Search PubMed⌕ Search

Biomedical subjects

L Vanhamme

Publications and source records attributed to L Vanhamme.

At least 19 recordsLinked to original sources
PubMed2006

Java-based framework for processing and displaying short-echo-time magnetic resonance spectroscopy signals.

Magnetic resonance spectroscopy (MRS) can be used to determine in a non-invasive way the concentrations of certain chemical substances, also called metabolites. The spectra of MRS signals contain peaks that correspond to the metabolites of interest. Short-echo-time signals are characterized by heavily overlapping metabolite peaks and require sophisticated processing methods. To be useful in a clinical environment tools are needed that can process those signals in an accurate and fast way. Therefore, we developed novel processing methods and we designed a freely available and open-source framework (http://www.esat.kuleuven.ac.be/sista/members/biomed) in which the processing methods can be integrated. The framework has a set of abstract classes, called hot spots, and its goal is to provide a general structure and determine the control flow of the program. It provides building blocks or components in order to help developers with integrating their methods in the framework via a plug-in system. The framework is designed with the unified modeling language (UML) and implemented in Java. When a developer implements the framework he gets an application that acts like a simple and user-friendly graphical user interface (GUI) for processing MRS data. This article describes in detail the structure and implementation of the framework and the integration of our processing methods in it.

Belgium↗
PubMed2005

The use of multivariate MR imaging intensities versus metabolic data from MR spectroscopic imaging for brain tumour classification.

This study investigated the value of information from both magnetic resonance imaging and magnetic resonance spectroscopic imaging (MRSI) to automated discrimination of brain tumours. The influence of imaging intensities and metabolic data was tested by comparing the use of MR spectra from MRSI, MR imaging intensities, peak integration values obtained from the MR spectra and a combination of the latter two. Three classification techniques were objectively compared: linear discriminant analysis, least squares support vector machines (LS-SVM) with a linear kernel as linear techniques and LS-SVM with radial basis function kernel as a nonlinear technique. Classifiers were evaluated over 100 stratified random splittings of the dataset into training and test sets. The area under the receiver operating characteristic (ROC) curve (AUC) was used as a global performance measure on test data. In general, all techniques obtained a high performance when using peak integration values with or without MR imaging intensities. For example for low- versus high-grade tumours, low- versus high-grade gliomas and gliomas versus meningiomas, the mean test AUC was higher than 0.91, 0.94, and 0.99, respectively, when both MR imaging intensities and peak integration values were used. The use of metabolic data from MRSI significantly improved automated classification of brain tumour types compared to the use of MR imaging intensities solely.

Brain Chemistry↗
PubMed2004

Brain tumor classification based on long echo proton MRS signals.

There has been a growing research interest in brain tumor classification based on proton magnetic resonance spectroscopy (1H MRS) signals. Four research centers within the EU funded INTERPRET project have acquired a significant number of long echo 1H MRS signals for brain tumor classification. In this paper, we present an objective comparison of several classification techniques applied to the discrimination of four types of brain tumors: meningiomas, glioblastomas, astrocytomas grade II and metastases. Linear and non-linear classifiers are compared: linear discriminant analysis (LDA), support vector machines (SVM) and least squares SVM (LS-SVM) with a linear kernel as linear techniques and LS-SVM with a radial basis function (RBF) kernel as a non-linear technique. Kernel-based methods can perform well in processing high dimensional data. This motivates the inclusion of SVM and LS-SVM in this study. The analysis includes optimal input variable selection, (hyper-) parameter estimation, followed by performance evaluation. The classification performance is evaluated over 200 stratified random samplings of the dataset into training and test sets. Receiver operating characteristic (ROC) curve analysis measures the performance of binary classification, while for multiclass classification, we consider the accuracy as performance measure. Based on the complete magnitude spectra, automated binary classifiers are able to reach an area under the ROC curve (AUC) of more than 0.9 except for the hard case glioblastomas versus metastases. Although, based on the available long echo 1H MRS data, we did not find any statistically significant difference between the performances of LDA and the kernel-based methods, the latter have the strength that no dimensionality reduction is required to obtain such a high performance.

