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 271 records · Page 15Linked to original sources
PubMed2005

Dirt detection on brown eggs by means of color computer vision.

In the last 20 yr, different methods for detecting defects in eggs were developed. Until now, no satisfying technique existed to sort and quantify dirt on eggshells. The work presented here focuses on the design of an off-line computer vision system to differentiate and quantify the presence of different dirt stains on brown eggs: dark (feces), white (uric acid), blood, and yolk stains. A system that provides uniform light exposure around the egg was designed. In this uniform light, pictures of dirty and clean eggs were taken, stored, and analyzed. The classification was based on a few standard logical operators, allowing for a quick implementation in an online set-up. In an experiment, 100 clean and 100 dirty eggs were used to validate the classification algorithm. The designed vision system showed an accuracy of 99% for the detection of dirt stains. Two percent of the clean eggs had a light-colored eggshell and were subsequently mistaken for showing large white stains. The accuracy of differentiation of the different kinds of dirt stains was 91%. Of the eggs with dark stains, 10.81% were mistaken for having bloodstains, and 33.33% of eggs with bloodstains were mistaken for having dark stains. The developed system is possibly a first step toward an on line dirt evaluation technique for brown eggs.

Animals↗
PubMed1995

Proposed criteria for the diagnosis of well-differentiated endometrial carcinoma. A diagnostic test for myoinvasion.

Existing criteria for separating clinically benign but architecturally complex or cytologically atypical endometrial proliferations (hyperplasia or metaplasia) from well-differentiated endometrial carcinoma are underspecified and poorly reproducible, in part due to the absence of a uniformly agreed on methodologically independent outcome against which to judge the efficacy of competing sets of criteria. Because myoinvasion is the first unambiguous indicator of clinically aggressive behavior for proliferations in this spectrum, we have employed the presence or absence of myoinvasion as a tool to develop clinically meaningful diagnostic criteria for the separation of complex atypical hyperplasia/metaplasia from well-differentiated carcinoma (CAHM/WDCA). We obtained the paired endometrial samplings and hysterectomy specimens of 520 patients; these were split into a training set of 306 cases and a test set of 214. The presence or absence of myoinvasion was assessed from an examination of the hysterectomy specimen. For the purposes of this study, myoinvasion was defined as the presence of irregular intramyometrial glands surrounded by a granulation tissue-like response. To determine the morphologic features that were most predictive of myoinvasion, a series of endometrial architectural, cytological, and stromal features was initially evaluated on the training set (149 myoinvasive and 157 nonmyoinvasive). Using a variety of exploratory data techniques including the classification algorithm CART, we developed a diagnostic rule for predicting myoinvasion that employed one architectural feature (glandular complexity captured by a pictorial architectural index) and two cytological features (nuclear pleomorphism and prominence of nucleoli). Extensive squamous differentiation, fibroblastic stroma, necrosis, stromal foam cells, and other cytologic features did not provide additional predictive value when cross-validated. The true misclassification rate of the CART-generated prediction rule was further assessed by applying the rule to the test set drawn largely from community hospitals. The sensitivity and specificity of this rule for detecting myoinvasion was 99.5 and 57%. The likelihood ratio was 2:1, (i.e., using prior odds of myoinvasion in the CAHM/WDCA spectrum of 1:10, the posterior odds on myoinvasion using the CART-generated rule would be 1:5). Comparison of the CART-generated myoinvasion prediction rule with the Kurman and Norris endometrial stromal invasion criteria for well differentiated endometrial carcinoma (25), using receiver operator characteristic curve (ROC) techniques, demonstrated a significant improvement in the ability to separate myoinvasive from non-myoinvasive endometrial proliferations with the CART-generated rule; the average area under the curve for the CART-generated rule was 0.78 (SE = 0.02) versus 0.67 (SE = 0.03) for the endometrial stromal invasion criteria.(ABSTRACT TRUNCATED AT 400 WORDS)

Adenocarcinoma↗
PubMed1999

EEG source localization and imaging using multiple signal classification approaches.

