PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Classification Algorithms”

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 1,459 records · Page 81Linked to original sources
PubMed1998

The rational clinical examination. Is this patient having a myocardial infarction?

When faced with a patient with acute chest pain, clinicians must distinguish myocardial infarction (MI) from all other causes of acute chest pain. If MI is suspected, current therapeutic practice includes deciding whether to administer thrombolysis or primary percutaneous transluminal coronary angioplasty and whether to admit patients to a coronary care unit. The former decision is based on electrocardiographic (ECG) changes, including ST-segment elevation or left bundle-branch block, the latter on the likelihood of the patient's having unstable high-risk ischemia or MI without ECG changes. Despite advances in investigative modalities, a focused history and physical examination followed by an ECG remain the key tools for the diagnosis of MI. The most powerful features that increase the probability of MI, and their associated likelihood ratios (LRs), are new ST-segment elevation (LR range, 5.7-53.9); new Q wave (LR range, 5.3-24.8); chest pain radiating to both the left and right arm simultaneously (LR, 7.1); presence of a third heart sound (LR, 3.2); and hypotension (LR, 3.1). The most powerful features that decrease the probability of MI are a normal ECG result (LR range, 0.1-0.3), pleuritic chest pain (LR, 0.2), chest pain reproduced by palpation (LR range, 0.2-0.4), sharp or stabbing chest pain (LR, 0.3), and positional chest pain (LR, 0.3). Computer-derived algorithms that depend on clinical examination and ECG findings might improve the classification of patients according to the probability that an MI is causing their chest pain.

Acute Disease↗
PubMed1994

An algorithmic approach to aspergillus sinusitis.

The effective management of paranasal sinus aspergillosis requires early diagnosis, histological classification, surgery and where appropriate, chemotherapy. Fungal sinusitis may be easily missed unless a high index of suspicion is maintained and specific culture and histology requested. The disease is classified into invasive and noninvasive types, each being divided into two subgroups: invasive aspergillosis may be either fulminant or indolent and noninvasive disease localized or allergic. The literature is reviewed and an algorithmic approach to aspergillus sinusitis proposed. The importance of histologically differentiating invasive from noninvasive aspergillosis prior to selecting the appropriate treatment options is stressed. CT scan should precede definitive surgery, and be used in follow-up. Close and prolonged follow-up is essential.

Algorithms↗
PubMed2004

Machine learning based pattern recognition applied to microarray data.

MOTIVATION: Microarrays have allowed the expression level of thousands of genes or proteins to be measured simultaneously. Data sets generated by these arrays consist of a small number of observations (e.g., 20-100 samples) on a very large number of variables (e.g., 10,000 genes or proteins). The observations in these data sets often have other attributes associated with them such as a class label denoting the pathology of the subject. Finding the genes or proteins that are correlated to these attributes is often a difficult task since most of the variables do not contain information about the pathology and as such can mask the identity of the relevant features. We describe a genetic algorithm (GA) that employs both supervised and unsupervised learning to mine gene expression and proteomic data. The pattern recognition GA selects features that increase clustering, while simultaneously searching for features that optimize the separation of the classes in a plot of the two or three largest principal components of the data. Because the largest principal components capture the bulk of the variance in the data, the features chosen by the GA contain information primarily about differences between classes in the data set. The principal component analysis routine embedded in the fitness function of the GA acts as an information filter, significantly reducing the size of the search space since it restricts the search to feature sets whose principal component plots show clustering on the basis of class. The algorithm integrates aspects of artificial intelligence and evolutionary computations to yield a smart one pass procedure for feature selection, clustering, classification, and prediction.

Algorithms↗
PubMed2002

Pre-endoscopic prognostic factors in non-varicose upper gastrointestinal bleeding. Development of a predictive algorithm.

