490 Hits in 3.0 sec

High-performance numerical algorithms and software for subspace-based linear multivariable system identification

Vasile Sima, Diana Maria Sima, Sabine Van Huffel
2004 Journal of Computational and Applied Mathematics  
Basic algorithmic and numerical issues involved in subspace-based linear multivariable discrete-time system identiÿcation are described. A new identiÿcation toolbox-SLIDENT-has been developed and incorporated in the freely available Subroutine Library in Control Theory (SLICOT). Reliability, e ciency, and ability to solve industrial identiÿcation problems received a special consideration. Two algorithmic subspace-based approaches (MOESP and N4SID) and their combination, and both standard and
more » ... t techniques for data compression are provided. Structure exploiting algorithms and dedicated linear algebra tools enhance the computational e ciency and reliability. Extensive comparisons with the available computational tools based on subspace techniques show the better e ciency of the SLIDENT toolbox, at comparable numerical accuracy, and its capabilities to solve identiÿcation problems with many thousands of samples and hundreds of parameters.
doi:10.1016/ fatcat:yp55ybryrremlkg2lidvgd3vpu

An augmentation strategy to mimic multi-scanner variability in MRI [article]

Maria Ines Meyer, Ezequiel de la Rosa, Nuno Barros, Roberto Paolella, Koen Van Leemput, Diana M. Sima
2021 arXiv   pre-print
Most publicly available brain MRI datasets are very homogeneous in terms of scanner and protocols, and it is difficult for models that learn from such data to generalize to multi-center and multi-scanner data. We propose a novel data augmentation approach with the aim of approximating the variability in terms of intensities and contrasts present in real world clinical data. We use a Gaussian Mixture Model based approach to change tissue intensities individually, producing new contrasts while
more » ... serving anatomical information. We train a deep learning model on a single scanner dataset and evaluate it on a multi-center and multi-scanner dataset. The proposed approach improves the generalization capability of the model to other scanners not present in the training data.
arXiv:2103.12595v1 fatcat:t54qcdzoonhsbgqatk4xpgbbdy

The Influence of Fructification Managing and Fertilization

Danut MANIUTIU, Rodica Maria SIMA, Diana FICIOR
2008 Notulae Botanicae Horti Agrobotanici Cluj-Napoca  
In the experiments effectuated in 2007, at the University of Agricultural Sciences and Veterinary Medicine Cluj-Napoca, Romania, in a "Venlo" type greenhouse, the fructification managing and fertilization of medium-long fruit cucumbers was studied. During the experiment, observations regarding early and total yield as well as fruit commercial quality have been done. For the variant for which fruits were placed both on the main stem and on the shoots early yield efficiencies of 30.7% and total
more » ... eld efficiencies of 22.9% were recorded in comparison with the variants for which the fruits were placed just on the main stem. The fertilization way influenced both early and total yield difference significant positive being recorded in case of root + foliar fertilization given the root fertilization. Under combined influence of experimental factors, both for early and for total yield, the best results were obtained by variant with managing of fructification both on main stem and shoots, root + foliar fertilized.
doi:10.15835/nbha36273 doaj:71c858485a724bb7b7001566cea25ab2 fatcat:crtpmi5hvrdxfp6fkrgiexyor4

Improved inter-scanner MS lesion segmentation by adversarial training on longitudinal data [article]

Mattias Billast, Maria Ines Meyer, Diana M. Sima, David Robben
2020 arXiv   pre-print
Sima, D. Robben  ... 
arXiv:2002.00952v1 fatcat:cpgavach55funmeew5yejj2jam

