SEARCH · PubMed Health

Results for “cross-validation”

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 685 records · Page 38Linked to original sources

PubMed1984

Prediction of chronicity in acute low back pain.

Forty-eight patients seeking treatment for low back pain were assessed in the acute stage to see if they exhibited characteristics that would predict response to medical treatment. It was hoped that, if there were personality and demographic differences between patients who did and those who did not respond to standard medical treatment, potential chronic pain patients could be identified and targeted for special treatment. Stepwise discriminant analysis yielded a prediction equation that correctly identified 41 of the 48 cases (85.4%). Acute patients who became chronic complained of pain over a wider area of the body; had deeper, more central pain; were highly anxious; and had a lower activity level. Although cross-validation studies are needed, identification of potentially chronic pain patients may be feasible, making appropriate early intervention possible.

Back Pain↗

PubMed1984

Research design and analysis of data procedures for predicting body density.

Statistical methods used to develop body composition prediction equations were examined. Equations published in the 1960s and early 1970s have been called "population-specific" equations because they were developed on homogeneous samples. A major limitation of population-specific equations was a low ratio of subjects per variable, which reduces validity. The more recent method has been to develop "generalized" equations with large samples varying greatly in age and body fatness. The statistical models used for generalized equations were nonlinear regression analysis to account for the quadratic relation between body density and skinfold fat. Age was used as an independent variable to adjust for aging. Generalized equations exhibit accuracy consistent with population-specific equations and increase the generalizability of prediction. Future body composition research should focus on the cross-validation of published equations to define the limits of equation accuracy.

Adult↗

PubMed1984

Relationship between physical, psychological, social, and environmental variables and subjective sleep quality.

In a survey study of patients of a general practitioner the relationship between sleep quality and a heterogeneous set of other variables was examined. The data file was divided randomly, and a two-staged multiple regression analysis was performed on each half. The two resulting regression equations were cross-validated on the data of the other data file. The variables mood, age, and use of medicine proved to have the most significant relationship to sleep quality.

Adult↗

PubMed1984

The influence of somatotype on anthropometric prediction of body composition in young women.

The influence of somatotype on the validity of anthropometric prediction of body density (Db) in young women (N = 92) was investigated. Three groups of predominantly endomorph (N = 27), mesomorph (N = 35), and ectomorph (N = 30) women were identified by the Heath-Carter and Sheldon somatotyping methods. Discriminant analysis revealed a 100% accuracy in somatotype group determination. Thirteen diameters, 26 girths, and 8 skinfolds were measured and used in a STEPWISE regression analysis to derive somatotype-specific regression equations to predict body density. Combining all the measures provided very good prediction accuracy in all three groups with multiple correlation coefficients (R) of 0.98, 0.90, and 0.90, and the standard error of estimate (SEE) of 0.003, 0.005, and 0.005 for Db (gm X cc-1) in the endomorphs, mesomorphs, and ectomorphs, respectively. A cross-validation study confirmed the accuracy of the somatotype-specific regression equations and demonstrated an inherent weakness in using some generalized equations on specific somatotypes. The use of non-somatotype-specific equations resulted in mean Db prediction errors ranging from -0.018 to +0.023 gm X cc-1 (8.5 to -10.7% Fat). Although all equations tested demonstrated specific weaknesses in one or more of the somatotype groups when predicting Db, the Jackson et al. (1980) equation performed better than most of the non-somatotype-specific prediction equations. These findings suggest that the anthropometric estimation of Db may not be sample specific in the same manner as had been previously thought and that greater accuracy may be achieved by using regression equations which have been generated on a previously somatotyped population sample.

Adolescent↗

PubMed1984

Predictors of performance on National Board Examinations Part II.

An effort was made to determine successful predictors of performance on National Board Examinations Part II. Preliminary results, using stepwise regression analysis, identified two predictor variables accounting for 65 percent of the variance in the criterion variable. Cross-validation regression analysis again identified the same two predictor variables accounting for 69 percent of the variance in the criterion variable. In both investigations, the correlation coefficients for the predicted National Board averages and the actual National Board averages are statistically significant. Implications of the studies are discussed.

