Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Classification Algorithms”

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

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

At least 145 records · Page 8Linked to original sources
PubMed2006

Decision threshold adjustment in class prediction.

Standard classification algorithms are generally designed to maximize the number of correct predictions (concordance). The criterion of maximizing the concordance may not be appropriate in certain applications. In practice, some applications may emphasize high sensitivity (e.g., clinical diagnostic tests) and others may emphasize high specificity (e.g., epidemiology screening studies). This paper considers effects of the decision threshold on sensitivity, specificity, and concordance for four classification methods: logistic regression, classification tree, Fisher's linear discriminant analysis, and a weighted k-nearest neighbor. We investigated the use of decision threshold adjustment to improve performance of either sensitivity or specificity of a classifier under specific conditions. We conducted a Monte Carlo simulation showing that as the decision threshold increases, the sensitivity decreases and the specificity increases; but, the concordance values in an interval around the maximum concordance are similar. For specified sensitivity and specificity levels, an optimal decision threshold might be determined in an interval around the maximum concordance that meets the specified requirement. Three example data sets were analyzed for illustrations.

Algorithms↗
PubMed2004

Automated ischemic beat classification using genetic algorithms and multicriteria decision analysis.

Cardiac beat classification is a key process in the detection of myocardial ischemic episodes in the electrocardiographic signal. In the present study, we propose a multicriteria sorting method for classifying the cardiac beats as ischemic or not. Through a supervised learning procedure, each beat is compared to preclassified category prototypes under five criteria. These criteria refer to ST segment changes, T wave alterations, and the patient's age. The difficulty in applying the above criteria is the determination of the required method parameters, namely the thresholds and weight values. To overcome this problem, we employed a genetic algorithm, which, after proper training, automatically calculates the optimum values for the above parameters. A task-specific cardiac beat database was developed for training and testing the proposed method using data from the European Society of Cardiology ST-T database. Various experimental tests were carried out in order to adjust each module of the classification system. The obtained performance was 91% in terms of both sensitivity and specificity and compares favorably to other beat classification approaches proposed in the literature.

Age Factors↗
PubMed1997

[Development and evaluation of a spectroscopy system for classification of laser-induced arterial fluorescence spectra].

The present study evaluated the potential of fluorescence guidance of laser angioplasty without using a second laser for fluorescence excitation. A prototype spectroscopy system with a grating spectrograph, microchannel plate, CCD array and digital image processor on a personal computer was developed and coupled to a clinical XeCl excimer laser. Using multifibre catheters, specimens of human aorta were ablated in physiological saline and blood. The spectra thus generated were recorded and validated histologically. Five types of spectra could be differentiated. Based on a training set, classification algorithms were developed using multiple linear regression and linear discriminant analysis with intensity ratios as predictor variables. Discriminant analysis yielded prospective classification of the remaining validation spectra with high sensitivity and specificity for each type. These data demonstrate that fluorescence spectroscopy during excimer laser ablation at 308 nm does not require a diagnostic laser. Principal types of atherosclerotic lesions and the media can be differentiated spectroscopically in physiological saline and blood.

Adult↗
PubMed2004

Methods for multi-category cancer diagnosis from gene expression data: a comprehensive evaluation to inform decision support system development.

Cancer diagnosis is a major clinical applications area of gene expression microarray technology. We are seeking to develop a system for cancer diagnostic model creation based on microarray data. In order to equip the system with the optimal combination of data modeling methods, we performed a comprehensive evaluation of several major classification algorithms, gene selection methods, and cross-validation designs using 11 datasets spanning 74 diagnostic categories (41 cancer types and 12 normal tissue types). The Multi-Category Support Vector Machine techniques by Crammer and Singer, Weston and Watkins, and one-versus-rest were found to be the best methods and they outperform other learning algorithms such as K-Nearest Neighbors and Neural Networks often to a remarkable degree. Gene selection techniques are shown to significantly improve classification performance. These results guided the development of a software system that fully automates cancer diagnostic model construction with quality on par with or better than previously published results derived by expert human analysts.

Algorithms↗
PubMedDate not supplied

Olecranon fractures in children: Part 1: a clinical review; Part 2: a new classification and management algorithm.

Part 1: A Clinical Review: Fractures of the olecranon in children are uncommon, and the outcome has historically been reported as good. Management varies from simple immobilization to include various methods of internal fixation. Recently a new technique of fracture fixation has been used in an attempt to improve the management of this fracture. This article reviews olecranon fractures seen at two children's hospitals over a 10-year period. The purpose of the study was to review and classify various fracture patterns, different fixation techniques, and to assess medium-term outcomes after this uncommon injury. Part 2: A New Classification and Management Algorithm: A new classification system and management algorithm for fractures of the olecranon in children is presented. This was developed after a 10-year review of olecranon fractures and their management at two children's hospitals (Part 1). Minimally displaced or undisplaced fractures account for 80% of all olecranon fractures, and these are managed with immobilization, rarely requiring hospital admission. Of those fractures admitted to hospital, oblique metaphyseal fractures are most commonly seen and are satisfactorily treated with tension-band techniques, but a wide variety of fracture patterns are less common and may require different management. Previous classifications of this uncommon fracture have not addressed all of the significant features to categorize the fracture accurately or to imply a management pathway. This new anatomically based classification system facilitates accurate classification for research purposes, and its principles are applied in the clinical-management algorithm presented here.

Adolescent↗
PubMed2006

Comparison of retinal nerve fiber layer thickness and optic disk algorithms with optical coherence tomography to detect glaucoma.

PURPOSE: To compare the performance of the retinal nerve fiber layer (RNFL) thickness and optic disk algorithms as determined by optical coherence tomography to detect glaucoma. DESIGN: Observational cross-sectional study. METHODS: setting: Academic tertiary-care center. study population: One eye from 42 control subjects and 65 patients with open-angle glaucoma with visual acuity of > or =20/40, and no other ocular pathologic condition. observation procedures: Two optical coherence tomography algorithms were used: "fast RNFL thickness" and "fast optic disk." main outcome measures: Area under the receiver operating characteristic curves and sensitivities at fixed specificities were used. Discriminating ability of the average RNFL thickness and RNFL thickness in clock-hour sectors and quadrants was compared with the parameters that were derived from the fast optic disk algorithm. Classification and regression trees were used to determine the best combination of parameters for the detection of glaucoma. RESULTS: The average visual field mean deviation (+/-SD) was 0.0 +/- 1.3 and -5.3 +/- 5.0 dB in the control and glaucoma groups, respectively. The RNFL thickness at the 7 o'clock sector, inferior quadrant, and the vertical C/D ratio had the highest area under the receiver operating characteristic curves (0.93 +/- 0.02, 0.92 +/- 0.03, and 0.90 +/- 0.03, respectively). At 90% specificity, the best sensitivities (+/-SE) from each algorithm were 86% +/- 3% for RNFL thickness at the 7 o'clock sector and 79% +/- 4% for horizontal integrated rim width (estimated rim area). The combination of inferior quadrant RNFL thickness and vertical C/D ratio achieved the best classification (misclassification rate, 6.2%). CONCLUSION: The fast optic disk algorithm performs as well as the fast RNFL thickness algorithm for discrimination of glaucoma from normal eyes. A combination of the two algorithms may provide enhanced diagnostic performance.

Adult↗
PubMed2004

Sliding window discretization: a new method for multiple band matching of bacterial genotyping fingerprints.

Microbiologists have traditionally applied hierarchical clustering algorithms as their mathematical tool of choice to unravel the taxonomic relationships between micro-organisms. However, the interpretation of such hierarchical classifications suffers from being subjective, in that a variety of ad hoc choices must be made during their construction. On the other hand, the application of more profound and objective mathematical methods--such as the minimization of stochastic complexity--for the classification of bacterial genotyping fingerprints data is hampered by the prerequisite that such methods only act upon vectorized data. In this paper we introduce a new method, coined sliding window discretization, for the transformation of genotypic fingerprint patterns into binary vector format. In the context of an extensive amplified fragment length polymorphism (AFLP) data set of 507 strains from the Vibrionaceae family that has previously been analysed, we demonstrate by comparison with a number of other discretization methods that this new discretization method results in minimal loss of the original information content captured in the banding patterns. Finally, we investigate the implications of the different discretization methods on the classification of bacterial genotyping fingerprints by minimization of stochastic complexity, as it is implemented in the BinClass software package for probabilistic clustering of binary vectors. The new taxonomic insights learned from the resulting classification of the AFLP patterns will prove the value of combining sliding window discretization with minimization of stochastic complexity, as an alternative classification algorithm for bacterial genotyping fingerprints.