OBJECTIVE: Upper gastrointestinal or digestive bleeding (UDB) is a common problem requiring hospitalization and implying important morbidity-mortality. The aim of the present study is to identify clinical and laboratory factors predictive of a serious course of non-varicose UDB, based on the development of a simple algorithm for application in the clinical setting. METHODS: A longitudinal study was made of 803 hospitalized patients with non-varicose UDB. Clinical and laboratory parameters were recorded in the Emergency Service before endoscopy. A first non-conditional logistic regression model was developed, including those variables exhibiting a greater capacity for predicting a serious outcome. This model was posteriorly simplified to facilitate clinical application. The prognostic performance was estimated by calculating the area under the ROC curve. RESULTS: The initial predictive model included 11 variables, with an area under the ROC curve of 0.74 (95%CI: 0.71-0.77), while in the second model these same variables were dichotomized, exhibiting an area under the ROC curve of 0.74 (95%CI: 0.70-0.77). Based on the corresponding regression coefficients, the prognostic variables were classified as either major (shock, liver disease, anticoagulant therapy, urea/creatinine ratio > 62) or minor (hematocrit < 33%, renal failure, presentation as hematemesis and melenas, systolic arterial pressure < 110 mmHg, regular alcohol consumption, a history of UDB, and a patient age of > 75 years). UDB was predicted to take an unfavorable course in the presence of one major factor or two minor factors. CONCLUSIONS: The proposed algorithm includes pre-endoscopy clinical and laboratory variables, and allows early classification of patients with non-varicose UDB in terms of their possible clinical course.

Adult↗
PubMed1994

A case-mix classification system for medical rehabilitation.

Dissatisfaction with Medicare's current system of paying for rehabilitation care has led to proposals for a rehabilitation prospective payment system, but first a classification system for rehabilitation patients must be created. Data for 36,980 patients admitted to and discharged from 125 rehabilitation facilities between January 1, 1990, and April 19, 1991, were provided by the Uniform Data System for Medical Rehabilitation. Classification rules were formed using clinical judgment and a recursive partitioning algorithm. The Functional Independence Measure version of the Function Related Groups (FIM-FRGs) uses four predictor variables: diagnosis leading to disability, admission scores for motor and cognitive functional status subscales as measured by the Functional Independence Measure, and patient age. The system contains 53 FRGs and explains 31.3% of the variance in the natural logarithm length of stay for patients in a validation sample. The FIM-FRG classification system is conceptually simple and stable when tested on a validation sample. The classification system contains a manageable number of groups, and may represent a solution to the problem of classifying medical rehabilitation patients for payment, facility planning, and research on the outcomes, quality, and cost of rehabilitation.

Activities of Daily Living↗
PubMed1998

A new quantitative criterion to distinguish between alpha/beta and alpha+beta proteins (domains).

According to the statistical analysis, it is shown that the differences of the content of alpha-helix and beta-strand between alpha/beta and alpha+beta proteins are of statistical significance. Based on the secondary structure content and the percentage of parallel or anti-parallel strands, any mixed alphabeta protein can be represented by a point in a three-dimensional prism. The distribution of the mapping points for 79 mixed alphabeta proteins (domains), of which 26 are class alpha/beta and 53 are class alpha+beta, shows that the two kinds of points are situated at distinct regions roughly. A new quantitative criterion based on the Fisher discriminant algorithm is proposed to distinguish between the alpha/beta and alpha+beta proteins (domains). Of the 79 proteins 77 are correctly classified (97.5%). As a stringent cross-validation test, the jackknife test shows that of the 79 proteins 77 are correctly classified. The jackknife test accuracy is still 97.5%. These figures indicate the self-consistence and the extrapolating effectiveness of the new quantitative criterion. Applying the new criterion to reclassify the alpha/beta and alpha+beta proteins (domains) in SCOP is also discussed. It is hoped that the new quantitative criterion will be useful for the development of protein classification databases.

Algorithms↗
PubMed1997

Assessment of potential indicators for protein-energy malnutrition in the algorithm for integrated management of childhood illness.

Potential indicators were assessed for the two classifications of protein-energy malnutrition in the guidelines for integrated management of childhood illness: severe malnutrition, which requires immediate referral to hospital, and very low weight, which calls for feeding assessment, nutritional counselling and follow-up. Children aged < 2 years require feeding assessment and counselling as a preventive intervention. For severe malnutrition, we examined 1202 children admitted to a Kenyan hospital for any association of the indicators with mortality within one month. Bipedal oedema indicating kwashiorkor, and two marasmus indicators (visible severe wasting and weight-for-height (WFH) Z-score of < -3) were associated with a significantly increased mortality risk (odds ratios, 3.1-3.9). Very low weight-for-age (WFA) (Z-score of < -4.4) was not associated with an increased risk of mortality. Because first-level health facilities generally lack length-boards, bipedal oedema and visible severe wasting were chosen as indicators of severe malnutrition. To assess potential WFA thresholds for the very low weight classification, our primary source of data came from 1785 Kenyan outpatient children, but we also examined data from surveys in Nepal, Bolivia, and Togo. We examined the performance of WFA at various thresholds to identify children with low WFH and, for children aged < or = 2 years, low height-for-age (HFA). Use of a WFA threshold Z-score of < -2 identified a considerable proportion of children (from 13% in Bolivia to 68% in Nepal) which, in most settings, would pose an enormous burden on the health facility. Among ill children in Kenya, a threshold WFA Z-score of < -3 had a sensitivity of 89-100% to detect children with WFH Z-scores of < -3, and, with an identification rate of 9%, would avoid overburdening the clinics. Potential modifications include use of a more restrictive cut-off in countries with high rates of stunting, or the elimination of the WFA screen in order to concentrate efforts on intervention for all children below the 2-year age cut-off. Key issues in every country include the capacity to provide counselling for many children and linkage to nutritional improvement programmes in the community.

