Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Multivariate statistics”

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 649 records · Page 36Linked to original sources

The contribution of image cytometry to the characterization of clinical subgroups of lipomas.

The aim of the present study was to investigate whether biological features determined through image cytometry are able to characterize clinical subpopulations of lipomas. Forty lipomas excised from 36 patients were studied. On the one hand, the tumors were clinically characterized by means of patient-related and pre- and post-operative features. On the other, the tumors were analyzed by means of the computer-assisted microscopy analysis of Feulgen-stained nuclei. This analysis generated 3 groups of biological quantitative variables describing morphonuclear aspects (i.e. the chromatin pattern of the cell nuclei), the nuclear DNA content (DNA ploidy level), and architectural features (such as the cell density and the topographical cell nuclei organization). Possible relations between the clinical and the biological features of the lipomas were investigated by means of univariate and multivariate statistical analyses. The results show the existence of such relations, in particular between the morphonuclear and architectural features of lipomas, on the one hand, and their consistency, volume and weight, on the other. Furthermore, multivariate analysis made it possible to distinguish two subpopulations of lipomas exhibiting different biological characteristics in terms of morphonuclear patterns.

Adult↗

DNA microarrays in prostate cancer.

DNA microarray technology provides a means to examine large numbers of molecular changes related to a biological process in a high throughput manner. This review discusses plausible utilities of this technology in prostate cancer research, including definition of prostate cancer predisposition, global profiling of gene expression patterns associated with cancer initiation and progression, identification of new diagnostic and prognostic markers, and discovery of novel patient classification schemes. The technology, at present, has only been explored in a limited fashion in prostate cancer research. Some hurdles to be overcome are the high cost of the technology, insufficient sample size and repeated experiments, and the inadequate use of bioinformatics. With the completion of the Human Genome Project and the advance of several highly complementary technologies, such as laser capture microdissection, unbiased RNA amplification, customized functional arrays (eg, single-nucleotide polymorphism chips), and amenable bioinformatics software, this technology will become widely used by investigators in the field. The large amount of novel, unbiased hypotheses and insights generated by this technology is expected to have a significant impact on the diagnosis, treatment, and prevention of prostate cancer. Finally, this review emphasizes existing, but currently underutilized, data-mining tools, such as multivariate statistical analyses, neural networking, and machine learning techniques, to stimulate wider usage.

Algorithms↗

Sexual health education for youths--a Malaysian experience.

A survey was conducted to assess student's sexual knowledge and attitudes using a questionnaire based on the Sex Knowledge and Attitude Test (SKAT-II) to compare medical and nursing students with students (non-medical/nursing) who registered for a sexual health course. 85 Sexual Health, 115 medical and 81 nursing students voluntarily participated in the survey. This study showed that all the student groups showed relatively low scores in knowledge. Furthermore, average knowledge scores differed significantly between the three student groups with medical students scoring highest and nursing students lowest. Besides student groups, several other factors were found to be significantly associated with Knowledge score namely, race, religion, age, perception of the importance of religion and the extent to which religious beliefs influence sexual attitudes. Furthermore, multivariate statistical analyses showed that among these factors, student group, race/religion and religious importance were significant predictors of sexual knowledge. Specifically, being a medical student was associated with higher scores relative to a non-medical student, being a Malay student was independently associated with a lower average score compared to other races, and perceiving religion as extremely important was associated with a lower score.

Adolescent↗

Retrospective statistical analysis of lyophilized protein formulations of progenipoietin using PLS: determination of the critical parameters for long-term storage stability.

