Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Dataset”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 829 records · Page 46Linked to original sources
PubMed2006

Comparative visualization for comprehensive two-dimensional gas chromatography.

This paper investigates methods for comparing datasets produced by comprehensive two-dimensional gas chromatography (GC x GC). Chemical comparisons are useful for process monitoring, sample classification or identification, correlative determinations, and other important tasks. GC x GC is a powerful new technology for chemical analysis, but methods for comparative visualization must address challenges posed by GC x GC data: inconsistency and complexity. The approach extends conventional techniques for image comparison by utilizing specific characteristics of GC x GC data and developing new methods for comparative visualization and analysis. The paper describes techniques that register (or align) GC x GC datasets to remove retention-time variations; normalize intensities to remove sample amount variations; compute differences in local regions to remove slight misregistrations and differences in peak shapes; employ color (hue), intensity, and saturation to simultaneously visualize differences and values; and use tools for masking, three-dimensional visualization, and tabular presentation with controls for graphical highlights to significantly improve comparative analysis of GC x GC datasets. Experimental results indicate that the comparative methods preserve chemical information and support qualitative and quantitative analyses.

Benzene Derivatives↗
PubMed2006

Morphological muscle and joint parameters for musculoskeletal modelling of the lower extremity.

BACKGROUND: To assist in the treatment of gait disorders, an inverse and forward 3D musculoskeletal model of the lower extremity will be useful that allows to evaluate if-then scenarios. Currently available anatomical datasets do not comprise sufficiently accurate and complete information to construct such a model. The aim of this paper is to present a complete and consistent anatomical dataset, containing the orientations of joints (hip, knee, ankle and subtalar joints), muscle parameters (optimum length, physiological cross sectional area), and geometrical parameters (attachment sites, 'via' points). METHODS: One lower extremity, taken from a male embalmed specimen, was studied. Position and geometry were measured with a 3D-digitizer. Optotrak was used for measurement of rotation axes of joints. Sarcomere length was measured by laser diffraction. FINDINGS: A total of 38 muscles were measured. Each muscle was divided in different muscle lines of action based on muscle morphology. 14 Ligaments of the hip, knee and ankle were included. INTERPRETATION: The presented anatomical dataset embraces all necessary data for state of the art musculoskeletal modelling of the lower extremity. Implementation of these data into an (existing) model is likely to significantly improve the estimation of muscle forces and will thus make the use of the model as a clinical tool more feasible.

Aged↗
PubMed2005

Optimizing principal components analysis of event-related potentials: matrix type, factor loading weighting, extraction, and rotations.

OBJECTIVE: Given conflicting recommendations in the literature, this report seeks to present a standard protocol for applying principal components analysis (PCA) to event-related potential (ERP) datasets. METHODS: The effects of a covariance versus a correlation matrix, Kaiser normalization vs. covariance loadings, truncated versus unrestricted solutions, and Varimax versus Promax rotations were tested on 100 simulation datasets. Also, whether the effects of these parameters are mediated by component size was examined. RESULTS: Parameters were evaluated according to time course reconstruction, source localization results, and misallocation of ANOVA effects. Correlation matrices resulted in dramatic misallocation of variance. The Promax rotation yielded much more accurate results than Varimax rotation. Covariance loadings were inferior to Kaiser Normalization and unweighted loadings. CONCLUSIONS: Based on the current simulation of two components, the evidence supports the use of a covariance matrix, Kaiser normalization, and Promax rotation. When these parameters are used, unrestricted solutions did not materially improve the results. We argue against their use. Results also suggest that optimized PCA procedures can measurably improve source localization results. SIGNIFICANCE: Continued development of PCA procedures can improve the results when PCA is applied to ERP datasets.

Electroencephalography↗
PubMed2025

Three-dimensional U-Net with transfer learning improves automated whole brain delineation from MRI brain scans of rats, mice, and monkeys.

