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 181 records · Page 10Linked to original sources

PubMed1980

Predicting speech discrimination from the audiometric thresholds.

To develop a method for predicting a speech discrimination score (SDS) from audiometric thresholds (SRT and pure-tones) three prediction systems were investigated: a stepwise multiple regression procedure, smear-and-sweep analysis and a clinical classification of the audiometric configuration. Test results of 529 ears with sensorineural hearing loss were taken from copies of audiograms obtained as part of a normal audiology clinic caseload. The three prediction systems had similar predictive ability and yielded slightly higher correlations with the SDS than those in previously reported studies. Squared correlations in this study ranged from 0.58 to 0.60. Smear-and-sweep analysis yielded the best results; however, its complexity makes clinical application difficult at this time. The stepwise multiple regression models or the clinical classification system provided more clinically useful methods for predicting the SDS. An over-riding influence of increasing variability in the SDS with increased hearing loss was observed and significantly limited the accuracy of prediction for the moderate-to-severe hearing loss groups. Small changes in the slope of the audiometric configuration were noted to affect the SDS only when the degree of hearing loss was slight.

Audiometry↗

PubMed2000

Computer aided gnathosonic analysis: distinguishing between single and multiple tooth impact sounds.

Watt developed a classification of tooth contact sounds that distinguished between the short sharp, reproducible sounds heard when the teeth meet simultaneously and the dull prolonged, poorly reproducible sounds heard when tooth contacts are sequential. However, when a large occlusal prematurity, for instance a high restoration, is introduced, tooth contact sounds are also short sharp and highly reproducible. In this study, a method of distinguishing single from multiple tooth contact sounds is described, based on an analysis of the phase and amplitude of sounds detected by headphones placed over the ears.

Adult↗

PubMed2001

[Role of HDL cholesterol in the prediction of exercise stress test abnormalities in asymptomatic high-risk patients].

The diagnosis of coronary artery disease in asymptomatic patients is useful in order to target therapeutic intervention in the patients at highest risk. Systematic testing of all asymptomatic adults with coronary risk factors is not feasible. The aim of this study, carried out in 950 healthy subjects, was to assess the predictive value of classical risk factors for positive exercise stress tests (EE). All subjects underwent stress testing using the Bruce protocol. Statistical analysis was performed by multiple logistic regression on half the samples, then by CART (Classification and Regression Trees) analysis on all subjects. Age, HDL-cholesterol and interaction between lipid lowering treatment and LDL-cholesterol were significantly correlated (p < 0.05) to a positive exercise stress test. In both groups, treated or untreated by lipid lowering drugs. CART identified HDL-cholesterol (< 0.40 g/l) as a predictive factor for positive stress testing. Subgroups of elderly patients (> or = 60 years) with probabilities of 20 to 28% for a positive stress test were identified. The authors conclude that the diagnosis of coronary artery disease by systematic exercise stress testing is potentially valuable in elderly patients with low HDL-cholesterol values.

Adult↗

PubMed2004

Ordered multiple-class ROC analysis with continuous measurements.

Receiver operating characteristic (ROC) curves have been useful in two-group classification problems. In three- and multiple-class diagnostic problems, an ROC surface or hyper-surface can be constructed. The volume under these surfaces can be used for inference using bootstrap techniques or U-statistics theory. In this article, ROC surfaces and hyper-surfaces are defined and their behaviour and utility in multi-group classification problems is investigated. The formulation of the problem is equivalent to what has previously been proposed in the general multi-category classification problem but the definition of ROC surfaces here is less complex and addresses directly the narrower problem of ordered categories in the three-class and, by extension, the multi-class problem applied to continuous and ordinal data. Non-parametric manipulation of both continuous and discrete test data and comparison between two diagnostic tests applied to the same subjects are considered. A three-group classification example in the context of HIV neurological disease is presented and the results are discussed.

AIDS Dementia Complex↗

PubMed1989

Cognitive errors in diagnosis: instantiation, classification, and consequences.