Age Factors↗
PubMed1998

A map of the protein space--an automatic hierarchical classification of all protein sequences.

We investigate the space of all protein sequences. We combine the standard measures of similarity (SW, FASTA, BLAST), to associate with each sequence an exhaustive list of neighboring sequences. These lists induce a (weighted directed) graph whose vertices are the sequences. The weight of an edge connecting two sequences represents their degree of similarity. This graph encodes much of the fundamental properties of the sequence space. We look for clusters of related proteins in this graph. These clusters correspond to strongly connected sets of vertices. Two main ideas underlie our work: i) Interesting homologies among proteins can be deduced by transitivity. ii) Transitivity should be applied restrictively in order to prevent unrelated proteins from clustering together. Our analysis starts from a very conservative classification, based on very significant similarities, that has many classes. Subsequently, classes are merged to include less significant similarities. Merging is performed via a novel two phase algorithm. First, the algorithm identifies groups of possibly related clusters (based on transitivity and strong connectivity) using local considerations, and merges them. Then, a global test is applied to identify nuclei of strong relationships within these groups of clusters, and the classification is refined accordingly. This process takes place at varying thresholds of statistical significance, where at each step the algorithm is applied on the classes of the previous classification, to obtain the next one, at the more permissive threshold. Consequently, a hierarchical organization of all proteins is obtained. The resulting classification splits the space of all protein sequences into well defined groups of proteins. The results show that the automatically induced sets of proteins are closely correlated with natural biological families and super families. The hierarchical organization reveals finer sub-families that make up known families of proteins as well as many interesting relations between protein families. The hierarchical organization proposed may be considered as the first map of the space of all protein sequences. An interactive web site including the results of our analysis has been constructed, and is now accessible through http:/(/)www.protomap.cs.huji.ac.il

Algorithms↗
PubMed2004

[Differential diagnosis of acute ischemic stroke and management on the basis of acute ischemic stroke subtype].

Recent advances in neuro-imaging technology assist early clinical diagnosis for ischemic stroke subtype. The precise early diagnosis for stroke subtype plays an important role for the management in patients with acute cerebral infarction. The differential diagnosis is made according to the algorithm in the modified the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification, with reference to the results of diffusion-weighted MRI and MRA. Artery-to-artery embolism and branch atheromatus disease are diagnosed as atherothrombotic infarction according to this algorithm. Thrombolytic therapy is recommended in cardioembolic stroke within 3 hours after onset. The anticoagulant therapy with heparin is often used to prevent the recurrence for thrombosis in acute cardioemboic infarction. The selective thrombin inhibitor has recently been used in the treatment of acute atherothrombotic infarction. The antiplatelet therapy with aspirin is recommended in acute atherothrombotic infarction and lacunar infarction. The sodium ozagrel is recommended in the treatment of acute lacunar infarction. The treatment of acute ischemic stroke should be managed according to Japanese Guidelines for the management of stroke (2004).

Acute Disease↗
PubMed1986

Allele and locus classification in electrophoretic population studies.

The electrophoretic separation of protein variants having slightly different mobilities is a basic tool of biochemical population genetics. In certain situations it is difficult to determine how to classify the variants as alleles of a number of genetic loci, that is, as variant subsets within each of which the Mendelian laws hold. In this article, we develop and analyze a series of algorithms for solving various versions and generalizations of this problem of optimal classification.

Alleles↗
PubMed2005

Graph embedding to improve supervised classification and novel class detection: application to prostate cancer.

Recently there has been a great deal of interest in algorithms for constructing low-dimensional feature-space embeddings of high dimensional data sets in order to visualize inter- and intra-class relationships. In this paper we present a novel application of graph embedding in improving the accuracy of supervised classification schemes, especially in cases where object class labels cannot be reliably ascertained. By refining the initial training set of class labels we seek to improve the prior class distributions and thus classification accuracy. We also present a novel way of visualizing the class embeddings which makes it easy to appreciate inter-class relationships and to infer the presence of new classes which were not part of the original classification. We demonstrate the utility of the method in detecting prostatic adenocarcinoma from high-resolution MRI.