Artificial Intelligence↗
PubMed2004

Subspace-based MRS data quantitation of multiplets using prior knowledge.

Accurate quantitation of Magnetic Resonance Spectroscopy (MRS) signals is an essential step before converting the estimated signal parameters, such as frequencies, damping factors, and amplitudes, into biochemical quantities (concentration, pH). Several subspace-based parameter estimators have been developed for this task, which are efficient and accurate time-domain algorithms. However, they suffer from a serious drawback: they allow only a limited inclusion of prior knowledge which is important for accuracy and resolution. In this paper, a new method is presented: KNOB-SVD and its improved variant KNOB-TLS. KNOB-SVD is a recently proposed method, based on the Singular Value Decomposition (SVD), which allows the use of more prior knowledge about the signal parameters than previously published subspace-based methods. We compare its performance in terms of robustness and accuracy with the performance of three commonly used methods for signal parameter estimation: HTLS, a subspace-based method which does not allow any inclusion of prior knowledge, except for the model order; HTLSPK(Delta fd(eq)), a subspace-based method obtained by incorporating in HTLS the prior information that the frequency differences between doublet components are known and the damping factors are equal; and AMARES, an interactive maximum likelihood method that allows the inclusion of a variety of prior knowledge. Extensive simulation and in vivo studies, using (31)P as well as proton MRS signals, show that the new method outperforms HTLS and HTLSPK(Delta fd(eq)) in robustness, accuracy, and resolution, and that it provides parameter estimates comparable to the AMARES ones.

Journal Article↗
PubMed2004

Classification of brain tumours using short echo time 1H MR spectra.

The purpose was to objectively compare the application of several techniques and the use of several input features for brain tumour classification using Magnetic Resonance Spectroscopy (MRS). Short echo time 1H MRS signals from patients with glioblastomas (n = 87), meningiomas (n = 57), metastases (n = 39), and astrocytomas grade II (n = 22) were provided by six centres in the European Union funded INTERPRET project. Linear discriminant analysis, least squares support vector machines (LS-SVM) with a linear kernel and LS-SVM with radial basis function kernel were applied and evaluated over 100 stratified random splittings of the dataset into training and test sets. The area under the receiver operating characteristic curve (AUC) was used to measure the performance of binary classifiers, while the percentage of correct classifications was used to evaluate the multiclass classifiers. The influence of several factors on the classification performance has been tested: L2- vs. water normalization, magnitude vs. real spectra and baseline correction. The effect of input feature reduction was also investigated by using only the selected frequency regions containing the most discriminatory information, and peak integrated values. Using L2-normalized complete spectra the automated binary classifiers reached a mean test AUC of more than 0.95, except for glioblastomas vs. metastases. Similar results were obtained for all classification techniques and input features except for water normalized spectra, where classification performance was lower. This indicates that data acquisition and processing can be simplified for classification purposes, excluding the need for separate water signal acquisition, baseline correction or phasing.

Brain Chemistry↗
PubMed2003

Quantification of the glycogen 13C-1 NMR signal during glycogen synthesis in perfused rat liver.

We studied glycogen synthesis from glucose in perfused livers of fed (n = 4) and 24 h starved (n = 7) rats. Glycogenolysis was inhibited by BAY R3401 (150 microM) and proglycosyn (100 microM). After 60 min, we replaced 99% (13)C-1 glucose by natural abundance glucose. This pulse-chase design allowed us to recognize residual ongoing futile glycogen turnover from the release of initially deposited (13)C-label, into the (13)C-free chase medium. Net residual turnover was less than 2 +/- 0.7% and 0.6 +/- 0.2% of 1-(13)C glycogen deposition rates of 0.31 +/- 0.04 and 0.99 +/- 0.04 micromol glucose g(-1) min(-1), in starved and fed livers, respectively. The 1-(13)C glycogen signal was monitored throughout the experiment with proton-decoupled (13)C NMR spectroscopy and analyzed in the time domain using AMARES. We noticed progressive line-broadening in any single experiment in the chase phase. One or a sum of two to three overlapping Lorentzians, with different exponential damping factors, were fitted to the signal. When the S/N was better than 40, the fit always delivered a small and a broad component. In the chase phase, the fit with a single Lorentzian resulted in a decline of glycogen signal by about 15 +/- 4 and 12 +/- 2% in starved and fed rats, respectively. This apparent decline in 1-(13)C glycogen signal could not be accounted for by the appearance of equivalent amounts of (13)C-labeled metabolites in the perfusate. The fit with a sum of two Lorentzians resulted in a decline of glycogen signal intensity of 7 +/- 5 and 5 +/- 3% in starved and fed rats, respectively, which reduced the apparent turnover to 8 +/- 9% and 6 +/- 4%, respectively. Quantification of the growing (13)C-1 glycogen signal requires a model function that accommodates changes in line shape throughout the period under study.