To identify diagnostic errors caused by faulty clinical cognition, we analyzed 40 consecutive transcripts of problem-solving exercises published in a pedagogic series of clinical reasoning. The analysis disclosed multiple errors in cognition and produced a provisional classification of these errors based on a framework derived from cognitive science. Faults in cognition were identified in all steps of the diagnostic process, including triggering, context formulation, information gathering and processing, and verification. We instantiated each type of error by providing detailed specific examples, and identified the consequences of each error. We conclude that cognitive errors can be identified and classified, that they can produce serious morbidity, and that a classification of cognitive errors is a step toward a deeper understanding of the epidemiology, causes, and prevention of diagnostic errors.

Clinical Competence↗

PubMed2005

Algorithm for ventricular capture verification based on the mechanical evoked response.

Automatic pacemaker capture verification is important for maintaining safety and low energy consumption in pacemaker patients. A new algorithm was developed, based on impedance measurement between pacing electrode poles, which reflects the distribution of the conducting medium between the poles and changes with effective contraction. Data acquired during pacemaker implant in 17 subjects were analysed, with intracardiac impedance recorded while pacing was performed in the ventricle at varying energies, resulting in multiple-captured and non-captured beats. The impedance signals of all captured/non-captured beats were analysed using three different algorithms, based on the morphology of the impedance signal. The algorithm decision for each beat was compared with an actual capture or non-capture, as determined from the simultaneous recording of surface ECG. Two of the three algorithms (Z1 and Zn) were based on impedance values, and one (Z'n) was based on the first derivative of the impedance. Z1 was based on a single sample, whereas Z'n and Z'n were based on several samples in each beat. The total accuracy for each was Z1: 43%, Zn: 87%, Z'n: 92%. It was concluded that impedance-based capture verification is feasible, that a multiple rather than single sample approach for signal classification is both feasible and superior, and that first derivative analysis with multiple samples (Z'n) provides the best results.

Aged↗

PubMed1986

Women at risk for developing osteoporosis: determination by total body neutron activation analysis and photon absorptiometry.

With stepwise multiple logistic regression (MLR), probabilistic classification equations were developed to identify asymptomatic women who are at risk for development of fracture of the spine. Clinically normal women with low TBCa/square root H ratios can be classified as at risk for osteoporosis prior to their developing spinal compression fractures. With receiver operating characteristic (ROC) analysis, it was possible to verify the accuracy of the MLR model to discriminate "normal" women at risk, with high sensitivity and specificity. With the MLR model, discrimination of osteoporotic women (50-59 years) was made correctly for 86.2% of the total osteoporotic subjects with the TBCa data. Similar models were derived from the photon absorptiometry data. From the spinal density (BDs) data, correct classification in the 50-59 year group was 55.6% of the total osteoporosis subjects; from the radius density (BMCr) data, the corresponding value was 31%. The highest probability of identifying osteoporosis in all age categories was, therefore, on the basis of TBCa data. Similar, but less accurate discrimination was achieved with the BDs and BMCr data. These conclusions were confirmed by the application of receiver operating characteristic (ROC) analysis. Correct identification of the population at risk permits the timely and efficient application of therapeutic programs prior to onset of fracture. In a serial study of 104 peri-menopausal women, for example, it was possible to determine the P value for individuals measured annually over a 3-10 year period and thus to predict normal individuals at risk for developing osteoporosis each year.

Age Factors↗

PubMedDate not supplied

[Surgical treatment and prognosis factors in spinal metastases of breast cancer].

AIM: The aim of this study was the evaluation of surgical therapy results and prognosis factors in patients with spinal metastases of breast cancer. METHODS: 55 patients with spinal metastases of breast cancer who were treated surgically were retrospectively evaluated. In 11 patients the cervical, in 27 patients the thoracic and in 17 patients the lumbar spine was affected. RESULTS: Postoperatively, 45 patients (81.8 %) described a reduction in pain and 5 patients (50 %) reported a neurological improvement. Perioperative complications appeared in 27 patients (49.1 %), 2 patients died. For the entire group, the mean postoperative survival was 27.2 +/- 28.6 months and the median survival 16.2 months. In patients with solitary metastasis the univariate analysis did not show a significantly longer postoperative survival than in patients with additional visceral metastases (p = 0.0659), but patients with solitary metastasis showed a significantly longer survival than those with multiple osseous and/or visceral metastases (p = 0.0325). In the univariate analysis, the classification of the primary tumour, the duration of symptoms, the localisation of the metastases, the patient's age and the kind of surgical procedure (posterior stabilising instrumentation versus combined posterior-anterior treatment with intralesional resection of the affected vertebra and vertebral body replacement) did not show a significant influence on the postoperative survival. The multivariate analysis did not show a significant prognostic influence for the potentially prognostic factors, however, solitary and multiple metastasis showed the highest statistical influence for the prognosis (p = 0.1187), followed by the classification of the primary tumour (p = 0.1243). CONCLUSION: Pain reduction and neurological improvement can be reached by a stabilisation of the diseased spinal region. Patients with spinal metastases due to breast cancer showed a relatively long postoperative median and mean survival. Therefore, the preoperative evaluation of extent of the disease and the therapy concept should be individually adapted. The surgical procedure (posterior stabilising instrumentation versus combined posterior-anterior approach with vertebrectomy and vertebral body replacement) does not significantly influence the survival.

Adult↗

PubMed2006

On the classification of experimental data modeled via a stochastic leaky integrate and fire model through boundary values.

We present a computational algorithm aimed to classify single unit spike trains on the basis of observed interspikes intervals (ISI). The neuronal activity is modeled with a stochastic leaky integrate and fire model and the inverse first passage time method is extended to the Ornstein-Uhlenbeck (OU) process. Differences between spike trains are detected in terms of the boundary shape. The proposed classification method is applied to the analysis of multiple single units recorded simultaneously in the thalamus and in the cerebral cortex of unanesthetized rats during spontaneous activity. We show the existence of at least three different firing patterns that could not be classified using the usual statistical indices.

Action Potentials↗

PubMed1985

Anthropometric assessment of nutritional status in newborn infants. Discriminative value of mid arm circumference and of skinfold thickness.

74 appropriate-for-gestational age (AGA) and 22 small-for-gestational age (SGA) caucasian infants were studied for anthropometric parameters: mid arm circumference (MAC), triceps and subscapular skinfold thickness (TSKF and SSKF) recorded at 15 and 60 s, chest circumference (cc), head circumference, birth weight and length. MAC is highly correlated with birth weight either in AGA (r = 0.936; P less than 0.001) or in SGA infants (r = 0.860; P less than 0.001). MAC is also correlated with gestational age in AGA (r = 0.850; P less than 0.001) and SGA infants (r = 0.76; P less than 0.001). Similar correlations were found between TSKF, SSKF and birth weight or gestational age. Arm muscle and fat areas are also positively correlated with birth weight and gestational age, in AGA and SGA infants. A multiple regression analysis of our data allowed a classification of the best discriminant anthropometric parameters between AGA and SGA infants. MAC, SSKF15, SSKF60 and chest circumference were selected. An equation was established in AGA infants with these four parameters giving a predictive gestational age: gestational age (weeks) = 1.216 MAC (cm)-3.588 SSKF15 (mm) + 0.263 CC (cm) + 17.9. The ratio of predicted gestational age to the real gestational age was 1.0 +/- 0.044 in AGA versus 0.896 +/- 0.034 in SGA infants. Our data suggest that MAC and SSKF provide a simple measure of body composition of neonates and a useful tool for determining the degree of maturity of a newborn independent of birth weight.

Anthropometry↗

PubMed2006

HPV typing and CGH analysis for the differentiation of primary and metastatic squamous cell carcinomas of the aerodigestive tract.

Human papilloma virus (HPV) typing and Comparative Genomic Hybridisation (CGH) analysis can be used in the classification of multiple tumours of the aerodigestive tract for the differentiation between secondary malignancy versus metastasis. We present 3 exemplary cases of patients with multiple squamous cell carcinomas, localised within the head and neck region, cervical lymph node and the lung. In two patients, HPV typing identified HPV type 16 in the tonsillar carcinomas and the corresponding cervical lymph node and lung carcinoma indicating that the latter were metastatic spreads. In case 1, CGH confirmed the clonal relationship. Case two showed a peculiar syncytial growth pattern with lymphocytic infiltration which may constitute a potential morphological marker for HPV infection. In case three, a vallecular carcinoma was HPV negative while a lung cancer was positive for HPV type 6 indicating two independent primary tumours. Our case triplet illustrates the variability of HPV infection in squamous cell cancer of the aerodigestive tract and power as well as limitations of morphology, HPV typing and tumour genetics in the classification of multiple tumours.

Carcinoma, Squamous Cell↗

PubMed1995

Exercise capacity and prognosis in patients with chronic atrial fibrillation.

To evaluate the response of patients with chronic atrial fibrillation (AF) to exercise and to demonstrate if prognosis could be predicted, 200 male patients (64 +/- 1 years) with AF were identified retrospectively who underwent resting echocardiography and symptom-limited treadmill testing. They were classified by underlying disease into three subgroups: hypertension or no underlying disease (LONE; n = 102), ischemic heart disease (IHD; n = 45) and history of congestive heart failure or valvular disease (CHF-VD; n = 53). Maximal exercise capacities for LONE, IHD and CHF-VD were (mean +/- 1 SEM) 8.0 +/- 0.3, 6.4 +/- 0.4 and 6.0 +/- 0.3 metabolic equivalents, respectively (p < 0.01), and resting left ventricular ejection fractions were 61.7 +/- 1.6, 60.1 +/- 2.2 and 49.5 +/- 1.9%, respectively (p < 0.01). Stepwise multiple regression analysis demonstrated that, except for group classification (R2 = 0.13, p < 0.01), no clinical, exercise or morphologic variables could predict exercise capacity. After a mean 39.1-month follow-up (range 1-78), 17 of the 200 had died from cardiovascular causes. The rate of cardiac death using Kaplan-Meier survival analysis was significantly greater in CHF-VD patients (p < 0.01). However, Cox hazard function and Kaplan-Meier survival analysis demonstrated that neither echocardiographic measurements of cardiac size or function at rest, nor exercise or clinical variables were significant predictors of outcome. AF patients with a history of CHF and/or VD demonstrated a reduced exercise tolerance ad a worse prognosis than those without morphologic heart disease or those with IHD.(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗

PubMed2002

Artificial intelligence techniques for bioinformatics.

This review provides an overview of the ways in which techniques from artificial intelligence (AI) can be usefully employed in bioinformatics, both for modelling biological data and for making new discoveries. The paper covers three techniques: symbolic machine learning approaches (nearest neighbour and identification tree techniques), artificial neural networks and genetic algorithms. Each technique is introduced and supported with examples taken from the bioinformatics literature. These examples include folding prediction, viral protease cleavage prediction, classification, multiple sequence alignment and microarray gene expression analysis.

Algorithms↗

PubMed1982

Determination of sex of white femora by discriminant function analysis: forensic science applications.

Stepwise multiple discriminant function analysis is used to establish classification functions for sex assessment of North American white femora. The functions correctly assign sex for 82% of a sample consisting of 85 femora of verified age, sex, and race, and for a similarly verified test sample of 30. The objectives are to provide criteria for sexing poorly preserved and fragmentary unknown specimens and a statement of the probable accuracy of such assessments in individual cases. The application of the method to forensic casework is illustrated by a sample case.

Anthropometry↗

PubMed1998

Spectral analysis of heart valve sound for detection of prosthetic heart valve diseases.