A Contrast Augmentation Approach to Improve Multi-Scanner Generalization in MRI

Maria Ines Meyer, Ezequiel de la Rosa, Nuno Pedrosa de Barros, Roberto Paolella, Koen Van Leemput, Diana M. Sima
2021 Frontiers in Neuroscience  
Most data-driven methods are very susceptible to data variability. This problem is particularly apparent when applying Deep Learning (DL) to brain Magnetic Resonance Imaging (MRI), where intensities and contrasts vary due to acquisition protocol, scanner- and center-specific factors. Most publicly available brain MRI datasets originate from the same center and are homogeneous in terms of scanner and used protocol. As such, devising robust methods that generalize to multi-scanner and
more » ... data is crucial for transferring these techniques into clinical practice. We propose a novel data augmentation approach based on Gaussian Mixture Models (GMM-DA) with the goal of increasing the variability of a given dataset in terms of intensities and contrasts. The approach allows to augment the training dataset such that the variability in the training set compares to what is seen in real world clinical data, while preserving anatomical information. We compare the performance of a state-of-the-art U-Net model trained for segmenting brain structures with and without the addition of GMM-DA. The models are trained and evaluated on single- and multi-scanner datasets. Additionally, we verify the consistency of test-retest results on same-patient images (same and different scanners). Finally, we investigate how the presence of bias field influences the performance of a model trained with GMM-DA. We found that the addition of the GMM-DA improves the generalization capability of the DL model to other scanners not present in the training data, even when the train set is already multi-scanner. Besides, the consistency between same-patient segmentation predictions is improved, both for same-scanner and different-scanner repetitions. We conclude that GMM-DA could increase the transferability of DL models into clinical scenarios.
doi:10.3389/fnins.2021.708196 pmid:34531715 pmcid:PMC8439197 fatcat:phwzu5vsercwhaalankai7veuq

Mature miR-99a Upregulation in the Amniotic Fluid Samples from Female Fetus Down Syndrome Pregnancies: A Pilot Study

Anda-Cornelia Vizitiu, Danae Stambouli, Anca-Gabriela Pavel, Maria-Cezara Muresan, Diana Maria Anastasiu, Cristina Bejinar, Anda Alexa, Catalin Marian, Ioan Ovidiu Sirbu, Laurentiu Sima
2019 Medicina  
and Objective: Although Down syndrome is the most frequent aneuploidy, its pathogenic molecular mechanisms are not yet fully understood. The aim of our study is to quantify—by qRT-PCR—the expression levels of both the mature forms and the pri-miRNAs of the microRNAs resident on chromosome 21 (miR(21)) in the amniotic fluid samples from Down syndrome singleton pregnancies and to estimate the impact of the differentially expressed microRNAs on Down syndrome fetal heart and amniocytes
more » ... s. Materials and methods: We collected amniotic fluid samples harvested by trained obstetricians as part of the second trimester screening/diagnostic procedure for aneuploidies to assess the trisomy 21 status by QF-PCR and karyotyping. Next, we evaluated—by Taqman qRT-PCR—the expression levels of both the mature forms and the pri-miRNA precursors of the microRNAs resident on chromosome 21 in amniotic fluid samples from singleton Down syndrome and euploid pregnancies. Further, we combined miRWalk 3.0 microRNA target prediction with GEO DataSets analysis to estimate the impact of hsa-miR-99a abnormal expression on Down syndrome heart and amniocytes transcriptome. Results: We found a statistically significant up-regulation of the mature form of miR-99a, but not pri-miR-99a, in the amniotic fluid samples from Down syndrome pregnancies with female fetuses. GATHER functional enrichment analysis of miRWalk3.0-predicted targets from Down syndrome amniocytes and fetal hearts transcriptome GEODataSets outlined both focal adhesion and cytokine–cytokine receptor interaction signaling as novel signaling pathways impacted by miR-99a and associated with cardiac defects in female Down syndrome patients. Conclusions: The significant overexpression of miR-99a, but not pri-miR-99a, points towards an alteration of the post-transcriptional mechanisms of hsa-miR-99a maturation and/or stability in the female trisomic milieu, with a potential impact on signaling pathways important for proper development of the heart.
doi:10.3390/medicina55110728 pmid:31703316 pmcid:PMC6915350 fatcat:b6byqz4g5bfqjf3r5pfeblolay

