SEARCH · Search PubMed

Results for “CLASSIFICATION”

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.

SourceResearch connectionFrom yearThrough year

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

At least 901 records · Page 50Linked to original sources

PubMed2006

Using classification trees to assess low birth weight outcomes.

OBJECTIVE: Low birth weight (LBW) is a major public health problem. Compared to normal weight infants, LBW is positively associated with infant mortality and negatively associated with normative childhood cognitive and physical development. In the past two decades, research has identified important risk factors of LBW. In this study, we used classification trees to study the interactive nature of these factors. In particular we: (1) identify subgroups of women who are at a high risk of a LBW outcome in seven geographical regions of Florida, and (2) study the predictive performance of classification trees by comparing the tree-based results to those obtained using logistic regression. METHODS: The data, 181,690 singleton births, were derived from Florida birth certificates recorded in 1998. Classification trees and logistic regression models were built based on seven geographical regions. The outcome variable consisted of two classes, namely LBW (< 2500 g) and normal birth weight (> or = 2500 g) cases, while a large number of known risk factors was examined. Tree and logistic regression models were compared using Receiving Operating Curves, and sensitivity and specificity analyses. RESULTS: The use of classification trees has revealed a number of high-risk subgroups. For instance, White, Hispanic or Other non-white mothers who were healthy and smoked with a weight gain less than 20 lbs had a higher risk of a LBW birth compared to those with the same characteristics but with a weight gain of more than 20 lbs. Factors such as parity and marital status were important predictors for pregnancy outcomes among nonsmoker White, Hispanic or Other non-white mothers. Furthermore, we found that Black mothers were directly classified as a high-risk subgroup in the regions of Panhandle, Northeast, North Central, while in the Southern regions a series of other characteristics further defined the high-risk subgroup of Black mothers. Overall, the differences in predictive performance between tree models and logistic regression were minimal. CONCLUSION: The present study demonstrated that classification trees can be used to identify high-risk subgroups of mothers who are at risk of LBW outcomes. Although these exploratory tree analyses revealed a number of distinctive variable interactions for each geographical area, the variable selection was similar across all seven regions. This study also demonstrated that classification trees did not outperform logistic regression models or vice versa; both approaches provided useful analyses of the data.

Adolescent↗

PubMed2005

Dynamic CE-MRA for endoleak classification after endovascular aneurysm repair.

AIM: To evaluate the value of dynamic contrast enhanced magnetic resonance angiography (CE-MRA) for classification of endoleaks after endovascular aneurysm repair (EVAR). MATERIALS AND METHODS: Twenty-eight patients, between 2 days and 54 months after EVAR, were evaluated with CTA, MRI and dynamic CE-MRA. The additional diagnostic value of the dynamic 3D CE-MRA was evaluated by determining the ability of the dynamic series in pinpointing the site of inflow of an endoleak. RESULTS: An endoleak was detected in 23 patients. Seventeen of the 23 dynamic series were technically successful (no disturbing artifacts limiting the diagnostic value). Using MRI our findings were: 2 type I, 6 type II, 1 type III, no type IV endoleaks and in 14 cases classification could not be made. The classification results for MRI plus the dynamic CE-MRA were: 2 type I, 12 type II, 1 type III, no type IV endoleaks and in eight cases classification could not be made. In six cases the dynamic MRA allowed classification of the endoleak, which was not possible with the non-dynamic images alone (p=0.091, Fisher exact). CONCLUSION: This pilot study shows that dynamic CE-MRA can have additional value in the classification of endoleaks. Dynamic CE-MRA might obviate the need for diagnostic digital subtraction angiography and aid planning for intervention.

Aged↗

PubMed2004

Non-invasive urothelial neoplasms: according to the most recent WHO classification.

The key points of the latest World Health Organization (WHO) classification of non-invasive urothelial tumors are: the description of the categories has been expanded in the current version to improve their recognition; one group (papillary urothelial neoplasm of low malignant potential) with particularly good prognosis does not carry the label of 'cancer'; it avoids use of ambiguous grading such as grade 1/2 or 2/3 (according to the WHO classification published in 1973, i.e., 1973 WHO classification); the group of non-invasive high grade carcinoma is large enough to contain virtually all those tumors that have biological properties (and a high level of genetic instability) similar to those seen in invasive urothelial carcinoma. This scheme is meant to replace the 1973 WHO classification. Changes in classification have their own inherent problems, tending to lead to confusion, at least for a period of time. From the practical point of view, the use of both the 1973 and the latest WHO classifications is recommended until the latter is sufficiently validated.

Adult↗

PubMed2004

Clinical classification of pulmonary hypertension.

In 1998, during the Second World Symposium on Pulmonary Hypertension (PH) held in Evian, France, a clinical classification of PH was proposed. The aim of the Evian classification was to individualize different categories sharing similarities in pathophysiological mechanisms, clinical presentation, and therapeutic options. The Evian classification is now well accepted and widely used in clinical practice, especially in specialized centers. In addition, this classification has been used by the U.S. Food and Drug Administration and the European Agency for Drug Evaluation for the labeling of newly approved medications in PH. In 2003, during the Third World Symposium on Pulmonary Arterial Hypertension held in Venice, Italy, it was decided to maintain the general architecture and philosophy of the Evian classification. However, some modifications have been proposed, mainly to abandon the term "primary pulmonary hypertension" and to replace it with "idiopathic pulmonary hypertension"; to reclassify pulmonary veno-occlusive disease and pulmonary capillary hemangiomatosis; to update risk factors and associated conditions for pulmonary arterial hypertension and to propose guidelines in order to improve the classification of congenital systemic-to-pulmonary shunts.

Genetic Predisposition to Disease↗

PubMed2005

Reliability of McKenzie classification of patients with cervical or lumbar pain.

BACKGROUND: In the McKenzie system, patients are classified first into syndromes, then into subsyndromes. At present, the reliability of classification with this system is unclear. No study has included patients with cervical pain, and the studies to date have reported conflicting results. OBJECTIVE: The aim of the study is to investigate the interexaminer reliability of the McKenzie classification system for patients with cervical or lumbar pain. SUBJECTS: Fifty patients with spinal pain (25 with lumbar pain and 25 with cervical pain) were included in the study. METHOD: The patients were assessed simultaneously by 2 physical therapists (14 in total) trained in the McKenzie method. Agreement was expressed using the multirater kappa coefficient and percent agreement for classification into (i) syndromes and (ii) subsyndromes. RESULTS: The reliability for syndrome classification was kappa = 0.84 with 96% agreement for the total patient pool, kappa = 1.0 with 100% agreement for lumbar patients, and kappa = 0.63 with 92% agreement for cervical patients. The reliability for subsyndrome classification was kappa = 0.87 with 90% agreement for the total patient pool, kappa = 0.89 with 92% agreement for lumbar patients, and kappa = 0.84 with 88% agreement for the cervical patients. CONCLUSION: The McKenzie assessment performed by persons trained in the McKenzie method may allow for reliable classification of patients with lumbar and cervical pain.

Female↗

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↗

PubMed2004

Neural spike classification using parallel selection of all algorithm parameters.

The Forster-Handwerker template-matching algorithm (J. Neurosci. Methods 31 (1990) 109) classifies neuronal spikes according to three parameters selected by the experimenter prior to running the algorithm. Thousands of different combinations of these parameter values are possible producing hundreds of different classifications for each input file. Using a 40-processor Linux-based parallel computing cluster, we ran their algorithm with an effective sampling of all combinations of parameter values in order to generate a list of the classifications that can be generated by the algorithm. A distance measure was used to quantify the similarity between classifications and then to create a distance table containing entries for the distances between all pairs of classifications. Using a self-organizing neural network (SON) and the distance table we group the classifications by similarity and select the best representative classifications that the Forster-Handwerker algorithm can produce.