Adenocarcinoma↗
PubMed2003

Greedy mixture learning for multiple motif discovery in biological sequences.

MOTIVATION: This paper studies the problem of discovering subsequences, known as motifs, that are common to a given collection of related biosequences, by proposing a greedy algorithm for learning a mixture of motifs model through likelihood maximization. The approach adds sequentially a new motif to a mixture model by performing a combined scheme of global and local search for appropriately initializing its parameters. In addition, a hierarchical partitioning scheme based on kd-trees is presented for partitioning the input dataset in order to speed-up the global searching procedure. The proposed method compares favorably over the well-known MEME approach and treats successfully several drawbacks of MEME. RESULTS: Experimental results indicate that the algorithm is advantageous in identifying larger groups of motifs characteristic of biological families with significant conservation. In addition, it offers better diagnostic capabilities by building more powerful statistical motif-models with improved classification accuracy.

Algorithms↗
PubMed2000

The Bayesian evidence scheme for regularizing probability-density estimating neural networks.

Training probability-density estimating neural networks with the expectation-maximization (EM) algorithm aims to maximize the likelihood of the training set and therefore leads to overfitting for sparse data. In this article, a regularization method for mixture models with generalized linear kernel centers is proposed, which adopts the Bayesian evidence approach and optimizes the hyperparameters of the prior by type II maximum likelihood. This includes a marginalization over the parameters, which is done by Laplace approximation and requires the derivation of the Hessian of the log-likelihood function. The incorporation of this approach into the standard training scheme leads to a modified form of the EM algorithm, which includes a regularization term and adapts the hyperparameters on-line after each EM cycle. The article presents applications of this scheme to classification problems, the prediction of stochastic time series, and latent space models.

Acquired Immunodeficiency Syndrome↗
PubMedDate not supplied

Scalable partitioning and exploration of chemical spaces using geometric hashing.

Virtual screening (VS) has become a preferred tool to augment high-throughput screening(1) and determine new leads in the drug discovery process. The core of a VS informatics pipeline includes several data mining algorithms that work on huge databases of chemical compounds containing millions of molecular structures and their associated data. Thus, scaling traditional applications such as classification, partitioning, and outlier detection for huge chemical data sets without a significant loss in accuracy is very important. In this paper, we introduce a data mining framework built on top of a recently developed fast approximate nearest-neighbor-finding algorithm(2) called locality-sensitive hashing (LSH) that can be used to mine huge chemical spaces in a scalable fashion using very modest computational resources. The core LSH algorithm hashes chemical descriptors so that points close to each other in the descriptor space are also close to each other in the hashed space. Using this data structure, one can perform approximate nearest-neighbor searches very quickly, in sublinear time. We validate the accuracy and performance of our framework on three real data sets of sizes ranging from 4337 to 249 071 molecules. Results indicate that the identification of nearest neighbors using the LSH algorithm is at least 2 orders of magnitude faster than the traditional k-nearest-neighbor method and is over 94% accurate for most query parameters. Furthermore, when viewed as a data-partitioning procedure, the LSH algorithm lends itself to easy parallelization of nearest-neighbor classification or regression. We also apply our framework to detect outlying (diverse) compounds in a given chemical space; this algorithm is extremely rapid in determining whether a compound is located in a sparse region of chemical space or not, and it is quite accurate when compared to results obtained using principal-component-analysis-based heuristics.

Journal Article↗
PubMed2005

Using LogitBoost classifier to predict protein structural classes.

Prediction of protein classification is an important topic in molecular biology. This is because it is able to not only provide useful information from the viewpoint of structure itself, but also greatly stimulate the characterization of many other features of proteins that may be closely correlated with their biological functions. In this paper, the LogitBoost, one of the boosting algorithms developed recently, is introduced for predicting protein structural classes. It performs classification using a regression scheme as the base learner, which can handle multi-class problems and is particularly superior in coping with noisy data. It was demonstrated that the LogitBoost outperformed the support vector machines in predicting the structural classes for a given dataset, indicating that the new classifier is very promising. It is anticipated that the power in predicting protein structural classes as well as many other bio-macromolecular attributes will be further strengthened if the LogitBoost and some other existing algorithms can be effectively complemented with each other.

Animals↗