Towards Multimodal Machine Learning Prediction of Individual Cognitive Evolution in Multiple Sclerosis

Stijn Denissen, Oliver Y. Chén, Johan De Mey, Maarten De Vos, Jeroen Van Schependom, Diana Maria Sima, Guy Nagels
2021 Journal of Personalized Medicine  
Multiple sclerosis (MS) manifests heterogeneously among persons suffering from it, making its disease course highly challenging to predict. At present, prognosis mostly relies on biomarkers that are unable to predict disease course on an individual level. Machine learning is a promising technique, both in terms of its ability to combine multimodal data and through the capability of making personalized predictions. However, most investigations on machine learning for prognosis in MS were geared
more » ... owards predicting physical deterioration, while cognitive deterioration, although prevalent and burdensome, remained largely overlooked. This review aims to boost the field of machine learning for cognitive prognosis in MS by means of an introduction to machine learning and its pitfalls, an overview of important elements for study design, and an overview of the current literature on cognitive prognosis in MS using machine learning. Furthermore, the review discusses new trends in the field of machine learning that might be adopted for future studies in the field.
doi:10.3390/jpm11121349 pmid:34945821 pmcid:PMC8707909 fatcat:75ssrzwlybacljuz3mrtnhs3p4

Interaction of silicon-based quantum dots with gibel carp liver: oxidative and structural modifications

Loredana Stanca, Sorina Petrache, Andreea Serban, Andrea Staicu, Cornelia Sima, Maria Munteanu, Otilia Zărnescu, Diana Dinu, Anca Dinischiotu
2013 Nanoscale Research Letters  
Quantum dots (QDs) interaction with living organisms is of central interest due to their various biological and medical applications. One of the most important mechanisms proposed for various silicon nanoparticle-mediated toxicity is oxidative stress. We investigated the basic processes of cellular damage by oxidative stress and tissue injury following QD accumulation in the gibel carp liver after intraperitoneal injection of a single dose of 2 mg/kg body weight Si/SiO 2 QDs after 1, 3, and 7
more » ... ys from their administration. QDs gradual accumulation was highlighted by fluorescence microscopy, and subsequent histological changes in the hepatic tissue were noted. After 1 and 3 days, QD-treated fish showed an increased number of macrophage clusters and fibrosis, while hepatocyte basophilia and isolated hepatolytic microlesions were observed only after substantial QDs accumulation in the liver parenchyma, at 7 days after IP injection. Induction of oxidative stress in fish liver was revealed by the formation of malondialdehyde and advanced oxidation protein products, as well as a decrease in protein thiol groups and reduced glutathione levels. The liver enzymatic antioxidant defense was modulated to maintain the redox status in response to the changes initiated by Si/SiO 2 QDs. So, catalase and glutathione peroxidase activities were upregulated starting from the first day after injection, while the activity of superoxide dismutase increased only after 7 days. The oxidative damage that still occurred may impair the activity of more sensitive enzymes. A significant inhibition in glucose-6-phosphate dehydrogenase and glutathione-S -transferase activity was noted, while glutathione reductase remained unaltered. Taking into account that the reduced glutathione level had a deep decline and the level of lipid peroxidation products remained highly increased in the time interval we studied, it appears that the liver antioxidant defense of Carassius gibelio does not counteract the oxidative stress induced 7 days after silicon-based QDs exposure in an efficient manner.
doi:10.1186/1556-276x-8-254 pmid:23718202 pmcid:PMC3680243 fatcat:ybbjzz7uofagrhw67klkeys4qq

Relevance Vector Machines for Harmonization of MRI Brain Volumes Using Image Descriptors [chapter]

