of interest in a different extent. We consider different kinds of snakes, those searching for contours with a certain direction, those preserving an object’s model, those seeking for symmetry, those expanding open, etc. The availability of such a collection of snakes allows not only the more complete use of the knowledge about the segmented object, but also to solve some problems of the existing snakes. Our experiments on segmentation of facial features justify the usefulness of snakes with different properties.  
  Address  
  Corporate Author Thesis  
  Publisher Place of Publication Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN ISBN Medium  
  Area Expedition Conference CAIP  
  Notes MILAB;IAM Approved no  
  Call Number IAM @ iam @ RaM1995b Serial 1632  
Permanent link to this record
 

 
Author Paula Fritzsche; C.Roig; Ana Ripoll; Emilio Luque; Aura Hernandez-Sabate edit   pdf
doi  openurl
  Title A Performance Prediction Methodology for Data-dependent Parallel Applications Type Conference Article
  Year 2006 Publication Proceedings of the IEEE International Conference on Cluster Computing Abbreviated Journal  
  Volume Issue Pages 1-8  
  Keywords  
  Abstract (down) The increase in the use of parallel distributed architectures in order to solve large-scale scientific problems has generated the need for performance prediction for both deterministic applications and non-deterministic applications. In particular, the performance prediction of data dependent programs is an extremely challenging problem because for a specific issue the input datasets may cause different execution times. Generally, a parallel application is characterized as a collection of tasks and their interrelations. If the application is time-critical it is not enough to work with only one value per task, and consequently knowledge of the distribution of task execution times is crucial. The development of a new prediction methodology to estimate the performance of data-dependent parallel applications is the primary target of this study. This approach makes it possible to evaluate the parallel performance of an application without the need of implementation. A real data-dependent arterial structure detection application model is used to apply the methodology proposed. The predicted times obtained using the new methodology for genuine datasets are compared with predicted times that arise from using only one execution value per task. Finally, the experimental study shows that the new methodology generates more precise predictions.  
  Address  
  Corporate Author Thesis  
  Publisher Place of Publication Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN ISBN Medium  
  Area Expedition Conference  
  Notes IAM Approved no  
  Call Number IAM @ iam @ FRR2006 Serial 1497  
Permanent link to this record
 

 
Author Aura Hernandez-Sabate; Debora Gil; Petia Radeva; E.N.Nofrerias edit   pdf
doi  openurl
  Title Anisotropic processing of image structures for adventitia detection in intravascular ultrasound images Type Conference Article
  Year 2004 Publication Proc. Computers in Cardiology Abbreviated Journal  
  Volume 31 Issue Pages 229-232  
  Keywords  
  Abstract (down) The adventitia layer appears as a weak edge in IVUS images with a non-uniform grey level, which difficulties its detection. In order to enhance edges, we apply an anisotropic filter that homogenizes the grey level along the image significant structures (ridges, valleys and edges). A standard edge detector applied to the filtered image yields a set of candidate points prone to be unconnected. The final model is obtained by interpolating the former line segments along the tangent direction to the level curves of the filtered image with an anisotropic contour closing technique based on functional extension principles  
  Address  
  Corporate Author Thesis  
  Publisher Place of Publication Chicago (USA) Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN ISBN Medium  
  Area Expedition Conference  
  Notes IAM; MILAB Approved no  
  Call Number IAM @ iam @ HGR2004 Serial 1555  
Permanent link to this record
 

 
Author David Roche; Debora Gil; Jesus Giraldo edit   pdf
doi  isbn
openurl 
  Title Detecting loss of diversity for an efficient termination of EAs Type Conference Article
  Year 2013 Publication 15th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing Abbreviated Journal  
  Volume Issue Pages 561 - 566  
  Keywords EA termination; EA population diversity; EA steady state  
  Abstract (down) Termination of Evolutionary Algorithms (EA) at its steady state so that useless iterations are not performed is a main point for its efficient application to black-box problems. Many EA algorithms evolve while there is still diversity in their population and, thus, they could be terminated by analyzing the behavior some measures of EA population diversity. This paper presents a numeric approximation to steady states that can be used to detect the moment EA population has lost its diversity for EA termination. Our condition has been applied to 3 EA paradigms based on diversity and a selection of functions
covering the properties most relevant for EA convergence.
Experiments show that our condition works regardless of the search space dimension and function landscape.
 
  Address Timisoara; Rumania;  
  Corporate Author Thesis  
  Publisher Place of Publication Editor  
  Language Summary Language Original Title  
  Series Editor Series Title Abbreviated Series Title  
  Series Volume Series Issue Edition  
  ISSN ISBN 978-1-4799-3035-7 Medium  
  Area Expedition Conference SYNASC  
  Notes IAM; 600.044; 600.060; 605.203 Approved no  
  Call Number Admin @ si @ RGG2013c Serial 2299  
Permanent link to this record
Select All    Deselect All
 |   | 
Details

Save Citations:
Export Records: