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 451 records · Page 25Linked to original sources
PubMed1991

Estimating allele frequencies of hypervariable DNA systems.

Several polymorphisms of human DNA have been shown to be hypervariable due to the recurrence of a variable number of tandem repeats (VNTRs) in the lengths of allelic restriction fragments. The recurrence of allelic variants in this novel class of polymorphisms seems to comply well with a model of continuous random variables. Based on this assumption, we have compiled some simple algorithms for classification of continuous data and estimation of classes of relative frequencies and have implemented these routines for the management of databases storing hypervariable single locus DNA genetic systems. The algorithms are compiled in BASIC language and can be incorporated in task-oriented computer programs. Three procedures are discussed, based in turn on: (a) using predetermined, arbitrary classes; (b) point estimations of frequencies for single fragments using error measurements associated with the kilobase value assignment; (c) estimates of phenotype frequencies according to error measurements. Error measurements are obtained from a statistic of values pertaining to several restriction fragments (genomic controls) repeatedly tested in different experiments. Problems related to these approaches are discussed.

Algorithms↗
PubMed2004

Autofluorescence and diffuse reflectance properties of malignant and benign breast tissues.

BACKGROUND: Fluorescence spectroscopy is an evolving technology that can rapidly differentiate between benign and malignant tissues. These differences are thought to be due to endogenous fluorophores, including nicotinamide adenine dinucleotide, flavin adenine dinucleotide, and tryptophan, and absorbers such as beta-carotene and hemoglobin. We hypothesized that a statistically significant difference would be demonstrated between benign and malignant breast tissues on the basis of their unique fluorescence and reflectance properties. METHODS: Optical measurements were performed on 56 samples of tumor or benign breast tissue. Autofluorescence spectra were measured at excitation wavelengths ranging from 300 to 460 nm, and diffuse reflectance was measured between 300 and 600 nm. Principal component analysis to dimensionally reduce the spectral data and a Wilcoxon ranked sum test were used to determine which wavelengths showed statistically significant differences. A support vector machine algorithm compared classification results with the histological diagnosis (gold standard). RESULTS: Several excitation wavelengths and diffuse reflectance spectra showed significant differences between tumor and benign tissues. By using the support vector machine algorithm to incorporate relevant spectral differences, a sensitivity of 70.0% and specificity of 91.7% were achieved. CONCLUSIONS: A statistically significant difference was demonstrated in the diffuse reflectance and fluorescence emission spectra of benign and malignant breast tissue. These differences could be exploited in the development of adjuncts to diagnostic and surgical procedures.

Algorithms↗
PubMed2005

Molecular voting for glioma classification reflecting heterogeneity in the continuum of cancer progression.

Gliomas, the most common brain tumors, are generally categorized into two lineages (astrocytic and oligodendrocytic) and further classified as low-grade (astrocytoma and oligodendroglioma), mid-grade (anaplastic astrocytoma and anaplastic oligodendroglioma), and high-grade (glioblastoma multiforme) based on morphological features. A strict classification scheme has limitations because a specific glioma can be at any stage of the continuum of cancer progression and may contain mixed features. Thus, a more comprehensive classification based on molecular signatures may reflect the biological nature of specific tumors more accurately. In this study, we used microarray technology to profile the gene expression of 49 human brain tumors and applied the k-nearest neighbor algorithm for classification. We first trained the classification gene set with 19 of the most typical glioma cases and selected a set of genes that provide the lowest cross-validation classification error with k=5. We then applied this gene set to the 30 remaining cases, including several that do not belong to gliomas such as atypical meningioma. The results showed that not only does the algorithm correctly classify most of the gliomas, but the detailed voting results also provide more subtle information regarding the molecular similarities to neighboring classes. For atypical meningioma, the voting was equally split among the four classes, indicating a difficulty in placement of meningioma into the four classes of gliomas. Thus, the actual voting results, which are typically used only to decide the winning class label in k-nearest neighbor algorithms, provide a useful method for gaining deeper insight into the stage of a tumor in the continuum of cancer development.

Algorithms↗
PubMed2004

Fractures of the proximal humerus in osteoporotic bone.

The purpose of this article was to review critically the current treatment options for fractures of the proximal humerus in patients with severe osteoporosis. The main difficulties lie in correctly diagnosing the fracture and hence selecting the most appropriate method of treatment. The reliability of the diagnosis can be increased by systematically appending additional information to a basic fracture classification. Classification is best carried out on a morphological basis, whereby a descriptor of bone quality can be added in order to introduce the degree of osteoporosis into the decision-making algorithm. Any classification system that claims to provide both treatment and prognosis is inappropriate, because prognosis will depend hopefully on the treatment. Approaches to treatment differ widely amongst centers and surgeons. It is still unclear as to what would be the optimal treatment. Factors such as the individual's functional requirements and ability to cooperate should be given careful consideration. At our institution, hemiarthroplasty is the method of choice for ischemic humeral heads and/or when anatomic reconstruction cannot be obtained. In all other displaced fractures, the main objective is preservation of the head since the best functional results can generally be obtained with internal fixation. Selection of a balanced osteosynthesis, adapted to the weak bone, is mandatory. Bulky, stiff implants are inadequate and may cause additional damage. Load sharing, not load bearing, compound constructions are the aim. Obtaining metaphyseal elastic buttressing is the key element in achieving the necessary load-sharing fixation. The system should allow controlled impaction and be forgiving towards occasional load peaks that will occur and are beyond patient control. Thin and flexible implants are required to realize this type of fixation. Given the polypragmatic approach that is current in clinical practice there is room for further improvement of techniques and implants.

Arthroplasty↗
PubMed2026

A unified benchmark of supervised and retrieval-based methods for viral genomic sequence classification.

The rapid growth of genomic sequencing demands fast, accurate, and scalable analysis methods. In viral genomic classification, expanding labeled reference collections can make supervised models costly to update and dependent on fixed label sets, motivating retrieval-based genomic classification as a simpler, more flexible alternative. We present a unified benchmark of supervised and retrieval-based methods for viral genomic sequence classification across three viral classification tasks: hepatitis C virus (HCV) genotyping, COVID-19 discrimination, and human papillomavirus (HPV) genotyping. We compare standard sequence encodings (one-hot, k-mers, FCGR) with dense embeddings (dna2vec, DNABERT). For each representation, we evaluate supervised classifiers (Random Forest, Decision Tree, XGBoost) and retrieval-based classification, where sequence vectors are indexed with FAISS and labels are assigned via similarity-weighted k-NN. Furthermore, we benchmark multiple FAISS index types (Flat, IVF, HNSW, IVFPQ, OPQ) to characterize accuracy-speed-memory trade-offs at scale. The results show that XGBoost and retrieval using Flat or IVF indexes achieve strong classification performance under different computational profiles. Compressed indexes such as IVFPQ and OPQ substantially reduce memory usage, although their accuracy loss depends on the dataset and representation. Overall, supervised XGBoost provides a favorable accuracy-size trade-off, while retrieval-based classification remains competitive and allows labeled reference sequences to be incorporated without retraining a global classifier. This benchmark provides practical guidance for selecting sequence representations, classifiers, and vector-search indexes under different accuracy, memory, and update requirements.

Genome, Viral↗
PubMed2000

The DYT1 phenotype and guidelines for diagnostic testing.