The spectral analysis of heart valve sound is a noninvasive diagnostic method known to be useful in evaluating the state of the heart valve function. This may provide early detection of valve calcification, thrombus or destruction, since previous studies have shown that the dominant frequency peak moved to a high frequency area when natural heart valve leaflets were calcified, stiffened or destroyed. However, it is important for a heart valve sound diagnostic system to find a proper spectral analysis method on phonocardiography. Until now, conventional frequency analyses such as the Fourier transform or autoregressive spectral estimation technique have been used to estimate spectral components of a phonocardiogram, but they are inappropriate because the signal frequency is assumed to remain constant during the transform interval. To overcome this problem, in this study, FOS (Fast Orthogonal Search) & MUSIC (MUltiple SIgnal Classification), which both appeared suitable for the analysis of biological data, were applied to prosthetic heart valve sound as the new heart valve sound spectral analysis methods. Five subjects with normally functioning mechanical heart valves and a patient with a malfunctioning one were selected to collect the heart valve sound signals. As a result, the second dominant peak frequency proved to be important along with the first dominant peak frequency in identifying the valve function. This study showed that the new heart valve sound spectral analysis method presented in this paper may be an effective method in heart valve sound analysis. Further study using this system in a large population of patients will aid in providing a diagnostic method in the early detection of valve failure.

Female↗

PubMed2003

Fragmental methods in the analysis of biological activities of diverse compound sets.

The current mini-review explains how fragmental methods (FMs) can be used in the analysis and prediction of physicochemical properties and biological activities. The considered properties include log P, solubility, pK(a), intestinal permeability, P-gp substrate specificity and toxicity. The focus will be a description of a "mechanistic" approach, which implies a gradual reduction of alternative explanations for any property or activity. This means a flexible construction of fragmental parameters using large amounts of experimental data. Since biological activities involve multiple (unknown) target macromolecules with multiple binding modes, a stepwise classification (C-SAR) analysis is most useful. It involves the following procedures: (i). construction of physicochemical profiles using parameters that can be reliably predicted, (ii). identification of reactive functional groups and the largest active skeletons, (iii). generalization of these groups and skeletons in terms of "site-specific physicochemical profiling". This entails a dynamic construction of 2D pharmacophores that can be converted into 3D models.

Algorithms↗

PubMed1996

[Structural analysis of the new model of primary care in the community of Valencia].

OBJECTIVES: To analyse the structure of the new model of primary care (NMPC) in the Community of Valencia, and to identify the strategic importance of its characteristic variables and the possibilities of intervention to affect these variables. DESIGN: A qualitative study through a method of structural analysis (crossed impact method-multiplication applied to a classification) of the relationships between 37 variables characterising the NMPC which were identified by prior qualitative research, with interpretation of the results using the Téniere-Buchot Model. SETTING: Community of Valencia. RESULTS: The structural variables identified were those relating to the political-legal framework and to the allocation of primary care resources; and the resultant variables, those relating to efficiency and primary care quality. Between these two categories, the intervention variables covered management, NMPC professionals, health needs and the community's use of services. CONCLUSIONS: The structural analysis gives the legal-political and economical framework a determining role in NMPC, which can hardly be influenced from within the system. Management and organisation are identified as key variables from which an intervention can be made in the short or medium term to achieve the aims of the system.

Models, Organizational↗

PubMed1997

In vivo detection of carotid plaque thrombus by ultrasonic tissue characterization.

The purpose of the study was to determine whether ultrasonic tissue characterization could detect carotid plaque thrombus in vivo. Patients undergoing carotid endarterectomy were examined preoperatively and the ultrasonic tissue characterization findings were compared to those of optical microscopy of the removed plaque specimens. Ten of 15 patients studied had plaque thrombus. Ultra-ultrasonic tissue characterization entailed an analysis of parameters obtained from the power spectrum of backscattered ultrasound signals. Data were obtained with a nominal 10 MHz sector scanning transducer with an effective bandwidth of 3 to 13 MHz. The parameters were the slope and intercept derived from the linear regression of the normalized spectrum and total power (log of the integrated power of the normalized spectrum over the effective bandwidth). The combined effect of the three parameters was determined by discriminant function analysis and showed a significant difference (P < 0.05) between nonthrombus and plaque thrombus in a small sample of patients with advance carotid atherosclerosis. These parameters applied singly could not provide such a distinction. Correct classification of carotid plaque thrombus using the multiple-parameter analysis revealed a sensitivity of 90%, specificity of 80%, and accuracy of 86.7%. This study demonstrates that analysis utilizing a combination of multiple spectral parameters was able to detect carotid plaque thrombus in vivo.

Aged↗