SEARCH · PubMed Health

Results for “cluster 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 343 records · Page 19Linked to original sources

PubMed2002

[Identification of fructus broussonetiae from different habitats and its confused species with clustering analysis by near-infrared diffuse reflectance spectrometry].

OBJECTIVE: To provide a new method for identifying Fructus Broussonetiae and its confused species. METHODS: Cluster analysis was adopted for the identification. RESULTS: The method can identify Fructus Broussonetiae and the results are coincident with traditional phytotaxnomy. CONCLUSION: This method can be used for the rapid and accurate differentiation of Fructus Broussonetiae.

Broussonetia↗

PubMedDate not supplied

[Cluster analysis in biogeographical classifications].

Some unavoidable methodic and methodological problems arising at each stage of the classification of objects by the methods of hierarchical clustering are shown on the concrete (biogeographical) and abstract--numerical examples. Reasons which cause these problems and possible ways of minimization of cluster analysis artifacts are discussed. Unavoidable constrains in interpretations of dendrograms obtained by means of agglomerate algorithms are indicated.

Algorithms↗

PubMed1997

Adaptive classification of two-dimensional gel electrophoretic spot patterns by neural networks and cluster analysis.

The interpretation of two-dimensional gel electrophoresis spot profiles can be facilitated by statistical and machine learning programs. Two different approaches to classification of spot profiles - cluster analysis and neural networks - are discussed. Neural networks for two different model patterns were designed and an algorithm for training of the net for the classification was developed. It was shown that the performance of neural networks is higher compared to cluster and principal component analysis. The possibility of combining both approaches into one process can increase reliability and speed of classification. Artificially created training sets with added random noise can be used for network training. The analysis was applied on the Streptomyces coelicolor developmental two-dimensional (2-D) gel database.

Cluster Analysis↗

PubMed1990

Magnetic resonance imaging of peripheral nerve sheath tumors. Assessment by numerical visual fuzzy cluster analysis.

A retrospective, nonblinded review of ten nerve sheath tumors (four malignant) selected for pathologic proof and complete magnetic resonance (MR) evaluation was performed to assess the primary tumor location, signal pattern, and extent of reactive zone. A modification of visual fuzzy cluster analysis (VFCA) that emphasized the number of visual fuzzy clusters in each mass was developed to assess the neural tumors. The MR findings were correlated with the findings at surgery and histopathology. There were six men and four women, aged 19 to 62 years (mean, 43). Nine tumors involved the lower extremity. In all tumors, MRI correctly identified the nerve trunk of origin. Tumor dimensions were generally overestimated by MRI. Three internal signal patterns were observed: homogeneous (1/1 benign), finitely inhomogeneous (5/5 benign), and hectically inhomogeneous (4/4 malignant). The number of visual fuzzy clusters (VFCRs) for each sequence did not allow reliable separation of benign and malignant entities, but when considered in aggregate, benign and malignant lesions segregated in different clusters. This implies that the likelihood of malignancy increases as the number of MR-identifiable tissue types per lesion increase. Three types of reaction (edema) were observed best on long repetition time/echo time (TR/TE) sequences, confined to immediate peritumoral region, intracompartmental, and extracompartmental. The first two patterns correlated well with clinicopathologic findings; however, the third pattern did not. Separation of indolent (benign) cellular masses from aggressive (malignant) ones by MR characteristics is difficult but VFCA shows promise for aiding this differentiation and deserves further investigation in larger study populations.

Adult↗

PubMed1994

Risk profiles for nursing home placement of rural elderly: a cluster analysis of psychogeriatric indicators.

In an effort to better understand the clinical and functional status of patients served by our Rural Elder Outreach Program, more effectively identify risk groups, and more efficiently target services, we performed a cluster analysis on 92 older adults served by our program. The first cluster included patients with very poor health, mild cognitive impairment, very high care demands, and migrating toward active risk for institutionalization. The second cluster included patients with poor physical but good mental health, intact cognition, high care demands, and at passive risk. The third cluster comprised patients with high functional, physical, and cognitive impairment, intensive care demands, moderate mental health problems, poor insight into their situation, and at active risk for institutionalization.