OBJECTIVE: To develop diagnostic testing guidelines for the DYT1 GAG deletion in the Ashkenazi Jewish (AJ) and non-Jewish (NJ) primary torsion dystonia (PTD) populations and to determine the range of dystonic features in affected DYT1 deletion carriers. METHODS: The authors screened 267 individuals with PTD; 170 were clinically ascertained for diagnosis and treatment, 87 were affected family members ascertained for genetic studies, and 10 were clinically and genetically ascertained and included in both groups. We used published primers and PCR amplification across the critical DYT1 region to determine GAG deletion status. Features of dystonia in clinically ascertained (affected) DYT1 GAG deletion carriers and noncarriers were compared to determine a classification scheme that optimized prediction of carriers. The authors assessed the range of clinical features in the genetically ascertained (affected) DYT1 deletion carriers and tested for differences between AJ and NJ patients. RESULTS: The optimal algorithm for classification of clinically ascertained carriers was disease onset before age 24 years in a limb (misclassification, 16.5%; sensitivity, 95%; specificity, 80%). Although application of this classification scheme provided good separation in the AJ group (sensitivity, 96%; specificity, 88%), as well as in the group overall, it was less specific in discriminating NJ carriers from noncarriers (sensitivity, 94%; specificity, 69%). Using age 26 years as the cut-off and any site at onset gave a sensitivity of 100%, but specificity decreased to 54% (63% in AJ and 43% in NJ). Among genetically ascertained carriers, onset up to age 44 years occurred, although the great majority displayed early limb onset. There were no significant differences between AJ and NJ genetically ascertained carriers, except that a higher proportion of NJ carriers had onset in a leg, rather than an arm, and widespread disease. CONCLUSIONS: Diagnostic DYT1 testing in conjunction with genetic counseling is recommended for patients with PTD with onset before age 26 years, as this single criterion detected 100% of clinically ascertained carriers, with specificities of 43% to 63%. Testing patients with onset after age 26 years also may be warranted in those having an affected relative with early onset, as the only carriers we observed with onset at age 26 or later were genetically ascertained relatives of individuals whose symptoms started before age 26 years.

Adolescent↗
PubMed1993

Classification of audiograms by sequential testing: reliability and validity of an automated behavioral hearing screening algorithm.

In 1990, CAST (classification of audiograms by sequential testing) was proposed and developed as an automated, innovative approach to screening infant hearing using a modified Bayesian method. The method generated a four-frequency audiogram in a minimal number of test trials using VRA (visual reinforcement audiometry) techniques. Computer simulations were used to explore the properties (efficiency and accuracy) of the paradigm. The current work is designed to further test the utility of the paradigm with human infants and young children. Accordingly, infants and children between 6 months and 2 years of age were screened for hearing loss. The algorithm's efficacy was studied with respect to validity and reliability. Validity was evaluated by comparing CAST results with tympanometric data and outcomes of staircase-based testing. Test-retest reliability was also assessed. Results indicate that CAST is a valid, efficient, reliable, and potentially cost-effective screening method.

Acoustic Impedance Tests↗
PubMed2004

Classifying G-protein coupled receptors with bagging classification tree.

G-protein coupled receptors (GPCRs) play a key role in different biological processes, such as regulation of growth, death and metabolism of cells. They are major therapeutic targets of numerous prescribed drugs. However, the ligand specificity of many receptors is unknown and there is little structural information available. Bioinformatics may offer one approach to bridge the gap between sequence data and functional knowledge of a receptor. In this paper, we use a bagging classification tree algorithm to predict the type of the receptor based on its amino acid composition. The prediction is performed for GPCR at the sub-family and sub-sub-family level. In a cross-validation test, we achieved an overall predictive accuracy of 91.1% for GPCR sub-family classification, and 82.4% for sub-sub-family classification. These results demonstrate the applicability of this relative simple method and its potential for improving prediction accuracy.

Algorithms↗
PubMed2004

Classification of human breast cancer using gene expression profiling as a component of the survival predictor algorithm.