Action Potentials↗

PubMed2006

Application and comparison of classification algorithms for recognition of Alzheimer's disease in electrical brain activity (EEG).

The early detection of subjects with probable Alzheimer's disease (AD) is crucial for effective appliance of treatment strategies. Here we explored the ability of a multitude of linear and non-linear classification algorithms to discriminate between the electroencephalograms (EEGs) of patients with varying degree of AD and their age-matched control subjects. Absolute and relative spectral power, distribution of spectral power, and measures of spatial synchronization were calculated from recordings of resting eyes-closed continuous EEGs of 45 healthy controls, 116 patients with mild AD and 81 patients with moderate AD, recruited in two different centers (Stockholm, New York). The applied classification algorithms were: principal component linear discriminant analysis (PC LDA), partial least squares LDA (PLS LDA), principal component logistic regression (PC LR), partial least squares logistic regression (PLS LR), bagging, random forest, support vector machines (SVM) and feed-forward neural network. Based on 10-fold cross-validation runs it could be demonstrated that even tough modern computer-intensive classification algorithms such as random forests, SVM and neural networks show a slight superiority, more classical classification algorithms performed nearly equally well. Using random forests classification a considerable sensitivity of up to 85% and a specificity of 78%, respectively for the test of even only mild AD patients has been reached, whereas for the comparison of moderate AD vs. controls, using SVM and neural networks, values of 89% and 88% for sensitivity and specificity were achieved. Such a remarkable performance proves the value of these classification algorithms for clinical diagnostics.

Aged↗

PubMed2004

Identifying spatial relationships in neural processing using a multiple classification approach.

The application of statistical classification methods to in vivo functional neuroimaging data makes it possible to explore spatial patterns in task-related changes in neural processing. Cluster analysis is one group of descriptive statistical procedures that can assist in identifying classes of brain regions that exhibit similar task-related functionality. In practice, a limitation of cluster analysis is that the performances of clustering algorithms rely on unknown characteristics of the data, making it difficult to determine which procedure best suits a particular analysis. We present a multiple classification approach that incorporates numerous algorithms, evaluates the associated classifications, and either selects a plausible partition relative to the others considered or pools the results from the numerous methods. The multiple classification approach utilizes a new performance criterion, called the relative information (RI) measure, to evaluate the quality of the candidate partitions and as the basis for producing a composite classification image. Employing multiple classifications, rather than a single algorithm, our methodology increases the chance of detecting the functional relationships within the data and, therefore, produces more reliable results. We apply our methodology to a PET study to explore spatial relationships in measured brain function associated with increasing blood alcohol concentration levels, and we perform a simulation study to evaluate the performance of RI.

Alcoholic Intoxication↗

PubMed2004

Stability of patient adaptation classifications on the multidimensional pain inventory.

This study examined the adaptational classification stability of the multidimensional pain inventory (MPI) in two samples of female fibromyalgia syndrome patients. Retest resulted in one-third of patients being assigned to a different classification. Twenty patients had four repeated MPI assessments over a 10-month period; 85% of them changed classification at least once. Prediction of classification stability using demographic variables and measures of pain, depression, anxiety, impression management, and self-deception was unsuccessful. Examination of the MPI Variable Response Scale and an index of the goodness of fit of the cluster for each patient did not yield sufficient predictive power. The implication of this study is that for a sizable number of chronic pain patients, MPI classifications may not be stable, trait-like characterizations. As such, caution must be applied when treatment is tailored to MPI clusters and when classification change is used as an outcome measure.

Adaptation, Psychological↗

PubMed2004

A comparison of the performance characteristics of classification criteria for the diagnosis of psoriatic arthritis.