Activities of Daily Living↗

PubMed2006

[A study on clustering analysis of arrhythmias].

According to the characteristics of ECG analysis, large data quantum, high accuracy demand and real-time, a classified algorithm of arrhythmia based on clustering analysis is presented in this paper. According to "things-of-one-kind-come-together" principle, this algorithm uses the similarities of cases with same kind of heart disease at the same time, includes the factors of the individual difference to analyze arrhythmias by clustering QRS complex waveform and rhythm analysis as the subordinate method. Verified by eight records of MIT-BIR standard heart electricity database, the probability of correct clustering reaches above 90%, which shows that this algorithm can analyze arrhythmias effectively.

Algorithms↗

PubMed2004

Cluster analysis of gene expression data based on self-splitting and merging competitive learning.

Cluster analysis of gene expression data from a cDNA microarray is useful for identifying biologically relevant groups of genes. However, finding the natural clusters in the data and estimating the correct number of clusters are still two largely unsolved problems. In this paper, we propose a new clustering framework that is able to address both these problems. By using the one-prototype-take-one-cluster (OPTOC) competitive learning paradigm, the proposed algorithm can find natural clusters in the input data, and the clustering solution is not sensitive to initialization. In order to estimate the number of distinct clusters in the data, we propose a cluster splitting and merging strategy. We have applied the new algorithm to simulated gene expression data for which the correct distribution of genes over clusters is known a priori. The results show that the proposed algorithm can find natural clusters and give the correct number of clusters. The algorithm has also been tested on real gene expression changes during yeast cell cycle, for which the fundamental patterns of gene expression and assignment of genes to clusters are well understood from numerous previous studies. Comparative studies with several clustering algorithms illustrate the effectiveness of our method.

Algorithms↗

PubMed1999

Fuzzy cluster analysis of positive stress tests: comparison with stress echocardiography and nuclear perfusion imaging in patients with triple vessel and left main coronary disease.

Fuzzy cluster analysis (FCA) was used to classify 166 outpatient positive treadmill stress tests as mildly, moderately, or severely abnormal. The method combines ST-segment change with five other stress test variables, and then computes a similarity measure to determine how closely each patient's stress test resembles a prototypical mildly, moderately, or severely abnormal stress test. All patients had coronary angiography within 1 month of their stress tests. For the 45 patients with triple vessel disease (TVD), FCA classified 34 of these stress tests as severely abnormal (sensitivity = 75%). For the 22 patients with left main disease (LM), FCA classified 19 stress tests as severely abnormal (sensitivity = 86%). For the combined group with high-grade disease (TVD + LM), the sensitivity was 79%. A literature review shows that for stress echocardiography, multiple exercise-induced wall motion abnormalities have a sensitivity in the 70-80% range for patients with high-grade disease. For nuclear stress testing, the high-risk pattern of multiple reversible defects, with or without increased lung uptake, has a sensitivity in the range of 70-80% for patients with high-grade disease. Thus classification of a positive stress test as severely abnormal by FCA has a sensitivity comparable to high-risk patterns on stress echocardiography and nuclear stress testing in patients with TVD or LM.

Adult↗

PubMed1989

The role of cluster analysis on traditional cephalometric dimensions.

In order to categorize (classify) craniofacial forms, varying linear dimensional arrays from a series of pretreatment cephalographs ('A' records) were subjected to cluster analysis. The derived subgroups (clusters) not only showed inconsistencies in their component patients, but also their 'nearest neighbors', i.e. cases identified as being most similar to one another. This study, therefore, emphasized the need to devise more appropriate cephalometric appraisal techniques for patient categorization (diagnosis).

Adolescent↗

PubMed1998

[Identification of polygenes in the system achaete-scute in Drosophila melanogaster using cluster analysis].