PURPOSE: Selection of treatment options with the highest likelihood of successful outcome for individual breast cancer patients is based to a large degree on accurate classification into subgroups with poor and good prognosis reflecting a different probability of disease recurrence and survival after therapy. Here we propose a breast cancer classification algorithm taking into account three main prognostic features determined at the time of diagnosis: estrogen receptor (ER) status; lymph node (LN) status; and gene expression signatures associated with distinct therapy outcome. EXPERIMENTAL DESIGN: Using microarray expression profiling and quantitative reverse transcription-PCR analyses, we compared expression profiles of the 70-gene breast cancer survival signature in established breast cancer cell lines and primary breast carcinomas from cancer patients. We classified 295 breast cancer patients using 14-, 13-, 6-, and 4-gene survival predictor signatures into subgroups having statistically distinct probability of therapy failure (P < 0.0001). We evaluated the prognostic power of breast cancer survival predictor signatures alone and in combination with ER and LN status using Kaplan-Meier analysis. RESULTS: The breast cancer survival predictor algorithm allowed highly accurate classification into subgroups with dramatically distinct 5- and 10-year survival after therapy of a large cohort of 295 breast cancer patients with either ER+ or ER- tumors as well as LN+ or LN- disease (P < 0.0001, log-rank test). CONCLUSIONS: Our data imply that quantitative laboratory tests measuring expression profiles of a limited set of identified small gene clusters may be useful in stratification of breast cancer patients at the time of diagnosis into subgroups with statistically distinct probability of positive outcome after therapy and assisting in selection of optimal treatment strategies. The estimated increase in survival due to the optimization of treatment protocols may reach many thousands of breast cancer survivors every year at the 10-year follow-up check point.

Algorithms↗
PubMedDate not supplied

Adult flatfoot/posterior tibial tendon dysfunction: outcomes analysis of surgical treatment utilizing an algorithmic approach.

A new classification scheme is presented along with an algorithmic surgical approach to the treatment of posterior tibial tendon dysfunction (PTTD). This classification scheme treats as separate entities the soft-tissue and osseous components of PTTD. The soft-tissue pathology is staged from tendinitis to tendinosis to that of complete rupture. The osseous element is graded as no planar deformity, reducible planar deformity, or rigid planar deformity. The stage of soft-tissue pathology is then combined with the grade of the osseous condition, leading to a comprehensive surgical treatment plan. Based on this classification, 13 patients were retrospectively evaluated. Only those patients falling into a specific classification (2A and 2B) were included in this analysis. The American Orthopedic Foot and Ankle Society clinical hindfoot-ankle scale was utilized. The mean preoperative clinical rating was 32.8. The mean postoperative clinical rating was 88.0. Pre- and postoperative radiographic criteria were also analyzed. Overall patient satisfaction utilizing this treatment algorithm was good to excellent.

Adolescent↗
PubMed2001

Contextual clustering for analysis of functional MRI data.

We present a contextual clustering procedure for statistical parametric maps (SPM) calculated from time varying three-dimensional images. The algorithm can be used for the detection of neural activations from functional magnetic resonance images (fMRI). An important characteristic of SPM is that the intensity distribution of background (nonactive area) is known whereas the distributions of activation areas are not. The developed contextual clustering algorithm divides an SPM into background and activation areas so that the probability of detecting false activations by chance is controlled, i.e., hypothesis testing is performed. Unlike the much used voxel-by-voxel testing, neighborhood information is utilized, an important difference. This is achieved by using a Markov random field prior and iterated conditional modes (ICM) algorithm. However, unlike in the conventional use of ICM algorithm, the classification is based only on the distribution of background. The results from our simulations and human fMRI experiments using visual stimulation demonstrate that a better sensitivity is achieved with a given specificity in comparison to the voxel-by-voxel thresholding technique. The algorithm is computationally efficient and can be used to detect and delineate objects from a noisy background in other applications.

Algorithms↗
PubMed1996

Applying locally optimal criterion for classifying highly-reliable products with a control.

At the research and development stage, decision-makers may wish to classify several competing designs with respect to a control (or standard) one. The classification problem may become very difficult when the products are highly reliable, since only a few (or even no) failures may be observed under normal use condition. The accelerated life test model resolves this difficulty by shortening the time of life testing and quickly provides life data of products. For highly-reliable products with a Weibull log-linear model, we propose a classification rule based on a locally optimal criterion. A suitable sampling plan based on this rule is also developed. The performance of this rule is compared with a pairwise comparison classification rule. It is shown that the sample sizes needed for the new rule are considerably lower than those needed for the pairwise comparison rule.

Algorithms↗
PubMed2003

Prevalence of rheumatoid arthritis in persons 60 years of age and older in the United States: effect of different methods of case classification.

