IMT Institutional Repository: No conditions. Results ordered -Date Deposited. 2024-03-19T13:38:10ZEPrintshttp://eprints.imtlucca.it/images/logowhite.pnghttp://eprints.imtlucca.it/2017-02-01T08:49:17Z2017-02-01T08:49:17Zhttp://eprints.imtlucca.it/id/eprint/3652This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/36522017-02-01T08:49:17ZStatistical shape modeling of the left ventricle: myocardial infarct classification challengeStatistical shape modeling is a powerful tool for visualizing and quantifying geometric and functional patterns of the heart. After myocardial infarction (MI), the left ventricle typically remodels in response to physiological challenges. Several methods have been proposed in the literature to describe statistical shape changes. Which method best characterizes left ventricular remodeling after MI is an open research question. A better descriptor of remodeling is expected to provide a more accurate evaluation of disease status in MI patients. We therefore designed a challenge to test shape characterization in MI given a set of three-dimensional left ventricular surface points. The training set comprised 100 MI patients, and 100 asymptomatic volunteers (AV). The challenge was initiated in 2015 at the Statistical Atlases and Computational Models of the Heart workshop, in conjunction with the MICCAI conference. The training set with labels was provided to participants, who were asked to submit the likelihood of MI from a different (validation) set of 200 cases (100 AV and 100 MI). Sensitivity, specificity, accuracy and area under the receiver operating characteristic curve were used as the outcome measures. The goals of this challenge were to (1) establish a common dataset for evaluating statistical shape modeling algorithms in MI, and (2) test whether statistical shape modeling provides additional information characterizing MI patients over standard clinical measures. Eleven groups with a wide variety of classification and feature extraction approaches participated in this challenge. All methods achieved excellent classification results with accuracy ranges from 0.83 to 0.98. The areas under the receiver operating characteristic curves were all above 0.90. Four methods showed significantly higher performance than standard clinical measures. The dataset and software for evaluation are available from the Cardiac Atlas Project website1.A. SuinesiaputraP. AblinX. AlbaM. AlessandriniJ. AllenW. BaiS. CimenP. ClaesB. R. CowanJ. D'hoogeN. DuchateauJ. EhrhardtA. F. FrangiA. GooyaV. GrauK. LekadirA. LuA. MukhopadhyayIlkay Oksuzilkay.oksuz@imtlucca.itX. PennecM. PereanezC. PintoP. PirasM. M. RoheD. RueckertM. SermesantK. SiddiqiM. TabassianL. TeresiS. A. TsaftarisM. WilmsA. A. YoungX. ZhangP. Medrano-Gracia2017-02-01T08:36:30Z2017-02-01T08:36:30Zhttp://eprints.imtlucca.it/id/eprint/3651This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/36512017-02-01T08:36:30ZMRI-TRUS Image Synthesis with Application to Image-Guided Prostate InterventionAccurate and robust fusion of pre-procedure magnetic resonance imaging (MRI) to intra-procedure trans-rectal ultrasound (TRUS) imaging is necessary for image-guided prostate cancer biopsy procedures. The current clinical standard for image fusion relies on non-rigid surface-based registration between semi-automatically segmented prostate surfaces in both the MRI and TRUS. This surface-based registration method does not take advantage of internal anatomical prostate structures, which have the potential to provide useful information for image registration. However, non-rigid, multi-modal intensity-based MRI-TRUS registration is challenging due to highly non-linear intensities relationships between MRI and TRUS. In this paper, we present preliminary work using image synthesis to cast this problem into a mono-modal registration task by using a large database of over 100 clinical MRI-TRUS image pairs to learn a joint model of MR-TRUS appearance. Thus, given an MRI, we use this learned joint appearance model to synthesize the patient’s corresponding TRUS image appearance with which we could potentially perform mono-modal intensity-based registration. We present preliminary results of this approach.John A. OnofreyIlkay Oksuzilkay.oksuz@imtlucca.itSaradwata SarkarRajesh VenkataramanLawrence H. StaibXenophon Papademetris2016-03-21T10:45:55Z2016-04-06T07:44:40Zhttp://eprints.imtlucca.it/id/eprint/3257This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/32572016-03-21T10:45:55ZControlled Release of the Anti-cancer Drug Paclitaxel from Bioresorbable Poly(ester-ether-ester) MicrospheresThe release of the anti-cancer drug paclitaxel (PTX) from microspheres of the bioresorbable poly(ε-caprolactoneoxyethylene-ε-caprolactone)tri-block copolymer was studied. The microspheres, both loaded and not with PTX, were prepared by emulsion-evaporation technique, then characterized by SEM, AFM, total reflection and spotlight FT-IR spectroscopy, and DSC. The quantities of PTX released were measured by HPLC. The results showed a slow and very regular release, which fits very well the Peppas