Data on location of mobile elements mdg1, Dm412, copia, and B104 in 33 isogenic lines of Drosophila melanogaster has been processed by means of cluster analysis to reveal the relationship between the penetrance for bristle reduction and the distribution of mobile elements. The presence of two groups of sites specific for lines with contrasting penetrance levels have been demonstrated. The specificity suggests that the sites can be associated with the location of corresponding polygenes, affecting the penetrance level in mutant lines.

Animals↗

PubMed1993

[Revelations of the origin of Chinese nation from clustering analysis and frequency distribution of HLA polymorphism in major minority nationalities in mainland China].

This paper is a joint report on the analysis of polymorphism of HLA class I, II and III antigens in Chinese major minority nationalities (Miao, Buyi, Mongol of Inner Mongolia, Man, Hui, Tibetan and Uygur), and southern and northern Han samples as a part of the workshop organized by the 11th International Histocompatibility Workshop and Conference (IHWC) held on 6-13 November 1991 in Yokohama. Clustering analysis showed that Miao and Buyi cluster with the southern Han and Mongol, Man, Hui and Tibetan cluster around the northern Han, suggesting that the Chinese nation comprises northern and southern two major populations. Uygur clusters with a large group of Caucasoid samples from very far point. Frequencies of HLA alleles of Caucasoid origin (A3, B8 and others) decrease eastwards and southwards, and those of HLA alleles of Southeast Asia Mongoloid origin (B46 and others) decrease northwards. The gradient distribution of HLA allele frequencies suggests the unceasing migration and mutual amalgamation of our ancestors since the remote pre-historic times. From the fact that Chinese nation comprises northern and southern two major populations, this paper discussed the possible revelations of the origin of Chinese nation.

Asian People↗

PubMed2003

[The use of cluster analysis in clinico-neurophysiological study].

Neurophysiological characteristic--an amplitude of P300 auditory evoked potentials (EP) wave registered in 12 and 5 EEG leads--was studied in 22 patients with endogenous psychoses. Using cluster analysis, it was shown that the differences between the EP amplitudes before and after treatment split into clusters thus indicating a fine structure of the data, reflecting an objective changes in the patient's state during the treatment. The results obtained correlated with an evaluation of patient' clinical state by the PANSS, namely with scores on general psychopathological and positive symptoms subscales.

Cluster Analysis↗

PubMed2004

Imaging of colorectal adenocarcinoma using FT-IR microspectroscopy and cluster analysis.

In this paper, three different clustering algorithms were applied to assemble infrared (IR) spectral maps from IR microspectra of tissues. Using spectra from a colorectal adenocarcinoma section, we show how IR images can be assembled by agglomerative hierarchical (AH) clustering (Ward's technique), fuzzy C-means (FCM) clustering, and k-means (KM) clustering. We discuss practical problems of IR imaging on tissues such as the influence of spectral quality and data pretreatment on image quality. Furthermore, the applicability of cluster algorithms to the spatially resolved microspectroscopic data and the degree of correlation between distinct cluster images and histopathology are compared. The use of any of the clustering algorithms dramatically increased the information content of the IR images, as compared to univariate methods of IR imaging (functional group mapping). Among the cluster imaging methods, AH clustering (Ward's algorithm) proved to be the best method in terms of tissue structure differentiation.

Adenocarcinoma↗

PubMedDate not supplied

Robust growing neural gas algorithm with application in cluster analysis.

We propose a novel robust clustering algorithm within the Growing Neural Gas (GNG) framework, called Robust Growing Neural Gas (RGNG) network.The Matlab codes are available from . By incorporating several robust strategies, such as outlier resistant scheme, adaptive modulation of learning rates and cluster repulsion method into the traditional GNG framework, the proposed RGNG network possesses better robustness properties. The RGNG is insensitive to initialization, input sequence ordering and the presence of outliers. Furthermore, the RGNG network can automatically determine the optimal number of clusters by seeking the extreme value of the Minimum Description Length (MDL) measure during network growing process. The resulting center positions of the optimal number of clusters represented by prototype vectors are close to the actual ones irrespective of the existence of outliers. Topology relationships among these prototypes can also be established. Experimental results have shown the superior performance of our proposed method over the original GNG incorporating MDL method, called GNG-M, in static data clustering tasks on both artificial and UCI data sets.