Bacteria↗
PubMed1998

A cluster computer system for the analysis and classification of massively large biomedical image data.

The current trend in medical image acquisition is towards the generation of image datasets which are massively large, either because they exhibit fine x, y, or z resolution, are volumetric, are multispectral, or a combination of all of the preceding. Such images pose a significant computational challenge in their analysis, not only in terms of data throughput, but also in terms of platform costs and simplicity. In this paper we describe the role of a cluster of workstations together with two quite different application programming interfaces (APIs) in the quantitative analysis of anatomic image data from the visible human project using an MRF-Gibbs classification algorithm. We describe the typical architecture of a cluster computer, two API options and the parallelization of the MRF-Gibbs procedure for the cluster. Finally, we show speedup results obtained on the cluster and sample classifications of visible human data.

Algorithms↗
PubMed2005

Skin segmentation using color pixel classification: analysis and comparison.

This paper presents a study of three important issues of the color pixel classification approach to skin segmentation: color representation, color quantization, and classification algorithm. Our analysis of several representative color spaces using the Bayesian classifier with the histogram technique shows that skin segmentation based on color pixel classification is largely unaffected by the choice of the color space. However, segmentation performance degrades when only chrominance channels are used in classification. Furthermore, we find that color quantization can be as low as 64 bins per channel, although higher histogram sizes give better segmentation performance. The Bayesian classifier with the histogram technique and the multilayer perceptron classifier are found to perform better compared to other tested classifiers, including three piecewise linear classifiers, three unimodal Gaussian classifiers, and a Gaussian mixture classifier.

Algorithms↗
PubMed2006

[Preliminary study of serum proteome biomarkers of hypopharyngeal squamous cell carcinoma].

OBJECTIVE: To screen the serum proteome biomarkers of hypopharyngeal squamous cell carcinoma (HSCC) and to establish a predictive model for early detection of HSCC. METHODS: Serum samples were collected from 48 HSCC patients before surgery and 52 age and sex-matched individuals without cancer used as controls. The samples were divided into 2 sets: training set (including 36 HSCC patients and 36 controls) and blind testing set (including 12 HSCC patients and 16 controls). With WCX2 and IMAC3 protein chips, surface-enhanced laser desorption/ionization (SELDI) was used to analyze the serum protein profiling. 72 samples of the training set were analyzed by a decision tree algorithm to be able to differentiate HSCC patients from controls. Double-blind test was used to determine the sensitivity and specificity of the classification model. RESULTS: Ranging from 2000 - 50000 (M/Z), 11 potential biomarkers on WCX2 and 19 biomarkers on IMAC3 protein chip could differentiate HSCC patients from the control set (P < 10(-5)). Among them 4 candidate protein peaks with the m/z values of 7796, 4216, 5927, and 5361 were selected to be used to establish a predictive model by Biomarker Pattern Software. The model separated effectively the HSCC samples from the control samples, achieving a sensitivity of 94.44%, and a specificity of 88.89%. An accuracy of 85.71% (24/28), sensitivity of 91.67% (11/12), specificity of 81.25% (13/16), positive predictive value of 78.57%% (11/14), and negative predictive value of 92.85% (13/14) were validated in the double-blind testing set. CONCLUSION: The SELDI-TOF-MS Protein Chip combined with artificial intelligence classification algorithm helps find serum proteome biomarkers and establish predictive model for early diagnosis of HSCC. This technique has potential for the development of a screening test for the detection of HSCC.

Adult↗
PubMed2006

Resolution and SNR effects on carotid plaque classification.

Multicontrast-weighted MRI, which is increasingly being used in combination with automatic classification algorithms, has the potential to become a powerful tool for assessing plaque composition. The current literature, however, does not address the relationship between imaging conditions and segmentation viability well. In this study 13 carotid endarterectomy samples were imaged with a 156-microm in-plane resolution and high signal-to-noise ratio (SNR) using proton density (PD), T1, T2, and diffusion weightings. The maximum likelihood (ML) algorithm was used to classify plaque components, with sets of three contrast weighting intensities used as features. The resolution and SNR of the images were then degraded. Classification accuracy was found to be independent of in-plane resolution between 156 microm and 1250 microm, but dependent on SNR. Accuracy decreased less than 10% for degradation in SNR down to 25% of original values, and decreased sharply thereafter. The robustness of automatic classifiers makes them applicable to a wide range of imaging conditions, including standard in vivo carotid imaging scenarios.