Certification↗

PubMed1984

Neuropsychological impairment and associated intellectual functions in schizophrenic and other psychiatric patients.

As a routine admission procedure of an ongoing research program, the Halstead-Reitan Battery (HRB) and the Wechsler Adult Intelligence Scale (WAIS) were completed by 37 chronic schizophrenic patients, 3 patients with major affective disorders, and 2 with atypical psychoses. These admission data provided a basis for a retrospective investigation into the possibility of a relationship between impairment on the HRB and IQ. It was found that greater impairment on the HRB was associated with lower full-scale IQ (FSIQ) scores (r = -0.78, p less than .001). When the sample was separated into impaired and normal groups based on performance on the HRB, the coefficients of two discriminant functions, i.e., Verbal IQ and Picture Arrangement, classified the two groups with 100% accuracy. Cross-validation of the stepwise discriminant function analysis, using the jackknife procedure, resulted in a 98% hit rate. The results of the present study together with those of two other recent studies suggest that intellectual functions are critical interactive variables for interpreting the performance of chronic schizophrenic and other psychiatric groups on the HRB. Cautions to be observed in interpreting impaired and normal functioning in the context of intellectual functions are discussed.

Adult↗

PubMed1982

Steps toward the development of generalized equations for predicting body composition of adults.

The estimation of body density from anthropometric variables is often used to measure body composition. The purpose of this report was to review the development of early published equations and follow the transition from population specific equations to generalized equations for adults. The accuracy of population specific equations were improved and refined with the addition of circumference and diameter measures in combination with skinfolds and the use of populations of varied ages and degrees of body fatness. The major weakness of the population specific equations was their inability to account for aging and the non linear relationship between subcutaneous fat and body density. Regression models using the quadratic sum of skinfolds and age have been developed to account for these sources of variation. Cross-validation research has shown that these generalized regression models are accurate for adult males and females varying greatly in age and body fatness.

Adolescent↗

PubMed1983

[Comparative studies of the psychopathology of bronchial asthma and cystic fibrosis].

Comparative examinations were performed on each N = 36 subjects, with bronchial asthma, cystic fibrosis and healthy controls, with regard to individual psychopathology, family functioning, and the relationship of clinical parameters to psychopathology. Children suffering from cystic fibrosis had the highest prevalence rate of psychiatric disorder. Findings of the psychiatric assessment in general were cross-validated by parent's and teacher's questionnaires. Data relating to family function were less differentiating between the groups. In both groups of diseases significant correlations between clinical parameters and psychopathology were detected.

Adolescent↗

PubMedDate not supplied

Measurement of the nurse organizational climate of neonatal intensive care units.

The concept of organizational climate is important for the study and understanding of work organizations. Researchers have not yet studied the organizational climate of hospital intensive care units, so it is not clear that the dimensions useful in description of other work settings will be appropriate for conceptualizing neonatal intensive care units (NICU's). This study investigated nurse perceptions of the organizational climate of NICU's in a sample of 18 hospitals. The Organizational Climate Description Questionnaire was modified for use in the NICU setting. A set of homogeneous climate scales was developed from the revised instrument. The six scales of the resulting instrument, the Nursing Organizational Climate Description Questionnaire, Form B (NOCDQ-B), demonstrated median internal-consistency reliability of .69 on cross-validation. Analysis of variance indicated that each of the six NOCDQ-B dimensions served to differentiate the NICU's in the study. Further, significant relationships were found at the unit mean level between three of the six NOCDQ-B scales and nurse satisfaction, as measured by the Minnesota Satisfaction Questionnaire.

Analysis of Variance↗

PubMed1982

EEG studies of sleep in the diagnosis of depression.

