SEARCH · PubMed Health

Results for “Multiple Classification Analysis”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 1,207 records · Page 67Linked to original sources

PubMed1987

Statistical analysis of femoral neck fractures based on 3053 cases.

Femoral neck fractures in 2289 women and 764 men (total 3053) from the years 1975-1977 in Stockholm County Council were retrospectively analyzed regarding social and demographic classification. Seventy-nine percent of the patients (2418 of 3053) were admitted directly from their homes, more commonly in central city areas (82%) than in rural areas (74%). Direct return home was possible in 36%, and 47% could return home after some weeks of institutionalized rehabilitation. There was no increase in fracture incidence found during the three-year period. More fractures occurred on Mondays and less on Sundays than during the rest of the week. The period of December through March showed the highest fracture rate. Diseases other than hip fractures were seen in approximately one-half (1492) of the patients. Internal fixation was performed in 2937 fractures, mainly with the von Bahr screws (1465 cases), Hessel/Nyström multiple pins (674 cases), Thornton nails (380 cases), or Rydell nails (224 cases). Primary hemiarthroplasties were performed in only 102 cases. The detailed analysis of this consecutive population-based femoral neck fracture material is intended as a base for more uniform treatment and rehabilitation of this group of resource-consuming patients.

Age Factors↗

PubMed2005

Implementation of high-dimensional feature map for segmentation of MR images.

A method that considerably reduces the computational and memory complexities associated with the generation of high-dimensional (> or =3) feature maps for image segmentation is described. The method is based on the K-nearest neighbor (KNN) classification and consists of two parts: preprocessing of feature space and fast KNN. This technique is implemented on a PC and applied for generating 3D and 4D feature maps for segmenting MR brain images of multiple sclerosis patients.

Algorithms↗

PubMed1999

Statistical evaluation of clusters derived by nonlinear mapping of EEG spatial patterns.

New methods were devised to improve the discrimination of EEG spatial amplitude patterns recorded from arrays of 64 electrodes placed on visual, auditory or somatic cortex. The 64 traces shared a spatially coherent, aperiodic carrier wave with a spatial pattern of amplitude modulation (AM). Previous observations on AM patterns from rabbits trained to discriminate conditioned stimuli with reinforcement (CS+) and without (CS-) had revealed epochs between the CS and the CR in which AM patterns on CS+ trials could be distinguished from AM patterns on CS- trials. The AM patterns were expressed by points in 64-space that formed clusters. Levels of CS-/CS+ pattern separation were quantified by a pair-wise Euclidean distance method with cross-validation. The present study documents use of the technique for nonlinear mapping (NLM) to project the 64-dimensional structure onto a plane while preserving the relative distances between all points. The goodness of classification by the Euclidean distance measure was the same or improved after projection. Whereas the Euclidean distance measure only gave pair-wise classifications, the planar displays showed the patterns for multiple clusters simultaneously. These NLM-based methods revealed previously unrecognized structures within distributions of AM patterns in sensory cortices in the time period between the CS and CR.

Animals↗

PubMedDate not supplied

Improving tissue classification in MRI: a three-dimensional multispectral discriminant analysis method with automated training class selection.

PURPOSE: To improve the reliability, accuracy, and computational efficiency of tissue classification with multispectral sequences [T1, T2, and proton density (PD)], we developed an automated method for identifying training classes to be used in a discriminant function analysis. We compared it with a supervised operator-dependent method, evaluating its reliability and validity. We also developed a fuzzy (continuous) classification to correct for partial voluming. METHOD: Images were obtained on a 1.5 T GE Signa MR scanner using three pulse sequences that were co-registered. Training classes for the discriminant analysis were obtained in two ways. The operator-dependent method involved defining circular ROIs containing 5-15 voxels that represented "pure" samples of gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF), using a total of 150-300 voxels for each tissue type. The automated method involved selecting a large number of samples of brain tissue with sufficiently low variance and randomly placed throughout the brain ("plugs"), partitioning these samples into GM, WM, and CSF, and minimizing the amount of variance within each partition of samples to optimize its "purity." The purity of the plug was estimated by calculating the variance of 8 voxels in all modalities (T1, T2, and PD). We also compared "sharp" (discrete) measurements (which classified tissue only as GM, WM, or CSF) and "fuzzy" (continuous) measurements (which corrected for partial voluming by weighting the classification based on the mixture of tissue types in each voxel). RESULTS: Reliability was compared for the operator-dependent and automated methods as well as for the fuzzy versus sharp classification. The automated sharp classifications consistently had the highest interrater and intrarater reliability. Validity was assessed in three ways: reproducibility of measurements when the same individuals were scanned on multiple occasions, sensitivity of the method to detecting changes associated with aging, and agreement between the automated segmentation values and those produced through expert manual segmentation. The sharp automated classification emerged as slightly superior to the other three methods according to each of these validators. Its reproducibility index (intraclass r) was 0.97, 0.98, and 0.98 for total CSF, total GM, and total WM, respectively. Its correlations with age were 0.54, -0.61, and -0.53, respectively. Its percent agreement with the expert manually segmented tissue for the three tissue types was 93, 90, and 94%, respectively. CONCLUSION: Automated identification of training classes for discriminant analysis was clearly superior to a method that required operator intervention. A sharp (discrete) classification into three tissue types was also slightly superior to one that used "fuzzy" classification to produce continuous measurements to correct for partial voluming. This multispectral automated discriminant analysis method produces a computationally efficient, reliable, and valid method for classifying brain tissue into GM, WM, and CSF. It corrects some of the problems with reliability and computational inefficiency previously observed for operator-dependent approaches to segmentation.

Adult↗

PubMedDate not supplied

Supervised self-organizing maps in drug discovery. 1. Robust behavior with overdetermined data sets.

The utility of the supervised Kohonen self-organizing map was assessed and compared to several statistical methods used in QSAR analysis. The self-organizing map (SOM) describes a family of nonlinear, topology preserving mapping methods with attributes of both vector quantization and clustering that provides visualization options unavailable with other nonlinear methods. In contrast to most chemometric methods, the supervised SOM (sSOM) is shown to be relatively insensitive to noise and feature redundancy. Additionally, sSOMs can make use of descriptors having only nominal linear correlation with the target property. Results herein are contrasted to partial least squares, stepwise multiple linear regression, the genetic functional algorithm, and genetic partial least squares, collectively referred to throughout as the "standard methods". The k-nearest neighbor (kNN) classification method was also performed to provide a direct comparison with a different classification method. The widely studied dihydrofolate reductase (DHFR) inhibition data set of Hansch and Silipo is used to evaluate the ability of sSOMs to classify unknowns as a function of increasing class resolution. The contribution of the sSOM neighborhood kernel to its predictive ability is assessed in two experiments: (1) training with the k-means clustering limit, where the neighborhood radius is zero throughout the training regimen, and (2) training the sSOM until the neighborhood radius is reduced to zero. Results demonstrate that sSOMs provide more accurate predictions than standard linear QSAR methods.

Algorithms↗

PubMed1994

Clinico-pathological criteria with prognostic relevance in hepatoblastoma.

We investigated clinical data and histological specimens of 46 patients with a hepatoblastoma (HB) for prognostic criteria. Disease-free survival (DFS) of 23 patients treated in the German Cooperative Study HB-89 (1988-1990) was 83%, in contrast to 40% in 10 children with other chemotherapy regimes (1977-1987) and 38% in 13 with only a tumour resection (P = 0.005). Tumour residence after resection (R category) correlated significantly with probability of DFS (P = 0.0001). This was also the case for pT status, according to the pTNM classification for liver carcinoma (P = 0.0007), involvement of one or both liver lobes (P = 0.004), multiplicity of tumour nodes (P = 0.001), vascular invasion (P = 0.0006) and expression of nucleolar organiser regions as an indicator for proliferation activity of tumour cells (P = 0.05). Patients' age and histopathological subtypes could only indicate outcome, while tumour size and serum alpha-fetoprotein values were not significantly related to prognosis. In multivariate analysis, pT status and R categories remained significant. These should be applied in all cooperative trials on HB.

Child↗

PubMed2000

Towards a covering set of protein family profiles.

Evolutionary classification leads to an economical description of the protein sequence universe because attributes of function and structure are inherited in protein families. Efficient strategies of functional and structural genomics therefore target one representative from each family. Enumerating all families and establishing family membership consistently based on sequence similarities are nontrivial computational problems. Emerging concepts and caveats of global sequence clustering are reviewed. Explicit multiple alignments coupled with neighbourhood analysis lead to domain segmentation, and hierarchical unification helps to resolve conflicts and validate clusters. Eventually, every part of every sequence will be assigned to a domain family which is uniquely associated with a fold and a molecular function.