Equivalent current dipoles are a powerful tool for modeling focal sources. The dipole is often sufficient to adequately represent sources of measured scalp potentials, even when the area of activation exceeds 1 cm2 of cortex. Traditional least-squares fitting techniques involve minimization of an error function with respect to the location and orientation of the dipoles. The existence of multiple local minima in this error function can result in gross errors in the computed source locations. The problem is further compounded by the requirement that the model order, i.e. the number of dipoles, be determined before error minimization can be performed. An incorrect model order can produce additional errors in the estimated source parameters. Both of these problems can be avoided using alternative search strategies based on the MUSIC (multiple signal classification) algorithm. Here the authors review the MUSIC approach and demonstrate its application to the localization of multiple current dipoles from EEG data. The authors also show that the number of detectable sources can be determined in a recursive manner from the data. Also, in contrast to least-squares, the method can find dipolar sources in the presence of additional non-dipolar sources. Finally, extensions of the MUSIC approach to allow the modeling of distributed sources are discussed.

Action Potentials↗
PubMed2004

Bioinformatic analysis of the urine proteome of acute allograft rejection.

The urinary proteome in health and disease attracts increasing attention because of the potential diagnostic and pathophysiologic biomarker information carried by specific excreted proteins or their constellations. This cross-sectional study aimed to analyze the urinary proteome in patients with biopsy-proven acute rejection (n = 23) compared with transplant recipients with stable graft function (n = 22) and healthy volunteers (n = 20) and to correlate this with clinical, morphologic, and laboratory data. Urine samples were preadsorbed on four different protein chip surfaces, and the protein composition was analyzed using a surface-enhanced laser desorption/ionization time-of-flight mass spectrometer platform. The data were analyzed using two independent approaches to sample classification. Patients who experienced acute rejection could be distinguished from stable patients with a sensitivity of 90.5 to 91.3% and a specificity of 77.2 to 83.3%, depending on the classifier used. Protein masses that were important in constructing the classification algorithms included those of mass 2003.0, 2802.6, 4756.3, 5872.4, 6990.6, 19,018.8, and 25,665.7 Da. Normal urine was distinguished from transplant urine using a protein marker of mass 78,531.2 Da with both a sensitivity and a specificity of 100%. In conclusion, (1) urine proteome in transplant recipients with stable graft function was significantly different from healthy control subjects, and (2) acute rejections were characterized by a constellation of excreted proteins. Analysis of the urinary proteome may expedite the noninvasive prediction of acute graft rejection, thus importantly assisting in establishing the diagnosis.

Acute Disease↗
PubMed1995

The application of cepstral coefficients and maximum likelihood method in EMG pattern recognition.

A new technique for classifying patterns of movement via electromyographic (EMG) signals is presented. Two methods (conventional autoregressive (AR) coefficients and cepstral coefficients) for extracting features from EMG signals and three classification algorithms (Euclidean Distance Measure (EDM), Weighted Distance Measure (WDM), and Maximum Likelihood Method (MLM)) for discriminating signals representative of broad classes of movements are described and compared. These three classifiers are derived from Bayes classifier with some assumptions, the relationship among them is discussed. The conventional MLM is modified to avoid heavy matrix inversion. Six able-bodied subjects with two pairs of surface electrodes located on bilateral sternocleidomastoid and upper trapezius muscles were studied in the experiment. The EMG signals of 20 repetitions of 10 motions were analyzed for each subject. Experimental results showed that mean recognition rate of the cepstral coefficients was at least 5% superior to that of the AR coefficients. The improvement achieved by the cepstral method was statistically significant for all the three classifiers. Reasons for the superiority of cepstral features were investigated from the feature space and frequency domain, respectively. The cepstral coefficients owned better cluster separability in feature space and they emphasized the more informative part in the frequency domain. The discrimination rate of the MLM was the highest among three classifiers. Incorporation of the cepstral features with the MLM could reduce the misclassification rate by 10.6% when compared with the combination of AR coefficients and EDM. Proper choice of five of ten motions could further raise the recognition rate to more than 95%.

Adult↗
PubMed2000

Estimating neural sources from each time-frequency component of magnetoencephalographic data.

We have developed a method that incorporates the time-frequency characteristics of neural sources into magnetoencephalographic (MEG) source estimation. This method, referred to as the time-frequency multiple-signal-classification algorithm, allows the locations of neural sources to be estimated from any time-frequency region of interest. In this paper, we formulate the method based on the most general form of the quadratic time-frequency representations. We then apply it to two kinds of nonstationary MEG data: gamma-band (frequency range between 30-100 Hz) auditory activity data and spontaneous MEG data. Our method successfully detected the gamma-band source slightly medial to the N1m source location. The method was able to selectively localize sources for alpha-rhythm bursts at different locations. It also detected the mu-rhythm source from the alpha-rhythm-dominant MEG data that was measured with the subject's eyes closed. The results of these applications validate the effectiveness of the time-frequency MUSIC algorithm for selectively localizing sources having different time-frequency signatures.

Acoustic Stimulation↗
PubMed2002

Optimal excitation wavelengths for discrimination of cervical neoplasia.

Fluorescence spectroscopy has shown promise for the in vivo, real-time detection of cervical neoplasia. However, selection of excitation wavelength has in the past been based on in vitro studies and the availability of light sources. The goal of this study was to determine optimal excitation wavelengths for in vivo detection of cervical neoplasia. Fluorescence excitation-emission matrices (EEMs) were measured in vivo from 351 sites in 146 patients. Data were analyzed in pairs of diagnostic classes to determine which combination of excitation wavelengths yields classification algorithms with the greatest sensitivity and specificity. We find that 330-340-, 350-380-, and 400-450-nm excitation yield the best performance. The sensitivity and specificity for discrimination of squamous normal tissue and high-grade squamous intraepithelial lesion (HGSIL) were 71% and 77% on cross validation using three excitation wavelengths. These results are comparable with those found in earlier in vivo studies; however, in this study we find that the proportion of samples which are HGSIL influences performance. Furthermore stratification of samples within low-grade squamous intraepithelial lesion and HGSIL also appears to influence diagnostic performance. Future diagnostic studies should be carried out at these excitation wavelengths in larger groups so that data can be stratified by diagnostic subcategory, age and menopausal status. Similarly, large studies should be done in screening populations.

Carcinoma, Squamous Cell↗
PubMed2004

The Goodman-Kruskal coefficient and its applications in genetic diagnosis of cancer.

Increasing interest in new pattern recognition methods has been motivated by bioinformatics research. The analysis of gene expression data originated from microarrays constitutes an important application area for classification algorithms and illustrates the need for identifying important predictors. We show that the Goodman-Kruskal coefficient can be used for constructing minimal classifiers for tabular data, and we give an algorithm that can construct such classifiers.

Algorithms↗
PubMed2003

Computer-assisted diagnosis system in digestive endoscopy.

The purpose of this paper is to present an intelligent atlas of indexed endoscopic lesions that could be used in computer-assisted diagnosis as reference data. The development of such a system requires a mix of medical and engineering skills for analyzing and reproducing the cognitive processes that underlie the medical decision-making process. The analysis of both endoscopists experience and endoscopic terminologies developed by professional associations shows that diagnostic reasoning in digestive endoscopy uses a scene-object approach. The objects correspond to the endoscopic findings and the medical context of examination and the scene to the endoscopic diagnosis. According to expert assessment, the classes of endoscopic findings and diagnoses, their primitive characteristics (or indices), and their relationships have been listed. Each class describes an endoscopic finding or diagnosis in an intensive way. The retrieval method is based on a similarity metric that estimates the membership value of the case under investigation and the prototype of the class. A simulation test with randomized objects demonstrates a good classification of endoscopic findings. The correct class is the unique response in 68% of the tested objects, the first of multiple responses in 28%. Four descriptors are shown to be of major importance in the classification algorithm: anatomic location, shape, color, and relief. At the present time, the application database contains approximately 150 endoscopic images and is accessible via Internet. Experiments are in progress with endoscopists for the validation of the system and for the understanding of the similarity between images. The next step will integrate the system in a learning tool for junior endoscopists.

Algorithms↗
PubMed2003

Response error correction--a demonstration of improved human-machine performance using real-time EEG monitoring.

We describe a brain-computer interface (BCI) system, which uses a set of adaptive linear preprocessing and classification algorithms for single-trial detection of error related negativity (ERN). We use the detected ERN as an estimate of a subject's perceived error during an alternative forced choice visual discrimination task. The detected ERN is used to correct subject errors. Our initial results show average improvement in subject performance of 21% when errors are automatically corrected via the BCI. We are currently investigating the generalization of the overall approach to other tasks and stimulus paradigms.

Algorithms↗
PubMed2005

Closed-loop cortical control of direction using support vector machines.

Motor neuroprosthetics research has focused on reproducing natural limb motions by correlating firing rates of cortical neurons to continuous movement parameters. We propose an alternative system where specific spatial-temporal spike patterns, emerging in tasks, allow detection of classes of behavior with the aid of sophisticated nonlinear classification algorithms. Specifically, we attempt to examine ensemble activity from motor cortical neurons, not to reproduce the action this neural activity normally precedes, but rather to predict an output supervisory command to potentially control a vehicle. To demonstrate the principle, this design approach was implemented in a discrete directional task taking a small number of motor cortical signals (8-10 single units) fed into a support vector machine (SVM) to produce the commands Left and Right. In this study, rats were placed in a conditioning chamber performing a binary paddle pressing task mimicking the control of a wheelchair turning left or right. Four animal subjects (male Sprague-Dawley rats) were able to use such a brain-machine interface (BMI) with an average accuracy of 78% on their first day of exposure. Additionally, one animal continued to use the interface for three consecutive days with an average accuracy over 90%.

Algorithms↗
PubMed2006

The BCI competition. III: Validating alternative approaches to actual BCI problems.

A brain-computer interface (BCI) is a system that allows its users to control external devices with brain activity. Although the proof-of-concept was given decades ago, the reliable translation of user intent into device control commands is still a major challenge. Success requires the effective interaction of two adaptive controllers: the user's brain, which produces brain activity that encodes intent, and the BCI system, which translates that activity into device control commands. In order to facilitate this interaction, many laboratories are exploring a variety of signal analysis techniques to improve the adaptation of the BCI system to the user. In the literature, many machine learning and pattern classification algorithms have been reported to give impressive results when applied to BCI data in offline analyses. However, it is more difficult to evaluate their relative value for actual online use. BCI data competitions have been organized to provide objective formal evaluations of alternative methods. Prompted by the great interest in the first two BCI Competitions, we organized the third BCI Competition to address several of the most difficult and important analysis problems in BCI research. The paper describes the data sets that were provided to the competitors and gives an overview of the results.

Algorithms↗
PubMed1994

A neural network system for detection of atrial fibrillation in ambulatory electrocardiograms.

INTRODUCTION: A neural network classifier has been designed, which is able to distinguish atrial fibrillation (AF) from other supraventricular arrhythmias in ambulatory (Holter) ECGs. METHOD AND RESULTS: The classification algorithm uses a rhythm analysis that considers the ECG to be a time series of RR interval durations. This is combined with an analysis of baseline morphology that considers the morphological characteristics of the non-QRS portions of the waveform. A backpropagation-based neural network has been used as part of the classifier implementation. When applied to a library consisting exclusively of 42,970 examples of AF and other supraventricular rhythm disturbances validated by an experienced cardiologist, the algorithm demonstrated a sensitivity of 82.4% for 10-beat runs of paroxysmal atrial fibrillation (PAF) and a specificity of 96.6%. Since this system has been implemented as a postprocessor to a conventional automated Holter system, operating only on segments of ECG that are known to contain supraventricular arrhythmias rather than ventricular arrhythmias or sinus rhythm, it can be added to most existing Holter processing systems without significantly increasing the average time to process a tape. CONCLUSION: A neural network system has been designed, which can potentially provide, for the first time, an accurate, quantitative technique to determine the natural history of PAF and to evaluate potential treatments for PAF.

Atrial Fibrillation↗
PubMed2005

A systematic heuristic approach for feature selection for melanoma discrimination using clinical images.

BACKGROUND: Numerous features are derived from the asymmetry, border irregularity, color variegation, and diameter of the skin lesion in dermatology for diagnosing malignant melanoma. Feature selection for the development of automated skin lesion discrimination systems is an important consideration. METHODS: In this research, a systematic heuristic approach is investigated for feature selection and lesion classification. The approach integrates statistical-, correlation-, histogram-, and expert system-based components. Using statistical and correlation measures, interrelationships among features are determined. Expert system analysis is performed to identify redundant features. The feature selection process is applied to 19 shape and color features for a clinical image data set containing 355 malignant melanomas, 125 basal cell carcinomas, 177 dysplastic nevi, 199 nevocellular nevi, 139 seborrheic keratoses, and 45 vascular lesions. RESULTS: Experimental results show reduced lesion classification error rates based on condensing the shape and color feature set from 19 features to 13 features using the feature selection process. Specifically, average test lesion classification error rates for discriminating malignant melanoma from non-melanoma lesions were reduced from 26.6% for 19 features to 23.2% for 13 features over five randomly generated training and test sets. CONCLUSIONS: The experimental results show that the systematic heuristic approach for feature reduction can be successfully applied to achieve improved lesion discrimination. The feature reduction technique facilitates the elimination of redundant information that may inhibit lesion classification performance. The clinical application of this result is that automated skin lesion classification algorithm development can be fostered with systematic feature selection techniques.