Psychiatric diagnoses have traditionally been made on the basis of clinical criteria, including current phenomenology and historical information. This traditional procedure presents several problems, including standardization of data gathering and interpretation. Biological criteria have been shown to be useful aids to diagnosis, but the same problems of standardization must be overcome. We present here the derivation of discriminant functions (DFs) using sleep EEG data to separate depressed from normal subjects. More important, we have cross-validated these DFs in a separate group of patients, using them to separate endogenous (ED) from nonendogenous depressed (ND) patients. ADF using the sleep variables REM latency and REM density can make this discrimination with sensitivity = 0.61 and specificity = 0.93. We also present our preliminary findings in support of the earlier conclusion that the sleep of unipolar ED patients is more disturbed than that of bipolar ED patients.

Adult↗

PubMed1978

Objective assessment of prior air traffic control-related experience through the use of an occupational knowledge test.

The Occupational Knowledge Test (OKT) 101-B was administered to 784 air traffic control trainees who entered the FAA Academy's 16-week training course in 1976. All trainees completed the nonradar laboratory portion of the training and, in addition, completed a preemployment questionnaire. Based on responses to the questionnaire, the trainees were assigned to one of three experience groups, corresponding to groups given credit for experience using Civil Service Commission (CSC) selection procedures. It was found that the OKT was highly correlated with experience (r = 0.64) and, in addition, had a higher correlation with successful completion of the nonradar lab than did experience (r = 0.25 vs. 0.12). It was determined that use of an OKT score of 75 or above to assign extra credit would result in a failure rate of 3.1% for those receiving credit, while use of the current experience scale would result in a failure rate of 7.6% for those receiving extra credit on the CSC selection battery. The results held up for a cross-validation sample of 432 trainees who entered the Academy during 1977. Based on the results, it is recommended that an OKT score of 75 or above be used to assign extra credit for experience in the selection of air traffic controllers.

Aviation↗

PubMed1981

Development of a simplified diagnostic scoring method for the school version of the adaptive behavior scale.

The development of a standard scoring procedure by which the factor scores from the School Version of the Adaptive Behavior Scale can be combined was described. Subjects were 1,650 pupils in regular classes, 3,052 pupils in EMR programs, and 828 pupils in TMR classes from California and Florida. Discriminant analysis of the factor scores was computed in order to specify the discriminant coefficients to be used in predicting school classification status. Cross-validation of the discriminant analysis was obtained, and a procedure for creating a composite score was described.

Adaptation, Psychological↗

PubMed1980

Generalized equations for predicting body density of women.

Previous research with women has shown that body composition regression equations derived from anthropometric variables were population specific. This study sought to derive generalized equations for women differing in age and body composition. The hydrostatic method was used to determine body density (BD) and percent fat (%F) on 249 women in 18 to 55 years (X = 31.4 +/- 10.8 yrs) and 4 to 44 %F (X = 24.1 +/-7.2 %F). Skinfold fat (S), gluteal circumference (C) and age were independent variables. The quadratic form of the sum of three, four and seven S in combination with age and gluteal C produced multiple correlations that ranged from 0.842 to 0.867 with standard errors of 3.6 to 3.8 %F. The equations were cross-validated on a different sample of 82 women with similar age and %F characteristics. The correlations between predicted and hydrostatically determined %F ranged from 0.815 to 0.820 with standard errors of 3.7 to 4.0 %F. This study showed that valid generalized body composition equations could be derived for women varying in age and body composition, but care need to be exercised with women over an age of forty.

Adipose Tissue↗

PubMed1980

A questionnaire study of psychological background factors in U.S. Navy aircraft accidents.

A questionnaire concerning life changes, personality factors, and adjustmental and leadership qualities of U.S. Naval aircrewmembers involved in aircraft accidents was sent to investigating flight surgeons during 1977-78. The responses were divided into two groups: those who were causally involved in accidents and those who were not. In order to cross-validate the results, data were collected and analyzed during 1979. Results indicate that aircrewmembers in the process of deciding about staying in the service are more likely to fall into the causally involved group. So were those who had trouble with interpersonal relationships, had no sense of humor or humility concerning themselves, were immature, or had recently lost a friend or family member through death.