Algorithms↗
PubMed2001

Employment of intra-individual variability to improve computerized ECG interpretation.

One of the reasons for the limited practical utility of computer programs for interpretation of electrocardiograms (ECGs) is their susceptibility to intra-individual variability. Two of the most prominent sources of intra-individual variability in ECGs, electrode placement variations and respiration, were studied for their effects on computerized ECG interpretation. Previous research has shown that the effects of intra-individual variability on computerized ECG interpretation depend largely on the individual ECG. To enable the assessment of chest electrode position variations for individual standard 12- lead ECGs, ECGs resulting from simulations of such position variations were interpreted. Variability due to respiration was assessed by interpreting all individual ECG beats instead of an averaged beat. In this paper two methods are presented that employ information about the intra-individual variability in individual ECGs. The first method provides an estimate of the reliability of the interpretation, the second attempts to improve the interpretation itself. In the first method we quantified the variation in interpretation caused by the two sources of intra-individual variability with the use of a stability index, a high index value indicating a low variation in interpretation. This index was subsequently studied using two sets of ECGs. For the first set a â clinical' reference interpretation was obtained from discharge letters. For the second set three cardiologists provided a â cardiologists' reference. The performance of subgroups of ECGs having stability indices higher than a particular value was computed. It appeared that for the â cardiologists' reference, the interpretations of ECGs with a high stability index were more often correct. No effect was found for the â clinical' reference. In the second method we attempted to improve the original interpretation by combining the alternative interpretations into a new interpretation. This was done by taking the median or the average of the quantified alternatives. These combined interpretations proved to perform better than the original interpretation when a cardiologist's interpretation was taken as a reference. This paper shows that intra-individual ECG variability can be used to improve original interpretations. This can be done without having to record multiple ECGs, provided that a model is available to simulate intra-individual variability. The presented methods do not depend on the classification algorithm that is used. They can be used both during classifier design to correct imperfections, and in routine use of the classifier to produce more representative classifications.

Decision Trees↗
PubMed2006

Enhanced real-time cursor control algorithm, based on the spectral analysis of electromyograms.

This paper presents a new version of an EMG-based, hands-free, cursor control system, and compares its performance to that of a previous version. Both systems use classification algorithms that rely on the periodogram estimation of the power spectral density (PSD) of electromyogram (EMG) signals from muscles in the face. The older system requires three electrodes for EMG input, and utilizes an algorithm that calculates partial power accumulations over the frequency ranges of 0Hz - 145Hz and 145Hz - 600Hz in the PSDs of the EMG signals. The new system requires four electrodes for EMG input, and utilizes an algorithm that calculates mean power frequency (MPF) values to assist in distinguishing the cranial muscle that contracted. An experiment was devised to gauge the point-and-click capabilities of both systems. The experimental results were evaluated using Fitts' Law analysis. The results show that the new algorithm provides improved point-and-click performance over the old algorithm.

Algorithms↗
PubMed2003

Cluster analysis of fMRI data using dendrogram sharpening.

The major disadvantage of hierarchical clustering in fMRI data analysis is that an appropriate clustering threshold needs to be specified. Upon grouping data into a hierarchical tree, clusters are identified either by specifying their number or by choosing an appropriate inconsistency coefficient. Since the number of clusters present in the data is not known beforehand, even a slight variation of the inconsistency coefficient can significantly affect the results. To address these limitations, the dendrogram sharpening method, combined with a hierarchical clustering algorithm, is used in this work to identify modality regions, which are, in essence, areas of activation in the human brain during an fMRI experiment. The objective of the algorithm is to remove data from the low-density regions in order to obtain a clearer representation of the data structure. Once cluster cores are identified, the classification algorithm is run on voxels, set aside during sharpening, attempting to reassign them to the detected groups. When applied to a paced motor paradigm, task-related activations in the motor cortex are detected. In order to evaluate the performance of the algorithm, the obtained clusters are compared to standard activation maps where the expected hemodynamic response function is specified as a regressor. The obtained patterns of both methods have a high concordance (correlation coefficient = 0.91). Furthermore, the dependence of the clustering results on the sharpening parameters is investigated and recommendations on the appropriate choice of these variables are offered. Hum. Brain Mapping 20:201-219, 2003.

Algorithms↗
PubMed2024

SIMS: A deep-learning label transfer tool for single-cell RNA sequencing analysis.