Algorithms↗
PubMed2002

Improved Lanczos algorithms for blackbox MRS data quantitation.

Magnetic resonance spectroscopy (MRS) has been shown to be a potentially important medical diagnostic tool. The success of MRS depends on the quantitative data analysis, i.e., the interpretation of the signal in terms of relevant physical parameters, such as frequencies, decay constants, and amplitudes. A variety of time-domain algorithms to extract parameters have been developed. On the one hand, there are so-called blackbox methods. Minimal user interaction and limited incorporation of prior knowledge are inherent to this type of method. On the other hand, interactive methods exist that are iterative, require user involvement, and allow inclusion of prior knowledge. We focus on blackbox methods. The computationally most intensive part of these blackbox methods is the computation of the singular value decomposition (SVD) of a Hankel matrix. Our goal is to reduce the needed computational time without affecting the accuracy of the parameters of interest. To this end, algorithms based on the Lanczos method are suitable because the main computation at each step, a matrix-vector product, can be efficiently performed by means of the fast Fourier transform exploiting the structure of the involved matrix. We compare the performance in terms of accuracy and efficiency of four algorithms: the classical SVD algorithm based on the QR decomposition, the Lanczos algorithm, the Lanczos algorithm with partial reorthogonalization, and the implicitly restarted Lanczos algorithm. Extensive simulation studies show that the latter two algorithms perform best.

Algorithms↗
PubMed2001

Control and function of the bloodstream variant surface glycoprotein expression sites in Trypanosoma brucei.

African trypanosomes escape the host immune response through a periodical change of their surface coat made of one major type of protein, the variant surface glycoprotein. From a repertoire of a thousand variant surface glycoprotein genes available, only one is expressed at a time, and this takes place in a specialised expression site itself selected from a collection of an estimated 20-30 sites. As the specialised expression sites are long polycistronic transcription units, the variant surface glycoprotein is co-transcribed with several other genes termed expression site-associated genes. How do the trypanosomes only use a single specialised expression site at a time? Why are there two dozen specialised expression sites? What are the functions of the other genes of these transcription units? We review the currently available answers to these questions.

Animals↗
PubMed2001

The VSG expression sites of Trypanosoma brucei: multipurpose tools for the adaptation of the parasite to mammalian hosts.