Maria Ines Meyer, Ezequiel de la Rosa, Koen Van Leemput, Diana M. Sima
2019 Lecture Notes in Computer Science  
With the increased need for multi-center magnetic resonance imaging studies, problems arise related to differences in hardware and software between centers. Namely, current algorithms for brain volume quantification are unreliable for the longitudinal assessment of volume changes in this type of setting. Currently most methods attempt to decrease this issue by regressing the scanner- and/or center-effects from the original data. In this work, we explore a novel approach to harmonize brain
more » ... measurements by using only image descriptors. First, we explore the relationships between volumes and image descriptors. Then, we train a Relevance Vector Machine (RVM) model over a large multi-site dataset of healthy subjects to perform volume harmonization. Finally, we validate the method over two different datasets: i) a subset of unseen healthy controls; and ii) a test-retest dataset of multiple sclerosis (MS) patients. The method decreases scanner and center variability while preserving measurements that did not require correction in MS patient data. We show that image descriptors can be used as input to a machine learning algorithm to improve the reliability of longitudinal volumetric studies.
doi:10.1007/978-3-030-32695-1_9 fatcat:rfcdxbcwobggfdaoww432smqqy

Automated MRI volumetry as a diagnostic tool for Alzheimer's disease: validation of icobrain dm

Hanne Struyfs, Diana M. Sima, Melissa Wittens, Annemie Ribbens, Nuno Pedrosa de Barros, Thanh Vân Phan, Maria Ines Ferraz Meyer, Lene Claes, Ellis Niemantsverdriet, Sebastiaan Engelborghs, Wim Van Hecke, Dirk Smeets
2020 NeuroImage: Clinical  
Sima, Annemie Ribbens, Nuno Pedrosa de Barros, Thanh Vân Phan, Lene Claes, Maria Ines Ferraz Meyer, Wim Van Hecke, Dirk Smeets. Melissa Wittens and Ellis Niemantsverdriet have no competing interests.  ...  Declaration of Competing Interest The following authors are employed (or have been employed at the time of performing the work relevant for this paper) by icometrix: Hanne Struyfs, Diana M.  ... 
doi:10.1016/j.nicl.2020.102243 pmid:32193172 pmcid:PMC7082216 fatcat:qtxh2gndz5dufca2ylco4mesou

Quantifying brain tumor tissue abundance in HR-MAS spectra using non-negative blind source separation techniques

Anca Ramona Croitor Sava, M. Carmen Martinez-Bisbal, Diana Maria Sima, Jorge Calvar, Vicente Esteve, Bernardo Celda, Uwe Himmelreich, Sabine Van Huffel
2012 Journal of Chemometrics  
doi:10.1002/cem.2456 fatcat:wz7b2jsd2vcz5h6nveug7u53ou

Initializing nonnegative matrix factorization using the successive projection algorithm for multi-parametric medical image segmentation

Nicolas Sauwen, Marjan Acou, Halandur Nagaraja Bharath, Diana Maria Sima, Jelle Veraart, Frederik Maes, Uwe Himmelreich, Eric Achten, Sabine Van Huffel
2016 The European Symposium on Artificial Neural Networks  
As nonnegative matrix factorization (NMF) represents a nonconvex problem, the quality of its solution will depend on the initialization of the factor matrices. This study proposes the Successive Projection Algorithm (SPA) as a feasible NMF initialization method. SPA is applied to a multi-parametric MRI dataset for automated NMF brain tumor segmentation. SPA provides fast and reproducible estimates of the tissue sources, and segmentation quality is found to be similar compared to repetitive random initialization.
dblp:conf/esann/SauwenABSVMHAH16 fatcat:z5fs67txxfe2lmg6yrlgyja6nu

Deciphering the Magainin Resistance Process of Escherichia coli Strains in Light of the Cytosolic Proteome