OBJECTIVE: To determine prevalence estimates for rheumatoid arthritis (RA) in noninstitutionalized older adults in the US. Prevalence estimates were compared using 3 different classification methods based on current classification criteria for RA. METHODS: Data from the Third National Health and Nutrition Examination Survey (NHANES-III) were used to generate prevalence estimates by 3 classification methods in persons 60 years of age and older (n = 5,302). Method 1 applied the "n of k" rule, such that subjects who met 3 of 6 of the American College of Rheumatology (ACR) 1987 criteria were classified as having RA (data from hand radiographs were not available). In method 2, the ACR classification tree algorithm was applied. For method 3, medication data were used to augment case identification via method 2. Population prevalence estimates and 95% confidence intervals (95% CIs) were determined using the 3 methods on data stratified by sex, race/ethnicity, age, and education. RESULTS: Overall prevalence estimates using the 3 classification methods were 2.03% (95% CI 1.30-2.76), 2.15% (95% CI 1.43-2.87), and 2.34% (95% CI 1.66-3.02), respectively. The prevalence of RA was generally greater in the following groups: women, Mexican Americans, respondents with less education, and respondents who were 70 years of age and older. CONCLUSION: The prevalence of RA in persons 60 years of age and older is approximately 2%, representing the proportion of the US elderly population who will most likely require medical intervention because of disease activity. Different classification methods yielded similar prevalence estimates, although detection of RA was enhanced by incorporation of data on use of prescription medications, an important consideration in large population surveys.

Aged↗
PubMed2001

Multiclass cancer diagnosis using tumor gene expression signatures.

The optimal treatment of patients with cancer depends on establishing accurate diagnoses by using a complex combination of clinical and histopathological data. In some instances, this task is difficult or impossible because of atypical clinical presentation or histopathology. To determine whether the diagnosis of multiple common adult malignancies could be achieved purely by molecular classification, we subjected 218 tumor samples, spanning 14 common tumor types, and 90 normal tissue samples to oligonucleotide microarray gene expression analysis. The expression levels of 16,063 genes and expressed sequence tags were used to evaluate the accuracy of a multiclass classifier based on a support vector machine algorithm. Overall classification accuracy was 78%, far exceeding the accuracy of random classification (9%). Poorly differentiated cancers resulted in low-confidence predictions and could not be accurately classified according to their tissue of origin, indicating that they are molecularly distinct entities with dramatically different gene expression patterns compared with their well differentiated counterparts. Taken together, these results demonstrate the feasibility of accurate, multiclass molecular cancer classification and suggest a strategy for future clinical implementation of molecular cancer diagnostics.

Biomarkers, Tumor↗
PubMed2006

Computer-assisted algorithms improve reliability of King classification and Cobb angle measurement of scoliosis.

STUDY DESIGN: Interobserver and intraobserver reliability study of improved method to evaluate radiographs of patients with scoliosis. OBJECTIVE: To determine the reliability of a computer-assisted measurement protocol for evaluating Cobb angle and King et al classification. SUMMARY OF BACKGROUND DATA: Evaluation of scoliosis radiographs is inherently unreliable because of technical and human judgmental errors. Objective, computer-assisted evaluation tools may improve reliability. METHODS: Posteroanterior preoperative radiographic images of 27 patients with adolescent idiopathic scoliosis were each displayed on a computer screen. They were marked 3 times in random sequence by each of 5 evaluators (observers) who marked 70 standardized points on the vertebrae and sacrum in each radiograph. A computer program (Spine 2002;27:2801-5) that identified curves, calculated Cobb angles, and generated the King et al classification automatically analyzed coordinates of these points. The interobserver and intraobserver variability of the Cobb angle and King et al classification evaluations were quantified and compared with values obtained by unassisted observers. RESULTS: Average Cobb angle intraobserver standard deviation was 2.0 degrees for both the thoracic and lumbar curves (range 0.1 to 8.3 degrees for different curves). Interobserver reliability was 2.5 degrees for thoracic curves and 2.6 degrees for lumbar curves. Among the 5 observers, there was an inverse relationship between repeatability and time spent marking images, and no correlation with image quality or curve magnitude. Kappa values for the variability of the King et al classification averaged 0.85 (intraobserver). CONCLUSIONS: Variability of Cobb measurements compares favorably with previously published series. The classification was more reliable than achieved by unassisted observers evaluating the same radiographs. The same principles may be applicable to other radiographic measurement and evaluation procedures.