Amino Acid Sequence↗

PubMed1995

A controlled multivariate chart review of multiple complex developmental disorder.

OBJECTIVE: The primary aim of the study was to ascertain the usefulness and the validity of the set of criteria proposed to define a subgroup within the DSM-III-R category of pervasive developmental disorder-not otherwise specified under the name of multiple complex developmental disorder (MCDD). METHOD: A multivariate analysis (cluster and principal-components analysis) was performed on the characteristics, reliably extracted from the charts of 105 children with MCDD, 32 with autistic disorder, 51 with externalizing disorders, and 56 with internalizing disorders, all with an IQ greater than 70, fully assessed in our department between 1984 and 1991. RESULTS: The main finding was a strong correspondence between the classification of the cases by DSM-III-R categories and the subdivision by means of a multivariate cluster analysis. The cluster analysis did not discriminate between children with emotional and disruptive disorders. Furthermore, children with MCDD and autistic disorder were significantly different both on symptom factors ("psychotic thinking/anxiety," "aggression," "deficient interaction/communication," "stereotyped and rigid behavior," and "suspiciousness/odd interaction") and on factors that reflected developmental and environmental background variables. CONCLUSION: The results of this study add to the nosology of autistic spectrum disorders and lend additional support to the need for a separate subcategory of MCDD within DSM-V. Further corroboration of these results in independent (multicenter) samples will be required.

Autistic Disorder↗

PubMed1975

The prediction of glaucoma from ocular biometric data. Part 1. An application of multiple regression analysis.

Two sets of multiple regression equations (prediction systems) were derived from the analysis of ocular biometric data obtained from glaucoma patients (16 open angle; 16 angle-closure) and 75 normal subjects. Discriminant scores were established for both sets of equations which minimised the number of false negatives. One set, the 'Glaucoma Equations' was applied to the data to segregate the glaucoma from the normal subjects. The other prediction system, the 'Classification Equations,' was then applied to the group defined as glaucomatous to discriminate between patients in the angle-closure and open angle categories. The performance of these equations, obtained by comparing the predicted and actual classifications for this sample, was such that between 9 and 12% of false positives and 0 and 3% false negatives were found on the 'Glaucoma Equations' and between 6 and 12% of false positives with no false negatives on the 'Classification Equations.'

Age Factors↗

PubMed1995

Final evaluation of clubfoot.

After analyzing the multiple evaluations described in the literature, we propose an evaluation that can be used during growth. The global score is based on 50 points. The normal foot is scored 0. The functional evaluation (passive motion, muscle function, and gait analysis) is based on 28 points. The morphology is based on 12 points. The radiographic analysis is scored on 10 points. The final evaluation score is linked to the initial classification at birth. The gain during the treatment is the difference between the score at birth and the score of the final evaluation. It is noted as a percentage.

Adolescent↗

PubMed2005

Multi-metric and multi-substructure biclustering analysis for gene expression data.

A good number of biclustering algorithms have been proposed for grouping gene expression data. Many of them have adopted matrix norms to define the similarity score of a bicluster. We shall show that almost all matrix metrics can be converted into vector norms while preserving the rank equivalence. Vector norms provide a much more efficient vehicle for biclustering analysis and computation. The advantages are two folds: ease of analysis and saving of computation. Most existing biclustering algorithms have also implicitly assumed the use of univariate (i.e., single metric) evaluation for identifying biclusters. Such an approach however overlooks the fundamental principle that genes (even though they may belong to the same gene group) (1) may be subdivided into different substructures; and (2) they may be co-expressed via a diversity of coherence models (a gene may participate in multiple pathways that may or may not be co-active under all conditions). The former leads to the adoption of a multi-substurcture analysis, while the latter to the multivariate analysis. This paper will show that the proposed multivariate and multi-subscluster analysis is very effective in identifying and classifying biologically relevant groups in genes and conditions. For example, it has successfully yielded highly discriminant and accurate classification based on known ribosomal gene groups.