equation, Mt/M∞ = k · t, where Mt is the amount of solute released at the time t, M∞ is the amount of drug released at the plateau condition, k represents the Peppas kinetic constant and n the diffusion order.Giulio D. GuerraMariacristina Gagliardimariacristina.gagliardi@imtlucca.itNiccoletta BarbaniCaterina Cristallini2016-03-21T09:35:11Z2016-03-21T09:35:11Zhttp://eprints.imtlucca.it/id/eprint/3250This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/32502016-03-21T09:35:11ZThe effect of bioartificial constructs that mimic myocardial structure and biomechanical properties on stem cell commitment towards cardiac lineageAbstract Despite the enormous progress in the treatment of coronary artery diseases, they remain the most common cause of heart failure in the Western countries. New translational therapeutic approaches explore cardiomyogenic differentiation of various types of stem cells in combination with tissue-engineered scaffolds. In this study we fabricated PHBHV/gelatin constructs mimicking myocardial structural properties. Chemical structure and molecular interaction between material components induced specific properties to the substrate in terms of hydrophilicity degree, porosity and mechanical characteristics. Viability and proliferation assays demonstrated that these constructs allow adhesion and growth of mesenchymal stem cells (MSCs) and cardiac resident non myocytic cells (NMCs). Immunofluorescence analysis demonstrated that stem cells cultured on these constructs adopt a distribution mimicking the three-dimensional cell alignment of myocardium. qPCR and immunofluorescence analyses showed the ability of this construct to direct initial {MSC} and {NMC} lineage specification towards cardiomyogenesis: both {MSCs} and {NMCs} showed the expression of the cardiac transcription factor GATA-4, fundamental for early cardiac commitment. Moreover {NMCs} also acquired the expression of the cardiac transcription factors Nkx2.5 and {TBX5} and produced sarcomeric proteins. This work may represent a new approach to induce both resident and non-resident stem cells to cardiac commitment in a 3-D structure, without using additional stimuli.Caterina CristalliniElisa Cibrario RocchiettiLisa AccomassoAnna FolinoClara GallinaLuisa MuratoriPasquale PagliaroRaffaella RastaldoStefania RaimondoSilvia SaviozziAndrea E. SprioMariacristina Gagliardimariacristina.gagliardi@imtlucca.itNiccoletta BarbaniClaudia Giachino2016-03-21T08:41:12Z2016-03-21T08:41:12Zhttp://eprints.imtlucca.it/id/eprint/3240This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/32402016-03-21T08:41:12ZParcellation-based connectome assessment by using structural and functional connectivityConnectome analysis of the human brain structural and functional architecture provides a unique opportunity to understand the organization of brain networks. In this work, we investigate a novel large scale parcellation-based connectome, merging together information coming from resting state fMRI (rs-fMRI) data and diffusion tensor imaging (DTI) measurements.Ying-Chia Linyingchia.lin@imtlucca.itTommaso GiliSotirios A. TsaftarisAndrea GabrielliMariangela IorioGianfranco SpallettaGuido Caldarelliguido.caldarelli@imtlucca.it2016-03-21T08:41:04Z2016-03-21T08:41:04Zhttp://eprints.imtlucca.it/id/eprint/3239This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/32392016-03-21T08:41:04ZA cortical and sub-cortical parcellation clustering by intrinsic functional connectivityNetwork analysis of resting-state fMRI (rsfMRI) has been widely utilized to investigate the functional architecture of the whole brain. Here we propose a robust parcellation method that first divides cortical and sub-cortical regions into sub-regions by clustering the rsfMRI data for each subject independently, and then merges those individual parcellations to obtain a global whole brain parcellation. To do so our method relies on majority voting (to merge parcellations of multiple subjects) and enforces spatial constraints within a hierarchical agglomerative clustering framework to define parcels that are spatially homogeneous.Ying-Chia Linyingchia.lin@imtlucca.itTommaso GiliSotirios A. TsaftarisAndrea GabrielliMariangela IorioGianfranco SpallettaGuido Caldarelliguido.caldarelli@imtlucca.it2016-03-14T13:02:02Z2016-04-06T10:06:19Zhttp://eprints.imtlucca.it/id/eprint/3223This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/32232016-03-14T13:02:02ZA Cortical and Sub-cortical Parcellation Clustering by Intrinsic Functional ConnectivityNetwork analysis of resting-state fMRI (rsfMRI) has been widely utilized to investigate the functional architecture of the whole brain. Such analysis can divide the brain into several discrete elements (nodes) connected by links (edges) representing the relation between two elements. The brain cortical and subcortical areas can be segmented or parcelled into several functional and/or structural regions. The connectome analysis of human-brain structure and functional connectivity provides a unique opportunity to understand the organisation of brain networks. However, such analyses require