OBJECTIVE: To compare the accuracy of published classification criteria for the diagnosis of psoriatic arthritis (PsA) and to see whether data-derived classification criteria would be more accurate. METHODS: Data were abstracted from case-note review and radiographic review of patients identified with PsA or rheumatoid arthritis (RA) from 2 clinical disease registers. Each patient was classified according to 7 criteria sets. The test performance characteristics were compared using conditional logistic regression analysis. In an attempt to overcome the problems of the diagnostic gold standard, latent class analysis also was used to calculate test-performance characteristics. Classification and regression-tree methodology was used to derive new criteria and to indicate the diagnostic importance of particular data items, especially rheumatoid factor (RF). RESULTS: Four hundred ninety-nine patients were identified with RA (n=156) or PsA (n=343). Excluding the criteria of Fournie, which could not be applied in 24% of subjects, 446 cases could be classified by all of the other 6 methods. The most sensitive criteria for the diagnosis of PsA were those of Vasey and Espinoza, McGonagle, and Gladman (99%), whereas the others were significantly less sensitive (between 56% and 94%). The specificity of the criteria was high and statistically similar (between 93% and 99%). The Fournie criteria were the most difficult to use, whereas the Vasey and Espinoza and Moll and Wright criteria were the easiest (98% of subjects were able to be classified). A 2-latent class model found very similar test-performance characteristics. Logistic regression and classification and regression-tree models suggested that negative RF was not necessary for diagnosis in the presence of other characteristic features of PsA. CONCLUSIONS: Apart from the Bennett and European Spondyloarthropathy Study Group criteria, which have inadequate sensitivity, the published classification criteria for PsA have similar test-performance characteristics. These data suggest that the criteria proposed by Vasey and Espinoza, Gladman, or McGonagle are the most accurate and feasible in distinguishing between PsA and RA. Relevance International agreement about classification criteria for PsA will assist the interpretation of clinical and epidemiologic research. However, further prospective studies on unselected patients with and without PsA, including controls with non-rheumatoid inflammatory arthritis, are required to confirm these findings.

Arthritis, Psoriatic↗

PubMedDate not supplied

Changing comorbidity classification patterns at radical prostatectomy during a 10-year period.

OBJECTIVE: To investigate the consistency of several comorbidity classifications and concomitant diseases at radical prostatectomy (RP) during a 10-year period. METHODS AND MATERIALS: In 1,297 patients who underwent RP between 1993 and 2002, age and several comorbidity classifications were derived from patient records and assigned to the year of surgery. Trends were evaluated using the Cochran-Armitage trend test. RESULTS: Parallel to an increasing frequency of RPs and a shift toward more organ-confined tumors (P = 0.0094), the proportion of patients aged > or =70 years increased (P = 0.0077). The proportion of the American Society of Anesthesiologists (ASA) Physical Status class 3 increased (P < 0.0001), whereas that of ASA class 1 decreased (P < 0.0001). A Charlson score > or =1 has been assigned with an increasing frequency (P = 0.0008), whereas the trend with a Charlson score of > or =2 did not reach statistical significance (P = 0.07). In contrast to the latter 2 classifications, no significant trends were observed with classifications related to diabetes mellitus and heart disease. CONCLUSIONS: This study shows that the application of the ASA classification may change significantly over time, whereas cardiac and diabetes-related conditions, as well as the Charlson score were apparently less sensitive to changing classification standards in the RP setting.

Aged↗

PubMed2006

Hazard classification of chemicals inducing haemolytic anaemia: An EU regulatory perspective.

Haemolytic anaemia is often induced following prolonged exposure to chemical substances. Currently, under EU Council Directive 67/548/EEC, substances which induce such effects are classified as dangerous and assigned the risk phrase R48 'Danger of serious damage to health by prolonged exposure.' Whilst the general classification criteria for this endpoint are outlined in Annex VI of this Directive, they do not provide specific information to assess haemolytic anaemia. This review produced by the EU Working Group on Haemolytic Anaemia provides a toxicological assessment of haemolytic anaemia and proposes criteria that can be used in the assessment for classification of substances which induce such effects. An overview of the primary and secondary effects of haemolytic anaemia which can occur in rodent repeated dose toxicity studies is given. A detailed analysis of the toxicological significance of such effects is then performed and correlated with the general classification criteria used for this endpoint. This review intends to give guidance when carrying out an assessment for classification for this endpoint and to allow for better transparency in the decision-making process on when to classify based on the presence of haemolytic anaemia in repeated dose toxicity studies. The extended classification criteria for haemolytic anaemia outlined in this review were accepted by the EU Commission Working Group on the Classification and Labelling of Dangerous Substances in September 2004.