Simone Maria-Neto, Elizabete de Souza Cândido, Diana Ribas Rodrigues, Daniel Amaro de Sousa, Ezequiel Marcelino da Silva, Lidia Maria Pepe de Moraes, Anselmo de Jesus Otero-Gonzalez, Beatriz Simas Magalhães, Simoni Campos Dias, Octávio Luiz Franco
2012 Antimicrobial Agents and Chemotherapy  
ABSTRACTAntimicrobial peptides (AMPs) are effective antibiotic agents commonly found in plants, animals, and microorganisms, and they have been suggested as the future of antimicrobial chemotherapies. It is vital to understand the molecular details that define the mechanism of action of resistance to AMPs for a rational planning of the next antibiotic generation and also to shed some light on the complex AMP mechanism of action. Here, the antibiotic resistance ofEscherichia coliATCC 8739 to
more » ... inin I was evaluated in the cytosolic subproteome. Magainin-resistant strains were selected after 10 subsequent spreads at subinhibitory concentrations of magainin I (37.5 mg · liter−1), and their cytosolic proteomes were further compared to those of magainin-susceptible strains through two-dimensional electrophoresis analysis. As a result, 41 differentially expressed proteins were detected byin silicoanalysis and further identified by tandem mass spectrometryde novosequencing. Functional categorization indicated an intense metabolic response mainly in energy and nitrogen uptake, stress response, amino acid conversion, and cell wall thickness. Indeed, data reported here show that resistance to cationic antimicrobial peptides possesses a greater molecular complexity than previously supposed, resulting in cell commitment to several metabolic pathways.
doi:10.1128/aac.05558-11 pmid:22290970 pmcid:PMC3318327 fatcat:xl5kbhbo6zgtpng7hsxd6wxvau

The Total Least Squares Problem in AX≈B: A New Classification with the Relationship to the Classical Works

Iveta Hnětynková, Martin Plešinger, Diana Maria Sima, Zdeněk Strakoš, Sabine Van Huffel
2011 SIAM Journal on Matrix Analysis and Applications  
This paper revisits the analysis of the total least squares (TLS) problem AX ≈ B with multiple right-hand sides given by Van Huffel and Vandewalle in the monograph, The Total Least Squares Problem: Computational Aspects and Analysis, SIAM, Philadelphia, 1991. The newly proposed classification is based on properties of the singular value decomposition of the extended matrix ½BjA. It aims at identifying the cases when a TLS solution does or does not exist and when the output computed by the
more » ... cal TLS algorithm, given by Van Huffel and Vandewalle, is actually a TLS solution. The presented results on existence and uniqueness of the TLS solution reveal subtleties that were not captured in the known literature. Introduction. This paper focuses on the total least squares (TLS) formulation of the linear approximation problem with multiple right-hand sides We concentrate on the incompatible problem (1.1), i.e., RðBÞ ⊄ RðAÞ. The compatible case reduces to finding a solution of a system of linear algebraic equations. In TLS, contrary to the ordinary least squares, the correction is allowed to compensate for errors in the system (data) matrix A as well as in the right-hand side (observation) matrix B, and the matrices E and G are sought to minimize the Frobenius norm in
doi:10.1137/100813348 fatcat:a53cymh3lfegvekfz5kojkqlqu

Comparison of manual and semi-manual delineations for classifying glioblastoma multiforme patients based on histogram and texture MRI features

Adrian Ion-Margineanu, Sofie Van Cauter, Diana Maria Sima, Frederik Maes, Stefan Sunaert, Uwe Himmelreich, Sabine Van Huffel
2017 The European Symposium on Artificial Neural Networks  
In this paper we study the task of classifying the follow-up course of brain tumour patients that had surgery. Multiple magnetic resonance imaging brain scans were taken for each patient. We propose a simple method of delineating the contrast enhancing tumour lesion based on the total tumour region. We compare balanced accuracy values after tuning SVM-lin and SVM-rbf on histogram and 3-D texture features extracted from semi-manual and manual delineations. Results show that our proposed delineating method outperforms the classical method.
dblp:conf/esann/Ion-MargineanuC17 fatcat:4ywixlphsrdl7otrhqgjfxghga
« Previous Showing results 1 — 15 out of 490 results