an appropriate definition of functional or structural nodes to efficiently represent cortical regions. In order to address this issue, here we propose a robust parcellation method based on resting-state fMRI, which can be generalized from the single-subject level to the multi-group one. Considering the input data of a single subject and constructing multi-resolution graph elements. We combined voting-based measurements to divide the cortical region into sub-regions in order to obtain the whole brain parcellation. Our parcellation relies on majority vote and poses spatial constraints within a hierarchical agglomerative clustering framework to define parcels that are spatially homogeneous. We used rsfMRI data collected from 40 healthy subjects and we showed that our purposed algorithm is able to compute stable and reproducible parcellations across the group of subjects at multi-resolution level. We find that, even though previous methods ensure on average larger overlap between parcels and regions in AAL atlas, the method proposed herein reduces inter-subject variability, especially when the number of parcels increases. Our high-resolution parcels seem to be functionally more consistent and reliable and can be a useful tool for future analysis that will aim to match functional and structural architecture of the brain.Ying-Chia Linyingchia.lin@imtlucca.itTommaso GiliSotirios A. Tsaftarissotirios.tsaftaris@imtlucca.itAndrea GabrielliMariangela IorioGianfranco SpallettaGuido Caldarelliguido.caldarelli@imtlucca.it2016-01-20T10:27:26Z2016-04-06T10:06:49Zhttp://eprints.imtlucca.it/id/eprint/3025This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/30252016-01-20T10:27:26ZSupervised Learning of Functional Maps for Infarct ClassificationOur submission to the STACOM Challenge at MICCAI 2015 is based on the supervised learning of functional map representation between End Systole (ES) and End Diastole (ED) phases of Left Ventricle (LV), for classifying infarcted LV from the healthy ones. The Laplace-Beltrami eigen-spectrum of the LV surfaces at ES and ED, represented by their triangular meshes, are used to compute the functional maps. Multi-scale distortions induced by the mapping, are further calculated by singular value decomposition of the functional map. During training, the information of whether an LV surface is healthy or diseased is known, and this information is used to train an SVM classifier for the singular values at multiple scales corresponding to the distorted areas augmented with surface area difference of epicardium and endocardium meshes. At testing similar augmented features are calculated and fed to the SVM model for classification. Promising results are obtained on both cross validation of training data as well as on testing data, which encourages us in believing that this algorithm will perform favourably in comparison to state of the art methods.Anirban MukhopadhyayIlkay Oksuzilkay.oksuz@imtlucca.itSotirios A. Tsaftarissotirios.tsaftaris@imtlucca.it2015-11-24T13:00:34Z2015-11-24T13:00:34Zhttp://eprints.imtlucca.it/id/eprint/2931This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/29312015-11-24T13:00:34ZThe eNanoMapper database for nanomaterial safety informationBackground: The NanoSafety Cluster, a cluster of projects funded by the European Commision, identified the need for a computational infrastructure for toxicological data management of engineered nanomaterials (ENMs). Ontologies, open standards, and interoperable designs were envisioned to empower a harmonized approach to European research in nanotechnology. This setting provides a number of opportunities and challenges in the representation of nanomaterials data and the integration of ENM information originating from diverse systems. Within this cluster, eNanoMapper works towards supporting the collaborative safety assessment for ENMs by creating a modular and extensible infrastructure for data sharing, data analysis, and building computational toxicology models for ENMs.
Results: The eNanoMapper database solution builds on the previous experience of the consortium partners in supporting diverse data through flexible data storage, open source components and web services. We have recently described the design of the eNanoMapper prototype database along with a summary of challenges in the representation of ENM data and an extensive review of existing nano-related data models, databases, and nanomaterials-related entries in chemical and toxicogenomic databases. This paper continues with a focus on the database functionality exposed through its application programming interface (API), and its use in visualisation and modelling. Considering the preferred community practice of using spreadsheet templates, we developed a configurable spreadsheet parser facilitating user friendly data preparation and data upload. We further present a web application able to retrieve the experimental data via the API and analyze it with multiple data preprocessing and machine learning algorithms.