Basal Cell Carcinoma↗
PubMed2003

Near-infrared thermo-optical response of the localized reflectance of intact diabetic and nondiabetic human skin.

We observed a difference in the thermal response of localized reflectance signal of human skin between type 2 diabetics and nondiabetics. We investigated the use of this thermo-optical behavior as the basis for a noninvasive method for the determination of the diabetic status of a subject. We used a two-site temperature differential method, which is predicated upon the measurement of localized reflectance from two areas on the surface of the skin. Each of these areas is subjected to a different thermal perturbation. The response of localized reflectance to temperature perturbation was measured and used in a classification algorithm. We used a discriminant function to classify subjects as diabetic or nondiabetic. In a prediction set of twenty-four noninvasive tests collected from six diabetic and six nondiabetic subjects, the sensitivity ranged between 73 and 100%, and the specificity ranged between 75 and 100%, depending on the thermal conditions and the probe-skin contact time. The difference in the thermo-optical response of the skin of the two groups is explained in terms of a difference in the response of cutaneous microcirculation, which is manifested as a difference in the near-infrared light absorption. Another factor is the difference in the temperature response of the scattering coefficient between the two groups, which may be caused by cutaneous structural differences induced by nonenzymatic glycation of skin protein fibers, and possibly by the difference in blood cell aggregation.

Algorithms↗
PubMedDate not supplied

Localization of fluorescence spots with space-space MUSIC for mammographylike measurement systems.

Breast cancer diagnosis may be improved by optical fluorescence imaging techniques in the near-infrared wavelength range. We have shown that the recently proposed space-space MUSIC (multiple signal classification) algorithm allows the 3-D localization of focal fluorophore-tagged lesions in a turbid medium from 2-D fluorescence data obtained from laser excitations at different positions. The data are assumed to be measured with two parallel planar sensor arrays on the top and bottom of the medium. The laser sources are integrated at different positions in one of the planes. The space-space data are arranged into an MxN matrix (M, number of sensors; N, number of excitation sources). A singular-value decomposition (SVD) of this matrix yields the detectable number of spot regions with linearly independent behavior with respect to the laser excitation positions and thus allows definition of a signal subspace. Matches between this signal subspace and data from model spots are tested at scanned points in a model medium viewed as the breast region under study. The locations of best matches are then considered the centers of gravity of focal lesions. The optical model used was unbounded and optically homogeneous. Nevertheless, simulated spots in bounded, inhomogeneous media modeling the breast could be localized accurately.

Algorithms↗
PubMedDate not supplied

Combined reflectance and fluorescence spectroscopy for in vivo detection of cervical pre-cancer.

Optical technologies, such as reflectance and fluorescence spectroscopy, have shown the potential to provide improved point-of-care detection methods for cervical neoplasia that are sensitive, specific, and cost-effective. Our specific goals are to analyze the diagnostic potential of reflectance and fluorescence spectra, alone and in combination, to discriminate normal and precancerous cervical tissue in vivo and to identify which classification features contain significant diagnostic information. Reflectance spectra are measured at four source-detector separations and fluorescence emission spectra are measured at 16 excitation wavelengths, from 324 sites in 161 patients. These 20 spectral features are permuted in all possible combinations of one, two, and three; and classification algorithms are developed to evaluate the diagnostic performance of each combination. Algorithms based on fluorescence spectra alone yield better diagnostic performance than those based on reflectance spectra alone. The combination of fluorescence and reflectance do not significantly improve diagnostic performance compared to fluorescence alone, except in the case of discriminating high-grade precancers from columnar normal tissue. In general, fluorescence emission spectra at 330- to 360-nm and 460- to 470-nm excitation provide the best diagnostic performance for separating all pairs of tissue categories.

Algorithms↗
PubMed2005

Category-specific cortical activity precedes retrieval during memory search.

Here we describe a functional magnetic resonance imaging study of humans engaged in memory search during a free recall task. Patterns of cortical activity associated with the study of three categories of pictures (faces, locations, and objects) were identified by a pattern-classification algorithm. The algorithm was used to track the reappearance of these activity patterns during the recall period. The reappearance of a given category's activity pattern correlates with verbal recalls made from that category and precedes the recall event by several seconds. This result is consistent with the hypothesis that category-specific activity is cueing the memory system to retrieve studied items.

Adult↗