Accidents, Aviation↗

PubMed1993

Knowledge discovery in GenBank.

We describe various methods designed to discover knowledge in the GenBank nucleic acid sequence database. Using a grammatical model of gene structure, we create a parse tree of a gene using features listed in the FEATURE TABLE. The parse tree infers features that are not explicitly listed, but which follow from the listed features. This method discovers 30% more introns and 40% more exons when applied to a globin gene subset of GenBank. Parse tree construction also entails resolving ambiguity and inconsistency within a FEATURE TABLE. We transform the parse tree into an augmented FEATURE TABLE that represents inferred gene structure explicitly and unambiguously, thereby greatly improving the utility of the FEATURE TABLE to researchers. We then describe various analogical reasoning techniques designed to exploit the homologous nature of genes. We build a classification hierarchy that reflects the evolutionary relationship between genes. Descriptive grammars of gene classes are then induced from the instance grammars of genes. Case based reasoning techniques use these abstract gene class descriptions to predict the presence and location of regulatory features not listed in the FEATURE TABLE. A cross-validation test shows a success rate of 87% on a globin gene subset of GenBank.

Algorithms↗

PubMed1994

The prediction of the degree of exposure to solvent of amino acid residues via genetic programming.

In this paper I evolve programs that predict the degree of exposure to solvent (the buriedness) of amino acid residues given only the primary structure. I use genetic programming (Koza 1992; Koza 1994) to evolve programs that take as input the primary structure and that output the buriedness of each residue. I trained these programs on a set of 82 proteins from the Brookhaven Protein Data Bank (PDB) (Bernstein et al. 1977) and cross-validated them on a separate testing set of 40 proteins, also from the PDB. The best program evolved had a correlation of 0.434 between the predicted and observed buriednesses on the testing set.

Amino Acids↗

PubMed1995

Predicting free energy contributions to the conformational stability of folded proteins from the residue sequence with radial basis function networks.

Radial basis function neural networks are trained on a data base comprising 38 globular proteins of well resolved crystallographic structure and the corresponding free energy contributions to the overall protein stability (as computed partially from chrystallographic analysis and partially with multiple regression from experimental thermodynamic data by Ponnuswamy and Gromiha (1994)). Starting from the residue sequence and using as input code the percentage of each residue and the total residue number of the protein, it is found with a cross-validation method that neural networks can optimally predict the free energy contributions due to hydrogen bonds, hydrophobic interactions and the unfolded state. Terms due to electrostatic and disulfide bonding free energies are poorly predicted. This is so also when other input codes, including the percentage of secondary structure type of the protein and/or residue-pair information are used. Furthermore, trained on the computed and/or experimental delta G values of the data base, neural networks predict a conformational stability ranging from about 10 to 20 kcal mol-1 rather independently of the residue sequence, with an average error per protein of about 9 kcal mol-1.

Calorimetry↗

PubMed1995

Image analysis and machine learning applied to breast cancer diagnosis and prognosis.

Fine needle aspiration (FNA) accuracy is limited by, among other factors, the subjective interpretation of the aspirate. We have increased breast FNA accuracy by coupling digital image analysis methods with machine learning techniques. Additionally, our mathematical approach captures nuclear features ("grade") that are prognostically more accurate than are estimates based on tumor size and lymph node status. An interactive computer system evaluates, diagnoses and determines prognosis based on nuclear features derived directly from a digital scan of FNA slides. A consecutive series of 569 patients provided the data for the diagnostic study. A 166-patient subset provided the data for the prognostic study. An additional 75 consecutive, new patients provided samples to test the diagnostic system. The projected prospective accuracy of the diagnostic system was estimated to be 97% by 10-fold cross-validation, and the actual accuracy on 75 new samples was 100%. The projected prospective accuracy of the prognostic system was estimated to be 86% by leave-one-out testing.

Biopsy, Needle↗