Anemia, Hemolytic↗

PubMed2001

Nutritional status classification in the Department of Veterans Affairs.

The Department of Veterans Affairs (VA) Nutrition Status Classification scheme uses clinical data that are routinely collected on admission or shortly thereafter for quick inpatient nutrition screening. In this scheme, patients are assigned to 1 of 4 classification levels according to 7 individual indicators. The indicators include nutrition history, unintentional weight loss as a percent of usual body weight, percent of ideal body weight, diet, diagnosis, albumin, and total lymphocyte count. After ratings (1 to 4) are assigned to each of the 7 indicators, overall nutritional status for each patient is determined by an algorithm. The VA classification system includes many of the same criteria used in other nutritional status classifications. Where it differs is in the greater emphasis on the use of objective criteria and in the rigorous evaluation of reliability and validity that went into its development. Because of these extra measures, the VA classification can be used for prioritizing workload, as well as for determining staff requirements and for comparing workload and productivity across health care facilities. So that others might benefit from using this system, this article provides information on how the classification scheme was developed and explains how it is used.

Algorithms↗

PubMed1987

Body surface maps and the conventional 12-lead ECG compared by studying their performances in classification of old myocardial infarction.

The performance of body surface potential maps and the 12-lead ECG in the detection of old myocardial infarction has been compared in a two-group (54 normals; 52 infarctions) classification procedure (linear discriminant analysis). Three methods for data reduction of body surface maps were compared: 1) time integration, 2) one-step reduction in eigenvectors and 3) two-step reduction in spatial and temporal eigenvectors. Features were taken from the reduction variables by a stepwise selection procedure. From 90% to 93% correct classifications could be obtained using three features from the map data over the initial 30 ms (Q interval) of the QRS wave for all three methods considered. Using the 100 ms (QRS) interval 86% correct classifications were obtained using method 1, and up to 90% and 87% for methods 2 and 3, respectively. In a further analysis the classification based on body surface maps was compared to the one based on the 12-lead ECG. The 12-lead ECG was treated as a restricted set of the body surface mapping leads, so the same methods of data reduction, feature extraction and classification could be applied to both sets of data. Applying method 1 (time integration) 89% correct classifications were obtained using data taken from the 30 ms interval of the 12-lead ECG and a subsequent reduction to three features. When using the 100 ms interval the result was 79% also using three features. The results of method 2 applied to the 12-lead ECG were 89% (30 ms interval, three features) and 78% (100 ms interval, three features).

Electrocardiography↗

PubMed2005

[Oligodendrogliomas: historical background of classifications].

The story of the classifications for gliomas is related to the development of the techniques used for cytological and histological examination of brain parenchyma. After a review of these techniques and the progressive discovery of the central nervous system cell types, the main classifications are presented. The first classification is due to Bailey and Cushing in 1926. It was based on histoembryogenetic theory. Then Kernohan introduced, in 1938, the concept of anaplasia. The WHO classification was published in 1979, then revised in 1993 and 2000. It took into account some data from both previous systems and introduced gradually the notion of histological criteria of malignancy. More recently; molecular genetics data and clinical evolution were retained. The Sainte-Anne classification for oligodendrogliomas is based on both histological and imaging data. It includes the notion of spatial histological structure of oligodendrogliomas. Contrast enhancement is closely related to endotheliocapillary hyperplasia. Gliomas classifications are changing and confusions can be made because of lack of reproductibility and misinterpretations of samples.

Brain Neoplasms↗

PubMed1998

Reproducibility of neuroendocrine lung tumor classification.