BACKGROUND: Automated whole-brain delineation (WBD) techniques often struggle to generalize across pre-clinical studies due to variations in animal models, magnetic resonance imaging (MRI) scanners, and tissue contrasts. We developed a 3D U-Net neural network for WBD pre-trained on organophosphate intoxication (OPI) rat brain MRI scans. We used transfer learning (TL) to adapt this OPI-pretrained network to other animal models: rat model of Alzheimer's disease (AD), mouse model of tetramethylenedisulfotetramine (TETS) intoxication, and titi monkey model of social bonding. METHODS: We assessed an OPI-pretrained 3D U-Net across animal models under three conditions: (1) direct application to each dataset; (2) utilizing TL; and (3) training disease-specific U-Net models. For each condition, training dataset size (TDS) was optimized, and output WBDs were compared to manual segmentations for accuracy. RESULTS: The OPI-pretrained 3D U-Net (TDS = 100) achieved the best accuracy [median[min-max]] for the test OPI dataset with a Dice coefficient (DC) = [0.987 [0.977-0.992]] and Hausdorff distance (HD) = [0.86 [0.55-1.27]]mm. TL improved generalization across all models [AD (TDS = 40): DC = 0.987 [0.977-0.992] and HD = 0.72 [0.54-1.00]mm; TETS (TDS = 10): DC = 0.992 [0.984-0.993] and HD = 0.40 [0.31-0.50]mm; Monkey (TDS = 8): DC = 0.977 [0.968-0.979] and HD = 3.03 [2.19-3.91]mm], showing performance comparable to disease-specific networks. CONCLUSIONS: The OPI-pretrained 3D U-Net with TL achieved accuracy comparable to disease-specific networks with reduced training data (TDS ≤ 40 scans) across all models. Future work will focus on developing a multi-region delineation pipeline for pre-clinical MRI brain data, utilizing the proposed WBD as an initial step.

Animals↗
PubMed2005

An anatomic coordinate system of the femoral neck for highly reproducible BMD measurements using 3D QCT.

In this paper, a procedure for the determination of an anatomically oriented coordinate system of the femoral neck (NCS) in 3D spiral CT datasets is described. The origin of the NCS is centered in the minimal cross-sectional area of the neck. Its three axes are defined as follows: the so called neck axis is perpendicular to this area and points towards the femoral head, the second axis is the principal axis of the minimal cross-sectional area and the third axis is perpendicular to the other two. After a semi-automatic 3D segmentation of the proximal femur the NCS is automatically determined in a two-step minimization procedure. Relative to the coordinate system volumes of interest (VOIs) are positioned in which bone mineral density (BMD) and cortical thickness are analyzed. We investigated intra- and inter-operator precision of the position of the NCS, the BMD in cortical and trabecular VOIs, and cortical thickness in nine pelvic CT datasets obtained from clinical routine examinations. We further investigated the effect of increased noise by adding Gaussian distributed noise to measured projections before tomographic reconstruction. The mean precision error (averaged form the results of the nine datasets) of the NCS position was less than 0.5 mm and smaller than 2.25 degrees . There were no significant differences between inter- and intra-operator analyses. Precision errors in trabecular BMD were smaller than 3% in a stack of five 1 mm thin slices cut perpendicularly to the neck axis and smaller than 1% in a spherical VOI encompassing the neck. Relative precision errors for cortical BMD were smaller than 3% for both VOIs. An increase of noise up to a factor of 5 caused a maximal displacement of the NCS origin position by less than 1mm and a rotation by less than 2 degrees .

Bone Density↗
PubMed2006

Adaptive reconstruction of pipe-shaped human organs from 3D ultrasonic volume.