The variant surface glycoprotein (VSG) genes of Trypanosoma brucei are transcribed in telomeric loci termed VSG expression sites (ESs). Despite permanent initiation of transcription in most if not all of these multiple loci, RNA elongation is abortive except in bloodstream forms where full transcription up to the VSG occurs only in a single ES at a time. The ESs active in bloodstream forms are polycistronic and contain several genes in addition to the VSG, named ES-associated genes (ESAGs). So far 12 ESAGs have been identified, some of which are present only in some ESs. Most of these genes encode surface proteins and this list includes different glycosyl phosphatidyl inositol (GPI)-anchored proteins such as the heterodimeric receptor for the host transferrin (ESAG7/6), integral membrane proteins such as the receptor-like transmembrane adenylyl cyclase (ESAG4) and a surface transporter (ESAG10). An interesting exception is ESAG8, which may encode a cell cycle regulator involved in the differentiation of long slender into short stumpy bloodstream forms. Several ESAGs belong to multigene families including pseudogenes and members transcribed out of the ESs, named genes related to ESAGs (GRESAGs). However, some ESAGs (7, 6 and 8) appear to be restricted to the ESs. Most of these genes can be deleted from the active ES without apparently affecting the phenotype of bloodstream form trypanosomes, probably either due to the expression of ESAGs from 'inactive' ESs (ESAG7/6) or due to the expression of GRESAGs (in particular, GRESAGs4 and GRESAGs1). At least three ESAGs (ESAG7, ESAG6 and SRA) share the evolutionary origin of VSGs. The presence of these latter genes in ESs may confer an increased capacity of the parasite for adaptation to various mammalian hosts, as suggested in the case of ESAG7/6 and proven for SRA, which allows T. brucei to infect humans. Similarly, the existence of a collection of slightly different ESAG4s in the multiple ESs might provide the parasite with adenylyl cyclase isoforms that may regulate growth in response to different environmental conditions. The high transcription rate and high recombination level that prevail in VSG ESs may have favored the generation and/or recruitment in these sites of genes whose hyper-evolution allows adaptation to a larger variety of hosts.

Animals↗
PubMed2001

MR spectroscopy quantitation: a review of time-domain methods.

In this article an overview of time-domain quantitation methods is given. Advantages of processing the data in the measurement domain are discussed. The basic underlying principles of the methods are outlined and from them the situations under which these algorithms perform well are derived. Also an overview of methods to preprocess the data is given. In that respect, signal-to-noise and/or resolution enhancement, the removal of unwanted components and corrections for model imperfections are discussed.

Algorithms↗
PubMed2001

The filtering approach to solvent peak suppression in MRS: a critical review.

Suppressing the solvent peak is important in many applications of biomedical NMR spectroscopy in order to quantify the metabolites with a great accuracy. Among the postprocessing methods proposed in the literature, many deal with the concept of filtering. However, several proposals lack a theoretical perspective and some have not been explicitly applied to quantification problems. The present article is intended to bridge this gap: five methods are analyzed from a theoretical perspective. Subsequently the different methods are applied to the same set of data, and then the latter are quantified using the model fitting method AMARES. With our set, the scheme proposed by T. Sundin et al. (J. Magn. Reson. 139(2), 189-204 (1999)) proved to be the most reliable method.

Deuterium↗
PubMed2001

An update on antigenic variation in African trypanosomes.

African trypanosomes can spend a long time in the blood of their mammalian host, where they are exposed to the immune system and are thought to take advantage of it to modulate their own numbers. Their major immunogenic protein is the variant surface glycoprotein (VSG), the gene for which must be in one of the 20--40 specialized telomeric expression sites in order to be transcribed. Trypanosomes escape antibody-mediated destruction through periodic changes of the expressed VSG gene from a repertoire of approximately 1000. How do trypanosomes exclusively express only one VSG and how do they switch between them?

Animals↗
PubMed2000

A transcript encoding a proteasome beta-subunit and a zinc finger protein in Trypanosoma brucei brucei.

During the screening of a Trypanosoma brucei brucei (T. b. brucei) cDNA library constructed from bloodstream form mRNA, we identified a 2.3kb cDNA encoding a proteasome beta subunit (ORF1) and a putative zinc finger protein (ORF2). Northern blot analysis indicated the presence of a digenic transcript as well as the two individual messengers in both procyclic and bloodstream forms of the parasite. Southern blot analysis showed the relevant locus to be unique. ORF1 encoded a 22.7kDa protein sharing over 50% identity with the eukaryotic PRCE (aka beta5) proteasome beta subunit. This protein contained a beta amino acid signature and residues involved in the catalytic activity. Further phylogenetic analysis indicated that this subunit as well as those from other kinetoplastids could be confidentially assigned to extant eukaryotic subfamilies such as beta1, beta2, and beta5. ORF2 encoded a 14.6kDa putative zinc finger protein containing five repeats of a CCHC motif commonly present in retroviral nucleocapsid proteins as well as proteins involved in vertebrate embryogenesis.