Cell atlases serve as vital references for automating cell labeling in new samples, yet existing classification algorithms struggle with accuracy. Here we introduce SIMS (scalable, interpretable machine learning for single cell), a low-code data-efficient pipeline for single-cell RNA classification. We benchmark SIMS against datasets from different tissues and species. We demonstrate SIMS's efficacy in classifying cells in the brain, achieving high accuracy even with small training sets (<3,500 cells) and across different samples. SIMS accurately predicts neuronal subtypes in the developing brain, shedding light on genetic changes during neuronal differentiation and postmitotic fate refinement. Finally, we apply SIMS to single-cell RNA datasets of cortical organoids to predict cell identities and uncover genetic variations between cell lines. SIMS identifies cell-line differences and misannotated cell lineages in human cortical organoids derived from different pluripotent stem cell lines. Altogether, we show that SIMS is a versatile and robust tool for cell-type classification from single-cell datasets.

Single-Cell Analysis↗
PubMed2001

ECG beat recognition using fuzzy hybrid neural network.

This paper presents the application of the fuzzy neural network for electrocardiographic (ECG) beat recognition and classification. The new classification algorithm of the ECG beats, applying the fuzzy hybrid neural network and the features drawn from the higher order statistics has been proposed in the paper. The cumulants of the second, third, and fourth orders have been used for the feature selection. The hybrid fuzzy neural network applied in the solution consists of the fuzzy self-organizing subnetwork connected in cascade with the multilayer perceptron, working as the final classifier. The c-means and Gustafson-Kessel algorithms for the self-organization of the neural network have been applied. The results of experiments of recognition of different types of beats on the basis of the ECG waveforms have confirmed good efficiency of the proposed solution. The investigations show that the method may find practical application in the recognition and classification of different type heart beats.

Algorithms↗
PubMed2001

Fingerprints classification using artificial neural networks: a combined structural and statistical approach.

This paper describes a fingerprint classification algorithm using Artificial Neural Networks (ANN). Fingerprints are classified into six categories: arches, tented arches, left loops, right loops, whorls and twin loops. The algorithm extracts a string of symbols using the block directional image of a fingerprint, which represents the set of structural features for this image. The moment representing the statistical feature of the pattern is computed for this string and its Euclidean Distance Measures (EDM) are computed by using this moment. Our discrimination system uses a multilayer artificial neural network composed of six subnetworks one for each class. The classifier was tested on 1,500 images of good quality in the Egyptian Fingerprints database; images with poor quality were rejected. In the six-class problem the network achieved 95% classification accuracy. In the five-class problem when we place whorls and twin loops together in the same category the classification accuracy was around 99%. In the four-class problem when we place arches and tented arches in the same class the classification accuracy was 99%.

Algorithms↗
PubMed2023

Unraveling Neuronal Identities Using SIMS: A Deep Learning Label Transfer Tool for Single-Cell RNA Sequencing Analysis.

Large single-cell RNA datasets have contributed to unprecedented biological insight. Often, these take the form of cell atlases and serve as a reference for automating cell labeling of newly sequenced samples. Yet, classification algorithms have lacked the capacity to accurately annotate cells, particularly in complex datasets. Here we present SIMS (Scalable, Interpretable Machine Learning for Single-Cell), an end-to-end data-efficient machine learning pipeline for discrete classification of single-cell data that can be applied to new datasets with minimal coding. We benchmarked SIMS against common single-cell label transfer tools and demonstrated that it performs as well or better than state of the art algorithms. We then use SIMS to classify cells in one of the most complex tissues: the brain. We show that SIMS classifies cells of the adult cerebral cortex and hippocampus at a remarkably high accuracy. This accuracy is maintained in trans-sample label transfers of the adult human cerebral cortex. We then apply SIMS to classify cells in the developing brain and demonstrate a high level of accuracy at predicting neuronal subtypes, even in periods of fate refinement, shedding light on genetic changes affecting specific cell types across development. Finally, we apply SIMS to single cell datasets of cortical organoids to predict cell identities and unveil genetic variations between cell lines. SIMS identifies cell-line differences and misannotated cell lineages in human cortical organoids derived from different pluripotent stem cell lines. When cell types are obscured by stress signals, label transfer from primary tissue improves the accuracy of cortical organoid annotations, serving as a reliable ground truth. Altogether, we show that SIMS is a versatile and robust tool for cell-type classification from single-cell datasets.

Brain organoids↗