Algorithms↗

PubMed2002

Phenotypic and genotypic characterization of serologically atypical strains of Shigella flexneri type 4 isolated in Dhaka, Bangladesh.

Twenty-one atypical Shigella flexneri type 4 strains isolated from patients attending the Dhaka treatment center of the International Centre for Diarrhoeal Disease Research, Bangladesh, were extensively characterized and compared with S. flexneri serotypes 4a and 4b. The atypical strains agglutinated only with the type antigen factor 4 and did not agglutinate with any group factors, thereby excluding their characterization into serotype 4a or 4b. Of the 21 strains, 85.7% did not ferment mannitol but were able to ferment most of the sugars, whereas the remaining 14.3% strains fermented mannitol but were unable to ferment most of the sugars. Most of the strains were resistant to ampicillin, tetracycline, and trimethoprim-sulfomethoxazole. All of the strains harbored the 140-MDa plasmid, had the ipaH gene, had the sen gene (encoding Shigella enterotoxin 2), had the ability to bind Congo red, and were positive for keratoconjunctivitis in the guinea pig eye, attesting their invasive properties. All of the strains contained a middle-range plasmid (35 to 62 MDa) as well as a number of stable small plasmids, yielding mainly two plasmid profiles which were different from those of 4a and 4b strains. Conjugation and curing experiments suggested that the middle-range plasmids harbored a self-transferable multiple antibiotic resistance marker. Pulsed-field gel electrophoresis analysis of all of the tested strains yielded two types with numerous subtypes, whereas ribotyping yielded only two types which were completely different from those of types 4a and 4b. This study concluded that two different clones of atypical S. flexneri type 4 exist and strongly suggests that these are new subserotypes of S. flexneri that await further serological classification.

Animals↗

PubMedDate not supplied

[Neurotic depression versus endogenous depression: are they part of the same disorder? (Controlled study)].

Authors, studying in patients with RDC-endogenous major depression and distinct quality of depressed mood, using a Factor Analysis and two Stepwise Multiple Regression Analysis with data: 17 items of the Hamilton Rating Scale for Depression, its total score, age, number of episode, presence/absence of life events, Newcastle Index, EPI-Extroversion and EPI-Neuroticism. In the sample two profiles are found, in agreement with topics for "neurotic" and "endogenous" depression. More "neurotic" subject suffering the first episode, and more "endogenous" the others. Authors believe confirmed their hypothesis: that classification show the same illness in two different evolutive moments.

Adult↗

PubMed2003

A comprehensive approach to the analysis of matrix-assisted laser desorption/ionization-time of flight proteomics spectra from serum samples.

For our analysis of the data from the First Annual Proteomics Data Mining Conference, we attempted to discriminate between 24 disease spectra (group A) and 17 normal spectra (group B). First, we processed the raw spectra by (i) correcting for additive sinusoidal noise (periodic on the time scale) affecting most spectra, (ii) correcting for the overall baseline level, (iii) normalizing, (iv) recombining fractions, and (v) using variable-width windows for data reduction. Also, we identified a set of polymeric peaks (at multiples of 180.6 Da) that is present in several normal spectra (B1-B8). After data processing, we found the intensities at the following mass to charge (m/z) values to be useful discriminators: 3077, 12 886 and 74 263. Using these values, we were able to achieve an overall classification accuracy of 38/41 (92.6%). Perfect classification could be achieved by adding two additional peaks, at 2476 and 6955. We identified these values by applying a genetic algorithm to a filtered list of m/z values using Mahalanobis distance between the group means as a fitness function.

Blood Proteins↗

PubMed1987

[Determinants of plasma digoxin and digitoxin concentrations in elderly patients. A multivariate analysis].

In 1063 patients (greater than or equal to 60 years, 531 men, 532 women) the plasma concentration during digitalis maintenance therapy (metildigoxin, n = 356, beta-acetyldigoxin, n = 359, and digitoxin, n = 348) was determined and related to sex, age, body weight, serum potassium, renal function and the prescribed daily maintenance dose. Classification of treatment groups according to renal function (Crea less than or equal to 1.3 mg/dl parallel greater than 1.3 mg/dl) did not show any difference of the mean maintenance doses. In multiple linear regression analyses only a weak relationship between plasma digitalis concentration and the studied variables was found, which could be equally attributed to dose, creatinine and serum potassium in the digoxin derivative groups, whereas for digitoxin only body weight had a significant effect on the plasma concentration. During a maintenance dose of 0.07 or 0.1 mg/die which was given to 87% of patients in the digitoxin group, 70% were found to have plasma levels within the therapeutic range.