For a tumor classification scheme to be useful, it must be reproducible and it must show clinical significance. Classification of neuroendocrine lung tumors is a difficult problem with little information about interobserver reproducibility. We sought to evaluate the classification of typical carcinoid (TC), atypical carcinoid (AC), large-cell neuroendocrine carcinoma (LCNEC), and small-cell carcinoma (SCC) tumors as proposed by W.D. Travis et al (Am J Surg Pathol 15:529, 1991). Forty neuroendocrine tumors were retrieved from the Armed Forces Institute of Pathology (AFIP) files and independently evaluated by five lung pathologists and classified as TC, AC, LCNEC, or SCC (pure SCC, mixed small cell/large cell, and combined SCC). A single hematoxylin and eosin-stained slide from each case was reviewed. Each participant was provided a set of tables summarizing the criteria for separation of the four major categories. Agreement was regarded as unanimous if all five pathologists agreed, a majority if four agreed, and a consensus if three or more pathologists agreed. The kappa statistic was calculated to measure the degree of agreement between two observers. A consensus diagnosis was achieved in all 40 cases (100%), a majority agreement in 31 of 40 (78%), and unanimous agreement in 22 of 40 (55%) of cases. Unanimous agreement occurred in seven of SCC (70%), seven of TC (58%), four of AC (50%), and four of LCNEC (40%). A majority diagnosis was achieved in 11 of 12 (92%) of TC, 9 of 10 (90%) of SCC, 6 of 8 (75%) of AC, and 5 of 10 (50%) of LCNEC. Most of the kappa values were 0.70 or greater, falling into the substantial agreement category. The most common disagreements fell between LCNEC and SCC, followed by TC and AC, and AC and LCNEC. The highest reproducibility occurred for SCC and TC, with disagreement in 8% and 10% of the diagnoses, respectively. For TC, 10% of the diagnoses rendered were AC. For AC, 15% of the diagnoses were rendered as TC, with 2.5% called LCNEC and 2.5% called SCC. For LCNEC, 18% and 4% of the diagnoses were called SCC and AC, respectively. For SCC, 4% of the diagnoses were called AC and 4% were called LCNEC. Thus, using the classification scheme tested, a consensus diagnosis can be achieved for virtually all neuroendocrine lung tumors with substantial agreement between experienced lung pathologists. Classification of NE tumors is most reproducible for classification of TC and SCC but less reproducible for AC and LCNEC. These results indicate a need for more careful definition and application of criteria for TC versus AC and SCC versus LCNEC.

Carcinoid Tumor↗

PubMed1998

Prediction of local recurrence of ductal carcinoma in situ of the breast using five histological classifications: a comparative study with long follow-up.

The increased detection of ductal carcinoma in situ (DCIS) by mammographic screening and the more widespread use of breast-conserving surgery have led to a search for histological features associated with the risk of recurrence. In a case control study of 141 patients with long follow-up, we compared the ability of five morphological classifications to predict recurrence after local excision. A significant correlation was not found between recurrence and growth pattern when a traditional classification based on architecture was used nor with necrosis when a scheme based principally on this feature was employed. A correlation was, however, found between recurrence and "differentiation" as defined by nuclear features and cell polarization in a classification recently formulated by the European Pathologists Working Group (EPWG), but this failed to reach statistical significance at the 5% level. A stronger and statistically significant correlation was found between nuclear grade as defined by the EPWG and recurrence when cell polarization was disregarded, using the classification currently employed by the UK National Health Service and European Commission-funded Breast Screening Programmes. This was attributable to a small number of recurring cases being downgraded as a consequence of exhibiting polarized cells. A significant correlation between histology and recurrence was also observed using the Van Nuys classification, which is based on nuclear grade and necrosis. Whether the tumor recurred as in situ or invasive carcinoma was unrelated to histological classification, as was the time course over which it occurred. These findings strongly support the use of nuclear grade to identify cases of DCIS at high risk of recurrence after local excision, but further work is necessary to determine whether nuclear grade or necrosis is more appropriate to subdivide the non-high-grade cases.

Adult↗