Algorithms↗
PubMed2000

Intracranial pressure processing with artificial neural networks: classification of signal properties.

Intracranial pressure (ICP) is commonly used by neurosurgeons as a source of valuable information about the current condition of the neurosurgical patient. Nevertheless, despite years of effort, extracting clinically valuable information from the ICP signal is still problematical. Approaches, using current values of ICP, may fail to disclose imminent risk, because unpredictable factors can rapidly change the properties of the signal. An alternative approach is to determine some global characteristics of the signal within a longer time interval and such statistical analyses have been proposed by several authors. A further, rarely considered, problem is assessment of the results obtained from the point of view of their practical utility and/or such classification of the obtained properties of the signal that they correspond to certain clinical states of the patient. While this might be a typical task for discriminant analysis, we approached the analysis using an alternative methodology, that of computational intelligence, implemented in artificial neural networks (ANN). We tested two variants of the ANN algorithms for classification and discrimination of global properties of the ICP signal. In a "dynamic pattern classification" the network was presented with several sections of ICP records together with information from the expert-neurosurgeon, classifying 4 risk groups. In this mode no data pre-processing was carried out, in contrast to our second approach, in which the signal had been pre-processed using published statistical analyses and only these intermediate coefficients were fed into the ANN classifier. The results obtained with both classification methods at their current stage of training were similar and approximated to a 70% rate of judgements consistent with the expert scoring. Nevertheless, the method based on the assessment of global parameters from the ICP record looks more promising, because it leaves the possibility for modification of the set of parameters analysed. The new parameters may include information extracted not only from the ICP signal, but also from other diagnostic modalities, like colour coded Doppler ultrasonography. The ultimate goal of this work is to build up a pseudo-intelligent computer expert system, which would be able to reason from a reduced set of input information, available from a standard monitoring modality, because it had been taught salient links between these data and higher-order data, upon which expert scoring was based.

Cerebral Hemorrhage↗
PubMed1989

A psychiatric patient classification system. An alternative to diagnosis-related groups.

It is generally accepted that diagnosis-related groups (DRGs) for alcohol, drug, and mental disorders are inappropriate for inpatient prospective payment. To address this issue, the Veterans Administration (VA) supported a project to construct alternative classes that are more clinically meaningful, more homogeneous in their resource use, and that account for more variation in resource use among psychiatric and substance use cases than existing DRGs. This paper reports on this project. Using a data set containing universally available discharge data plus behavioral, social, and functional information obtained by a survey of 116,191 discharges from VA psychiatric beds, and with AUTOGRP as the classifying algorithm, a classification system was formed. Twelve psychiatric diagnostic groupings (PDGs) were identified, analogous to major diagnostic groups in the DRG system. Within each PDG, from 4 to 9 terminal groups of Psychiatric Patient Classes (PPCs) were formed and validated. The 12 substance abuse PPCs explain greater than 31% of the variation in length of stay; for the mental disorder PPCs the variance explanation is greater than 11%, a substantial improvement over DRGs that, for the same data set, explain less than 2 and 3%, respectively. With the addition of only 5 variables beyond those presently included in discharge data sets, greater precision for payment purposes can be achieved. Implications for adoption of this classification system are discussed.

Algorithms↗
PubMed2004

Active set support vector regression.

This paper presents active set support vector regression (ASVR), a new active set strategy to solve a straightforward reformulation of the standard support vector regression problem. This new algorithm is based on the successful ASVM algorithm for classification problems, and consists of solving a finite number of linear equations with a typically large dimensionality equal to the number of points to be approximated. However, by making use of the Sherman-Morrison-Woodbury formula, a much smaller matrix of the order of the original input space is inverted at each step. The algorithm requires no specialized quadratic or linear programming code, but merely a linear equation solver which is publicly available. ASVR is extremely fast, produces comparable generalization error to other popular algorithms, and is available on the web for download.

Algorithms↗