Acetyldigoxins↗

PubMed2001

High order statistics for image classification.

This paper addresses the problem of image classification using local information which is aggregated to provide global representation of different image classes. Local information is adaptively extracted from an image database using Independent Component Analysis (ICA) which provides a set of localized, oriented, and band-pass filters selective to independent features of the images. Local representation using ICA techniques has been previously investigated by several researchers. However, very little work has been done on further use of these representations to provide more complex and global description of images. In this paper, we present an algorithm which uses the energy of a minimal set of ICA filters to provide class-specific signatures which are shown to be strongly discriminant. Computer simulations are carried on two image databases, one consisting of five classes--referred to as categories--(buildings, rooms, mountains, forests and beaches) and one consisting of a set of 30 objects from multiple views for viewpoint invariant object recognition. The classification performance of the algorithm using both Independent and Principal Component Analyses are reported and discussed.

Image Processing, Computer-Assisted↗

PubMed2004

[Azathioprine-associated myelodysplastic syndrome with cytogenetic aberrations].

HISTORY AND ADMISSION FINDINGS: A 49-years-old female with a history of multiple sclerosis and on medication with azathioprine over 5 years amounting to a cumulative dosage of 45 g presented with fatigue and sinus tachycardia. INVESTIGATIONS: Haematological analysis revealed a pancytopenia with a sustained normochromic anemia (hemoglobin 6,2 g/dl), a mild leukopenia (leucocytes 3500/ microl) and thrombocytopenia (platelets 22000/ microl) requiring platelet substitution. Bone marrow aspirate revealed a dysplasia of all three lineages with reduced thrombopoiesis and ineffective erythropoiesis. Cytogenetic analysis detected a complex aberrant clone including loss of the critically deleted regions in 5q31 and 7q31, as well as structural changes of 12p. The diagnosis was refractory cytopenia with multilineage dysplasia according to the WHO classification of myelodysplastic syndromes. TREATMENT AND COURSE: While allogeneic sibling transplantation was planned the patient developed spontaneous recurrent subdural haematoma and died due to persistent bleeding refractory to platelet substitution. CONCLUSION: Treatment with azathioprine may - depending on its cumulative dosage - lead to pancytopenia and subsequent development of myelodysplasia or secondary leukaemia, respectively. Complex genetic alterations involving chromosome 7 are characteristic. Long-lasting treatment with azathioprine needs critical assessment.

Anemia↗

PubMed1995

Pediatric myelodysplasia: a study of 68 children and a new prognostic scoring system.

Clinical, morphologic, and cytogenetic features were examined in a group of 68 children with myelodysplasia (MDS) referred to a single institution between 1971-1991. The morphologic French-American-British (FAB) system of classification proved of limited value in this group of patients because 50% of the cases were categorized as chronic myelomonocytic leukemia and three patients with eosinophilia and MDS were unclassifiable. Cytogenetic analysis was performed in 63 cases and clonal abnormalities were detected in 55%; the most common chromosome involved was number 7. Modification of the FAB system to incorporate additional diagnostic features such as pretreatment fetal hemoglobin (Hb F) and cytogenetics allowed incorporation of the categories of juvenile chronic myeloid leukemia (JCML) and infantile monosomy 7 syndrome (IMo7). The resulting groups of patients had highly significant differences in survival (P = .00009). The overall 5-year survival for the patients was 31.9% (95% CI 21.7 to 44.1) and factors influencing prognosis included: modified FAB type, platelet count, Hb F level, and cytogenetic complexity. We developed a scoring system ("FPC") where each of the following findings at diagnosis scored one point: HbF greater than 10%, platelets < or = 40 x 10(9)/L, and complex karyotypic changes (two or more clonal structural/numerical abnormalities), which produced groups with highly significant differences, patients with a score of 0 having a 5-year survival of 61.6% (CI 33% to 84%), whereas those with a score of two or three all died within 4 years of diagnosis. The revised classification and scoring system may prove helpful in making treatment choices in pediatric MDS and now needs to be tested prospectively in large scale population-based studies.

Abnormalities, Multiple↗