In this paper, we introduce an adaptive scheme for reconstructing pipe-shaped human organs from the volume data acquired by 3D ultrasonic devices. No other methods but the contour-based scheme was used in the process of reconstructing the volume data into a 3D polygonal surface. In the first step, the algorithm extracts contours from the sampled slices of the volume data using the modified radial gradient method, in which the points are sampled on the boundary of the region of interest by radiating rays and connected through making use of the chain code algorithm. The contours are represented as the context-free grammar, and their parsing trees are traversed during the reconstruction. The generated polygonal surface is refined as the contours are being refined at the casting of the new rays between the existing rays to sample new points and to modify the contours according to these newly derived points. An adaptive scheme is achieved in casting the rays adaptively on the slices. The proposed algorithm is to be applied in reconstructing the pipe-shaped human organs, such as arteries or blood vessels, to a polygonal surface. In this paper, we present an innovative tiling algorithm that reconstructs pipe-shaped human organ from 3D ultrasonic datasets. A set of contours on slices through the ultrasonic datasets is extracted using a modified radial gradient method, and our algorithm tiles these to make a polygonal surface. The tiling is performed by traversing a set of parsing trees which represent the contours in a context-free grammar. This makes our algorithm more efficient than previous algorithms that reconstruct surfaces from a set of contours. The first step of the algorithm is to determine a contour on each slice of the 3D ultrasonic dataset. After removing unwanted artifacts from the slice by applying several noise-removing operators, the centroid pixel of region of interest on the slice is designated. A radial gradient method casts a set of rays from the centroid pixel to the boundary of the slice and computes the intersection points between the rays and the boundary cells of the object so as to determine the contours. The second step uses context-free grammar that represents the contours. Each edge of a contour can be classified into six categories according to its relation with the rays cast from the centroid pixel, and the contour can then be represented by a string in a context-free grammar whose terminal symbols are the six types of the edges. A polygonal surface between two contours is constructed by traversing the parsing trees of the contours and determining the corresponding edges. The third step is to refine the smooth surface constructed in the second step by casting more rays. Additional rays refine the contour by decomposing the edges on the contour and convert leaf node of the parsing tree to the root of a new sub-tree whose leaf nodes denote the newly created edges. Our algorithm was tested on a phantom object and an artery from the neck. Results show that the performance of the algorithm and the quality of the resulting surface are better than those of existing algorithms. We have implemented a navigation facility that allows users to investigate the pipe-shaped human organs interactively.

Algorithms↗
PubMed2006

The use of 3D surface fitting for robust polyp detection and classification in CT colonography.

In this paper we describe the development of a computationally efficient computer-aided detection (CAD) algorithm based on the evaluation of the surface morphology that is employed for the detection of colonic polyps in computed tomography (CT) colonography. Initial polyp candidate voxels were detected using the surface normal intersection values. These candidate voxels were clustered using the normal direction, convexity test, region growing and Gaussian distribution. The local colonic surface was classified as polyp or fold using a feature normalized nearest neighborhood classifier. The main merit of this paper is the methodology applied to select the robust features derived from the colon surface that have a high discriminative power for polyp/fold classification. The devised polyp detection scheme entails a low computational overhead (typically takes 2.20min per dataset) and shows 100% sensitivity for phantom polyps greater than 5mm. It also shows 100% sensitivity for real polyps larger than 10mm and 91.67% sensitivity for polyps between 5 to 10mm with an average of 4.5 false positives per dataset. The experimental data indicates that the proposed CAD polyp detection scheme outperforms other techniques that identify the polyps using features that sample the colon surface curvature especially when applied to low-dose datasets.

Algorithms↗
PubMed2006

Quantification of regional left ventricular wall motion from real-time 3-dimensional echocardiography in patients with poor acoustic windows: effects of contrast enhancement tested against cardiac magnetic resonance.

OBJECTIVE: Regional left ventricular function can be assessed by real-time 3-dimensional echocardiography (RT3DE) in patients with good image quality. Our goals were to: (1) test the feasibility of RT3DE quantification of regional wall motion (RWM) in patients with poor acoustic windows who require contrast for endocardial visualization; and (2) validate these measurements against cardiac magnetic resonance (CMR) reference. METHODS: RT3DE datasets and CMR images were obtained in 24 patients. In 16 of 24 patients with suboptimal endocardial definition, RT3DE imaging was repeated with intravenous contrast and triggering at end systole and end diastole. RT3DE datasets were analyzed using custom software designed to semiautomatically detect and segment the endocardial surface and calculate RWM values. CMR images were analyzed using commercial software to obtain reference values for RWM. RESULTS: In 8 of 24 patients with good endocardial definition, RT3DE values of RWM correlated well with CMR (r = 0.73) with a small bias (-1.0 mm). In the remaining 16 patients, analysis of nonenhanced RT3DE datasets yielded lower correlation with CMR (r = 0.61) and a slightly greater bias (-1.5 mm). The agreement with CMR improved significantly (r = 0.76, bias -1.1 mm) with contrast enhancement. CONCLUSIONS: The agreement between RT3DE and CMR values of RWM is directly related to RT3DE image quality. In patients with poor acoustic windows, dual-triggered contrast enhancement improves the accuracy of RWM quantification to a level similar to that noted in patients with good images without contrast.

Computer Systems↗
PubMed2004

OPCAB versus early mortality and morbidity: an issue between clinical relevance and statistical significance.