Algorithms↗

PubMed2003

Genomic gene clustering analysis of pathways in eukaryotes.

Genomic clustering of genes in a pathway is commonly found in prokaryotes due to transcriptional operons, but these are not present in most eukaryotes. Yet, there might be clustering to a lesser extent of pathway members in eukaryotic genomes, that assist coregulation of a set of functionally cooperating genes. We analyzed five sequenced eukaryotic genomes for clustering of genes assigned to the same pathway in the KEGG database. Between 98% and 30% of the analyzed pathways in a genome were found to exhibit significantly higher clustering levels than expected by chance. In descending order by the level of clustering, the genomes studied were Saccharomyces cerevisiae, Homo sapiens, Caenorhabditis elegans, Arabidopsis thaliana, and Drosophila melanogaster. Surprisingly, there is not much agreement between genomes in terms of which pathways are most clustered. Only seven of 69 pathways found in all species were significantly clustered in all five of them. This species-specific pattern of pathway clustering may reflect adaptations or evolutionary events unique to a particular lineage. We note that although operons are common in C. elegans, only 58% of the pathways showed significant clustering, which is less than in human. Virtually all pathways in S. cerevisiae showed significant clustering.

Animals↗

PubMed2001

Subsets in psoriatic arthritis formed by cluster analysis.

The aim of the study was to create subgroups among psoriatic arthritis patients on the basis of dermatological features, clinical pattern of arthritis, and laboratory, immunological and radiological findings. Data on 100 patients were expressed in a standardised form and entered into hierarchical cluster analysis according to Ward's method. Seven subgroups were created. Fifty-six patients with mild psoriasis were sorted into a 'polyarticular group'. Two 'RA-like groups' were formed, differing from each other serologically and in axial involvement. In an 'oligoarticular group' (18 patients) serious skin disease and female gender predominancy were found to be characteristic. Eight patients with polyarticular arthritis were assigned to an 'erythrodermal group', in which polyarticular arthritis, mutilating, severe arthritis and a history of erythroderma were characteristic. Close to this group on the dendrogram eight women were sorted into a 'distal form'. Sausage fingers were frequent, and nail dystrophy was present in every case. In a 'pustular group' (three patients) the different type of skin involvement was considered and nail dystrophy was common. In the newly created subgroups not only the arthritic status, but also the type of the skin disease, played a determining role.

Adult↗

PubMed2005

Performance evaluation of functional medical imaging compression via optimal sampling schedule designs and cluster analysis.

In previous work we have described a technique for the compression of positron emission tomography (PET) image data in the spatial and temporal domains based on optimal sampling schedule designs (OSS) and cluster analysis. It can potentially achieve a high data compression ratio greater than 80:1. However, the number of distinguishable cluster groups in dynamic PET image data is a critical issue for this algorithm that has not been experimentally analyzed on clinical data. In this paper, the problem of experimentally determining the ideal cluster number for the algorithm for PET brain data is addressed.

Algorithms↗

PubMed1991

Evaluation of monoclonal antibodies to Listeria monocytogenes flagella by checkerboard ELISA and cluster analysis.

A simple ELISA technique is described which utilizes whole bacteria as solid phase antigens for the evaluation of monoclonal antibodies against flagellar proteins of Listeria monocytogenes. Cluster analysis of the ELISA extinction values gave distinct reaction patterns from 23 tested supernatants. The method offers the possibility of screening supernatants without the necessity for standardization of antibody concentration or the labelling of established monoclonal antibodies.

Animals↗