Conclusion: We demonstrate how the eNanoMapper database is used to import and publish online ENM and assay data from several data sources, how the “representational state transfer” (REST) API enables building user friendly interfaces and graphical summaries of the data, and how these resources facilitate the modelling of reproducible quantitative structure–activity relationships for nanomaterials (NanoQSAR).Nina JeliazkovaHaralambos ChomenidesPhilip DoganisBengt FadeelRoland GrafströmBarry HardyJanna HastingsMarkus HegiVedrin JeliazkovNikolay KochevPekka KohonenCristian MunteanuHaralambos SarimveisBart SmeetsPantelis Sopasakispantelis.sopasakis@imtlucca.itGeorgia TsilikiDavid VorgrimmlerEgon Willighagen2015-10-19T09:40:53Z2016-04-06T08:50:40Zhttp://eprints.imtlucca.it/id/eprint/2776This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/27762015-10-19T09:40:53ZBinary and Multi-class Parkinsonian Disorders Classification Using Support Vector MachinesThis paper presents a method for an automated Parkinsonian disorders classification using Support Vector Machines (SVMs). Magnetic Resonance quantitative markers are used as features to train SVMs with the aim of automatically diagnosing patients with different Parkinsonian disorders. Binary and multi–class classification problems are investigated and applied with the aim of automatically distinguishing the subjects with different forms of disorders. A ranking feature selection method is also used as a preprocessing step in order to asses the significance of the different features in diagnosing Parkinsonian disorders. In particular, it turns out that the features selected as the most meaningful ones reflect the opinions of the clinicians as the most important markers in the diagnosis of these disorders. Concerning the results achieved in the classification phase, they are promising; in the two multi–class classification problems investigated, an average accuracy of 81% and 90% is obtained, while in the binary scenarios taken in consideration, the accuracy is never less than 88%.Rita Morisirita.morisi@imtlucca.itGiorgio Gneccogiorgio.gnecco@imtlucca.itNico LanconelliStefano ZanigniDavid Neil MannersClaudia TestaStefania EvangelistiLauraLudovica GramegnaClaudio BianchiniPietro CortelliCaterina TononRaffaele Lodi2015-10-19T09:31:34Z2015-10-19T09:31:34Zhttp://eprints.imtlucca.it/id/eprint/2775This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/27752015-10-19T09:31:34ZSemi-automated scar detection in delayed enhanced cardiac magnetic resonance imagesLate enhancement cardiac magnetic resonance images (MRI) has the ability to precisely delineate myocardial scars. We present a semi-automated method for detecting scars in cardiac MRI. This model has the potential to improve routine clinical practice since quantification is not currently offered due to time constraints. A first segmentation step was developed for extracting the target regions for potential scar and determining pre-candidate objects. Pattern recognition methods are then applied to the segmented images in order to detect the position of the myocardial scar. The database of late gadolinium enhancement (LE) cardiac MR images consists of 111 blocks of images acquired from 63 patients at the University Hospital Southampton NHS Foundation Trust (UK). At least one scar was present for each patient, and all the scars were manually annotated by an expert. A group of images (around one third of the entire set) was used for training the system which was subsequently tested on all the remaining images. Four different classifiers were trained (Support Vector Machine (SVM), k-nearest neighbor (KNN), Bayesian and feed-forward neural network) and their performance was evaluated by using Free response Receiver Operating Characteristic (FROC) analysis. Feature selection was implemented for analyzing the importance of the various features. The segmentation method proposed allowed the region affected by the scar to be extracted correctly in 96% of the blocks of images. The SVM was shown to be the best classifier for our task, and our system reached an overall sensitivity of 80% with less than 7 false positives per patient. The method we present provides an effective tool for detection of scars on cardiac MRI. This may be of value in clinical practice by permitting routine reporting of scar quantification.Rita Morisirita.morisi@imtlucca.itBruno DoniniNico LanconelliJames RosengardenJohn MorganStephen HardenNick Curzen2013-10-25T08:37:29Z2013-10-25T08:37:29Zhttp://eprints.imtlucca.it/id/eprint/1846This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/18462013-10-25T08:37:29ZJAQPOT RESTful Web Services: An Implementation of the OpenTox Application Programming Interface for On-line Prediction of Toxicological PropertiesPantelis Sopasakispantelis.sopasakis@imtlucca.itHaralambos Sarimveis2013-10-25T08:36:07Z2013-10-25T08:36:07Zhttp://eprints.imtlucca.it/id/eprint/1845This item is in the repository with the URL: http://eprints.imtlucca.it/id/eprint/18452013-10-25T08:36:07ZToxOtis: A Java Interface to the OpenTox Predictive Toxicology NetworkPantelis Sopasakispantelis.sopasakis@imtlucca.itHaralambos Sarimveis