OBJECTIVE: To evaluate the impact of OPCAB on major postoperative events in a large consecutive cohort of patients, covering the complete spectrum of risk. METHODS: A consecutive series of 3333 CABG patients operated in a single institution (Jan/97-Jan/03) is analyzed after a complete (98%) midterm reengineering towards off-pump surgery (Oct/99). Patients in cardiogenic shock are excluded. The on- (N=1593) or off-pump (N=1740) datasets are comparable for most demographic and non-cardiac variability. The studied events are early mortality, early stroke, early infarct, early dialysis and hospital stay. Three methods adjust for possible patient selection: similar datasets, forced inclusion of a saturated OPCAB propensity score and finally multivariate correction. RESULTS: Non-risk adjusted. The 3-month survival was 96.7+/-0.4% (OPCAB) and 95.9+/-0.5% (ECC) (P=0.2). The 8-day freedom from stroke was 99.4+/-0.2% (OPCAB) and 98.5+/-0.3% (ECC) (P=0.004). The prevalence of dialysis was 1.67% in OPCAB and 2.27% in ECC (P=0.2). The 8-day freedom from infarct was 98.4+/-0.2% (OPCAB) and 98.3+/-0.2% (ECC) (P=0.7). The freedom from hospital discharge day 15 was 17.6+/-0.9% (OPCAB) and 18.4+/-0.8% (ECC) (P=0.001). Propensity score corrected and adjusted for event-related variability. The survival effect remained non-significant (P=0.3), also for patients with a EuroSCORE>8 (P=0.9). The stroke effect became non-significant (P=0.2), but stayed significant for patients with severe internal carotid artery stenosis (P=0.02). The dialysis-effect remained non-significant (P=0.6), also for patients with an elevated creatinine (P=0.7). The early infarct-effect remained non-significant (P=0.8), also for the female patients (P=0.8). The hospital discharge was significantly influenced by the OPCAB approach for the total group (P=0.02) as well as for the patients with EuroSCORE>8 (P=0.01). CONCLUSIONS: The observed 20% reduction of mortality, 60% reduction of stroke and 20% reduction of dialysis were partly neutralized by the adjusting methods and demand, at least, larger datasets to obtain statistical significance. Subdatasets with fewer patients but higher risk identified risk-reducing effects for stroke. Hospital stay was shortened by the OPCAB approach. The interactions between risk, number of patients and the risk-reducing effect are the cornerstones of evidence generation for the OPCAB approach. These results were obtained through a very strict reengineering and cannot be extended to all OPCAB programs.

Aged↗
PubMed2006

A novel method for apoptosis protein subcellular localization prediction combining encoding based on grouped weight and support vector machine.

Apoptosis proteins have a central role in the development and homeostasis of an organism. These proteins are very important for understanding the mechanism of programmed cell death. Based on the idea of coarse-grained description and grouping in physics, a new feature extraction method with grouped weight for protein sequence is presented, and applied to apoptosis protein subcellular localization prediction associated with support vector machine. For the same training dataset and the same predictive algorithm, the overall prediction accuracy of our method in Jackknife test is 13.2% and 15.3% higher than the accuracy based on the amino acid composition and instability index. Especially for the else class apoptosis proteins, the increment of prediction accuracy is 41.7 and 33.3 percentile, respectively. The experiment results show that the new feature extraction method is efficient to extract the structure information implicated in protein sequence and the method has reached a satisfied performance despite its simplicity. The overall prediction accuracy of EBGW_SVM model on dataset ZD98 reach 92.9% in Jackknife test, which is 8.2-20.4 percentile higher than other existing models. For a new dataset ZW225, the overall prediction accuracy of EBGW_SVM achieves 83.1%. Those implied that EBGW_SVM model is a simple but efficient prediction model for apoptosis protein subcellular location prediction.

Algorithms↗
PubMed2005

Iranian STR variation at the fringes of biogeographical demarcation.