Amino Acid Sequence↗
PubMed2000

A role for the dynamic acylation of a cluster of cysteine residues in regulating the activity of the glycosylphosphatidylinositol-specific phospholipase C of Trypanosoma brucei.

The glycosylphosphatidylinositol-specific phospholipase C or VSG lipase is the enzyme responsible for the cleavage of the glycosylphosphatidylinositol anchor of the variant surface glycoprotein (VSG) and concomitant release of the surface coat in Trypanosoma brucei during osmotic shock or extracellular acidic stress. In Xenopus laevis oocytes the VSG lipase was expressed as a nonacylated and a thioacylated form. This thioacylation occurred within a cluster of three cysteine residues but was not essential for catalytic activity per se. These two forms were also detected in trypanosomes and appeared to be present at roughly equivalent amounts. A reversible shift to the acylated form occurred when cells were triggered to release the VSG by either nonlytic acid stress or osmotic lysis. A wild type VSG lipase or a gene mutated in the three codons for the acylated cysteines were reinserted in the genome of a trypanosome null mutant for this gene. A comparative analysis of these revertant trypanosomes indicated that thioacylation might be involved in regulating enzyme access to the VSG substrate.

Acylation↗
PubMed2000

A single-stranded DNA-binding protein shared by telomeric repeats, the variant surface glycoprotein transcription promoter and the procyclin transcription terminator of Trypanosoma brucei.

In Trypanosoma brucei the genes are organised into long polycistronic transcription units and only three promoters for protein-encoding genes and a single terminator have been characterised. These promoters recruit a polI-like RNA polymerase for the transcription units encoding the two major stage-specific antigens of the parasite, the variant surface glycoprotein (VSG) of the bloodstream form and procyclin of the insect-specific procyclic form, while the terminator is that of a procyclin transcription unit. By deletional and mutational analysis we defined the two DNA sequences essential for the activity of the VSG promoter from a bloodstream form transcription unit and one of the functional elements of the procyclin terminator. These three short sequences are similar, and their C-rich strand binds the same protein of 40 kDa. In addition, this factor also binds to the C-rich strand of the telomeric repeats, the consensus target sequence being 5'-CCCTNN-3'. The factor-binding sequences are functionally interchangeable in chimeric promoter or terminator constructs, although additional elements are required for full activity.

Animals↗
PubMed2000

Frequency-selective quantification of biomedical magnetic resonance spectroscopy data.

In this paper the possibility of obtaining accurate estimates of parameters of selected peaks in the presence of unknown or uninteresting spectral features in biomedical magnetic resonance spectroscopy (MRS) signals is investigated. This problem is denoted by frequency-selective parameter estimation. A new time-domain technique based on maximum-phase finite impulse response (FIR) filters is presented. The proposed method is compared to a number of existing approaches: the application of a weighting function in the time domain, frequency domain fitting using a polynomial baseline, and the time-domain HSVD filter method. The ease of use and low computational complexity of the FIR filter method make it an attractive approach for frequency-selective parameter estimation. The methods are validated using simulations of relevant (13)C and (31)P MRS examples.

Adenosine Triphosphate↗
PubMed2000

Differential RNA elongation controls the variant surface glycoprotein gene expression sites of Trypanosoma brucei.

The protozoan parasite Trypanosoma brucei develops antigenic variation to escape the immune response of its host. To this end, the trypanosome genome contains multiple telomeric expression sites competent for transcription of variant surface glycoprotein genes, but as a rule only a single antigen is expressed at any time. We used reverse transcription-PCR (RT-PCR) to analyse transcription of different segments of the expression sites in different variant clones of two independent strains of T. brucei. The results indicated that RNA polymerase is installed and active at the beginning of many, if not all, expression sites simultaneously, but that a progressive arrest of RNA elongation occurs in all but one site. This defect is linked to inefficient RNA processing and RNA release from the nucleus. Therefore, functional transcription in the active site appears to depend on the selective recruitment of a RNA elongation/processing machinery.

Animals↗