The integrative relationship between population genetics and forensic biology allows for a thorough genetic characterization of extant human populations. This study aimed to genetically characterize 150 unrelated healthy donors from a general population in Iran both forensically and phylogenetically. The allelic frequencies of 15 STR loci (D8S1179, D21S11, D7S820, CSF1PO, D3S1358, TH01, D13S317, D16S539, D2S1338, D19S433, vWA, TPOX, D18S51, D5S818 and FGA) were generated. This constitutes the core of polymerase chain reaction (PCR)-based DNA genetic markers in the US Combined DNA Index System (CODIS) plus two additional loci (D2S1338 and D19S433) that together are consistent with several other worldwide database requirements. There were no deviations from Hardy-Weinberg expectations. Based upon the allelic frequencies, several important forensic parameters were calculated including: gene diversity (GD) index, power of discrimination (PD), polymorphic information content (PIC) and power of exclusion (PE). G-tests indicate the allelic frequencies of the Iranians are statistically non-significant compared to two Turkish populations yet, statistically different from the remaining 18 groups obtained from the literature and examined in this study. This suggests that the Iranian dataset may be forensically equivalent to the dataset from the Turkish region of Eastern Anatolia and the general population from Turkey. Phylogenetic analysis of our population with the full set of 15 loci indicate the Iranians occupy an intermediate position relative to the major Caucasian and East Asian clades on a global level. A regional phylogenetic analysis using 13 of the 15 loci indicate the Iranians segregate in a more compact association with groups from southeastern Spain, Arabs from Morocco and Syria, and especially with the general population from Turkey and those from Eastern Anatolia. These groups are flanked by highly differentiated populations from northern India and a Berber group from Tunisia on opposing ends of the regional phylogram. This report also demonstrates the necessity to thoroughly characterize the genetic composition of populations located in geographic intersections in order to choose the appropriate dataset on which to base forensic calculations, not only at an intra-population level, but also at an inter-population level as well.

DNA Fingerprinting↗
PubMed2006

Large-scale in-vivo Caucasian facial soft tissue thickness database for craniofacial reconstruction.

A large-scale study of facial soft tissue depths of Caucasian adults was conducted. Over a 2-years period, 967 Caucasian subjects of both sexes, varying age and varying body mass index (BMI) were studied. A user-friendly and mobile ultrasound-based system was used to measure, in about 20min per subject, the soft tissue thickness at 52 facial landmarks including most of the landmarks used in previous studies. This system was previously validated on repeatability and accuracy [S. De Greef, P. Claes, W. Mollemans, M. Loubele, D. Vandermeulen, P. Suetens, G. Willems, Semi-automated ultrasound facial soft tissue depth registration: method and validation. J. Forensic Sci. 50 (2005)]. The data of 510 women and 457 men were analyzed in order to update facial soft tissue depth charts of the contemporary Caucasian adult. Tables with the average thickness values for each landmark as well as the standard deviation and range, tabulated according to gender, age and BMI are reported. In addition, for each landmark and for both sexes separately, a multiple linear regression of thickness versus age and BMI is calculated. The lateral asymmetry of the face was analysed on an initial subset of 588 subjects showing negligible differences and thus warranting the unilateral measurements of the remaining subjects. The new dataset was statistically compared to three datasets for the Caucasian adults: the traditional datasets of Rhine and Moore [J.S. Rhine, C.E. Moore, Tables of facial tissue thickness of American Caucasoids in forensic anthropology. Maxwell Museum Technical series 1 (1984)] and Helmer [R. Helmer, Schädelidentifizierung durch elektronische bildmischung, Kriminalistik Verlag GmbH, Heidelberg, 1984] together with the most recent in vivo study by Manhein et al. [M.H. Manhein, G.A. Listi, R.E. Barsley, R. Musselman, N.E. Barrow, D.H. Ubelbaker, In vivo facial tissue depth measurements for children and adults. J. Forensic Sci. 45 (2000) 48-60]. The large-scale database presented in this paper offers a denser sampling of the facial soft tissue depths of a more representative subset of the actual Caucasian population over the different age and body posture subcategories. This database can be used as an updated chart for manual and computer-based craniofacial approximation and allows more refined analyses of the possible factors affecting facial soft tissue depth.

Adolescent↗
PubMed2003

Development and implementation of a multi-centre information system for paediatric and infant critical care.

BACKGROUND: With no UK collective information system, a need existed to establish an integrated information system for public and private sector hospitals providing paediatric and infant critical care services. A lack of information in the past made it difficult for those procuring, providing and monitoring services to make informed, evidence-based decisions using reliable integrated data. OBJECTIVES: To develop and implement a collective multi-purpose information system for paediatric and infant critical care that was easily adaptable to any UK infant or paediatric critical care setting. Information outputs had to fulfil policy requirements and meet the needs of stakeholders. METHOD: Two minimum datasets, corresponding data definitions, survey forms and a user database were developed through a process of consultation by utilising an information partnership. Design, content, development and implementation issues were identified, discussed and resolved through a co-ordinated collaborative process. RESULTS: Data collection was implemented in all London and Brighton National Health Service (NHS) general and cardio-thoracic paediatric intensive care (PIC) units, several private PIC units and one NHS tertiary referral neonatal unit (NNU) 24 months from project start. CONCLUSIONS: The development of universal integrated information systems for defined settings of care is achievable within reasonable timeframes; however, successful development and implementation requires working within an information partnership to maximise co-ordination, co-operation and collaboration. Those collecting and using data must be identified and involved in all aspects of development from project start. Financial and manpower resources must be well planned. Datasets should be as small as possible in order to make the collection of complete and valid data realistically achievable. When considering service-based information needs, considerable thought should be given to a multi-purpose; multi-use approach based on the most refined minimum dataset possible.

Child↗
PubMed2007

Impact of hybrid fluorodeoxyglucose positron-emission tomography/computed tomography on radiotherapy planning in esophageal and non-small-cell lung cancer.

PURPOSE: The aim of this study was to investigate the impact of a hybrid fluorodeoxyglucose positron-emission tomography/computed tomography (FDG-PET/CT) scanner in radiotherapy planning for esophageal and non-small-cell lung cancer (NSCLC). METHODS AND MATERIALS: A total of 30 patients (16 with esophageal cancer, 14 with NSCLC) underwent an FDG-PET/CT for radiotherapy planning purposes. Noncontrast total-body spiral CT scans were obtained first, followed immediately by FDG-PET imaging which was automatically co-registered to the CT scan. A physician not involved in the patients' original treatment planning designed a gross tumor volume (GTV) based first on the CT dataset alone, while blinded to the FDG-PET dataset. Afterward, the physician designed a GTV based on the fused PET/CT dataset. To standardize PET GTV margin definition, background liver PET activity was standardized in all images. The CT-based and PET/CT-based GTVs were then quantitatively compared by way of an index of conformality, which is the ratio of the intersection of the two GTVs to their union. RESULTS: The mean index of conformality was 0.44 (range, 0.00-0.70) for patients with NSCLC and 0.46 (range, 0.13-0.80) for patients with esophageal cancer. In 10 of the 16 (62.5%) esophageal cancer patients, and in 12 of the 14 (85.7%) NSCLC patients, the addition of the FDG-PET data led to the definition of a smaller GTV. CONCLUSION: The incorporation of a hybrid FDG-PET/CT scanner had an impact on the radiotherapy planning of esophageal cancer and NSCLC. In future studies, we recommend adoption of a conformality index for a more comprehensive comparison of newer treatment planning imaging modalities to conventional options.

Carcinoma, Non-Small-Cell Lung↗
PubMed2004

New method to assess the registration of CT-MR images of the head.

Due to their complementary information content, both x-ray computed tomography (CT) and magnetic resonance (MR) imaging are employed in certain clinical cases to improve the understanding of pathology involved. o spatially relate the two datasets, image registration and image fusion are employed. However, registration errors, either global or local, are common and are nonuniform within the image volume. In this paper, we propose a new algorithm that assesses the quality of the registration locally within the CT-MR volume and provides visual, color-coded feedback to the user about the location and extent of good and bad correspondence between the two images. The proposed registration assessment algorithm is based on a correspondence analysis of bone structures in the CT and MR images. For that purpose, a custom segmentation algorithm for bone in MR images has been developed that is based on a stochastic threshold computation method. This segmentation method for MR images and the CT-MR registration assessment algorithm were validated on simulated MR datasets and real CT-MR image pairs of the head. Some partial-volume effects occur at the borders of the bone structures and at the bone interfaces with air, which cannot be separated from bone in the MR image. The presented assessment method of CT-MR image registration offers the user a new tool to evaluate the overall and local quality of the registration. With this information, the user does not have to blindly trust the fused CT-MR datasets but can easily identify areas of inaccurate correspondence. The application of the algorithm is so far limited to T1-weighted MR and CT images of the head area.

Algorithms↗
PubMed2004

A simple risk score for prediction of contrast-induced nephropathy after percutaneous coronary intervention: development and initial validation.

OBJECTIVES: We sought to develop a simple risk score of contrast-induced nephropathy (CIN) after percutaneous coronary intervention (PCI). BACKGROUND: Although several risk factors for CIN have been identified, the cumulative risk rendered by their combination is unknown. METHODS: A total of 8,357 patients were randomly assigned to a development and a validation dataset. The baseline clinical and procedural characteristics of the 5,571 patients in the development dataset were considered as candidate univariate predictors of CIN (increase >or=25% and/or >or=0.5 mg/dl in serum creatinine at 48 h after PCI vs. baseline). Multivariate logistic regression was then used to identify independent predictors of CIN with a p value <0.0001. Based on the odds ratio, eight identified variables (hypotension, intra-aortic balloon pump, congestive heart failure, chronic kidney disease, diabetes, age >75 years, anemia, and volume of contrast) were assigned a weighted integer; the sum of the integers was a total risk score for each patient. RESULTS: The overall occurrence of CIN in the development set was 13.1% (range 7.5% to 57.3% for a low [ or=16] risk score, respectively); the rate of CIN increased exponentially with increasing risk score (Cochran Armitage chi-square, p < 0.0001). In the 2,786 patients of the validation dataset, the model demonstrated good discriminative power (c statistic = 0.67); the increasing risk score was again strongly associated with CIN (range 8.4% to 55.9% for a low and high risk score, respectively). CONCLUSIONS: The risk of CIN after PCI can be simply assessed using readily available information. This risk score can be used for both clinical and investigational purposes.

Aged↗
PubMed2006

Identification of a common gene expression signature in dilated cardiomyopathy across independent microarray studies.

OBJECTIVES: This study was designed to identify a common gene expression signature in dilated cardiomyopathy (DCM) across different microarray studies. BACKGROUND: Dilated cardiomyopathy is a common cause of heart failure in Western countries. Although gene expression arrays have emerged as a powerful tool for delineating complex disease patterns, differences in platform technology, tissue heterogeneity, and small sample sizes obscure the underlying pathophysiologic events and hamper a comprehensive interpretation of different microarray studies in heart failure. METHODS: We accounted for tissue heterogeneity and technical aspects by performing 2 genome-wide expression studies based on cDNA and short-oligonucleotide microarray platforms which comprised independent septal and left ventricular tissue samples from nonfailing (NF) (n = 20) and DCM (n = 20) hearts. RESULTS: Concordant results emerged for major gene ontology classes between cDNA and oligonucleotide microarrays. Notably, immune response processes displayed the most pronounced down-regulation on both microarray types, linking this functional gene class to the pathogenesis of end-stage DCM. Furthermore, a robust set of 27 genes was identified that classified DCM and NF samples with >90% accuracy in a total of 108 myocardial samples from our cDNA and oligonucleotide microarray studies as well as 2 publicly available datasets. CONCLUSIONS: For the first time, independent microarray datasets pointed to significant involvement of immune response processes in end-stage DCM. Moreover, based on 4 independent microarray datasets, we present a robust gene expression signature of DCM, encouraging future prospective studies for the implementation of disease biomarkers in the management of patients with heart failure.

Antibody Formation↗
PubMed2004

Mapping high-dimensional data onto a relative distance plane--an exact method for visualizing and characterizing high-dimensional patterns.

We introduce a distance (similarity)-based mapping for the visualization of high-dimensional patterns and their relative relationships. The mapping preserves exactly the original distances between points with respect to any two reference patterns in a special two-dimensional coordinate system, the relative distance plane (RDP). As only a single calculation of a distance matrix is required, this method is computationally efficient, an essential requirement for any exploratory data analysis. The data visualization afforded by this representation permits a rapid assessment of class pattern distributions. In particular, we can determine with a simple statistical test whether both training and validation sets of a 2-class, high-dimensional dataset derive from the same class distributions. We can explore any dataset in detail by identifying the subset of reference pairs whose members belong to different classes, cycling through this subset, and for each pair, mapping the remaining patterns. These multiple viewpoints facilitate the identification and confirmation of outliers. We demonstrate the effectiveness of this method on several complex biomedical datasets. Because of its efficiency, effectiveness, and versatility, one may use the RDP representation as an initial, data mining exploration that precedes classification by some classifier. Once final enhancements to the RDP mapping software are completed, we plan to make it freely available to researchers.

Algorithms↗