Although certain criteria have become recognized as being essential for a stable lyophilized formulation, the relative importance of different stability criteria has not been demonstrated quantitatively. This study uses multivariate statistical methods to determine the relative importance of certain formulation variables that affect long-term storage stability of a therapeutic protein. Using the projection to latent structures (PLS) method, a retrospective analysis was conducted of 18 formulations of progenipoietin (ProGP), a potential protein therapeutic agent. The relative importance of composition, pH, maintenance of protein structure (as determined by infrared (IR) spectroscopy), and thermochemical properties of the glassy state (as measured by differential scanning calorimetry (DSC)) were evaluated. Various stability endpoints were assessed and validated models constructed for each using the PLS method. Retention of parent protein and the appearance of degradation products could be adequately modeled using PLS. The models demonstrate the importance of retention of native structure in the solid state and controlling the pH. The relative importance of T(g) in affecting storage stability was low, as all of the samples had T(g) values above the highest storage temperature (40 degrees C). However, other indicators of molecular mobility in the solid state, such as change in DeltaC(p) upon annealing, appear to be important, even for storage below T(g). For the first time, the relative importance of certain properties in controlling long-term storage stability could be assessed quantitatively. In general, the most important parameters appear to be pH and retention of native structure in the solid state. However, for some stability endpoints, the composition (concentration of protein or various excipients), as well as some DSC parameters, were found to be significant in predicting long-term stability.

Calorimetry, Differential Scanning↗

Quantitative EEG abnormalities and asymmetries in patients with intracranial tumors.

Quantified EEG data of 50 patients with intracranial tumors (T) were compared to the data profile of 60 normal adults (N), using univariate and multivariate statistical analyses. EEGs were recorded from 16 derivations. Twelve spectral parameters were computed from each lead and used for statistical comparisons. The number of significant deviations (P less than 0.05) among univariate features did not separate the T subjects from the N subjects with an acceptable sensitivity. A significant improvement was achieved by using multivariate features and particularly asymmetry multivariate features which detected 49/50 of the T subjects. The proportion of correctly lateralized tumors was 43/48.

Adult↗

Multiregion profiling of genomic and transcriptional heterogeneity in head and neck squamous-cell carcinoma.

BACKGROUND: Intratumoral heterogeneity (ITH) is thought to contribute to tumour evolution and treatment resistance but its biological and clinical significance in localised head and neck squamous-cell carcinoma (HNSCC) remains incompletely understood. PATIENTS AND METHODS: In the prospective SCANDARE study, we analysed 87 patients with resectable HNSCC treated with upfront surgery. Two to five spatially distinct tumour regions per patient underwent pathological evaluation, targeted DNA sequencing, and bulk RNA sequencing. Genomic ITH (gITH) was quantified using clonal deconvolution and Shannon diversity indices, whereas transcriptional heterogeneity (tITH) was assessed using the intratumour expression distance metric. Associations between ITH, molecular features, tumour microenvironment composition, and clinical outcomes were explored using multivariable statistical models. RESULTS: Pathology-based spatial heterogeneity showed limited prognostic value. gITH was common, with 37% of tumours displaying regionally heterogeneous pathogenic variants, including spatially actionable alterations in 10% of patients. In an initial multivariable Cox model, higher gITH was associated with shorter disease-free survival. However, after Ridge-penalised modelling and bootstrap internal validation, the effect size was attenuated [corrected hazard ratio 1.42, 95% confidence interval (CI) 0.91-2.75]. The overall model retained moderate discriminative performance (optimism-corrected C-index 0.69, 95% CI 0.59-0.79). gITH was associated with tumour cellularity, reduced estimated endothelial cell infiltration, and alterations in KMT2C and PIK3CA. tITH differed according to human papillomavirus (HPV) status, with lower tITH in HPV-positive tumours, and was associated with distinct biological pathways and genomic alterations. Genomic and tITH were not correlated. CONCLUSIONS: This prospective multiregion study provides a comprehensive characterisation of genomic and tITH in localised HNSCC. Our findings highlight substantial spatial molecular diversity within primary tumours and suggest potential associations between heterogeneity, tumour biology, and clinical outcome that warrant validation in independent cohorts.

head and neck squamous-cell carcinoma (HNSCC)↗

Subtypes of mania determined by grade of membership analysis.

Classical descriptions of mania subtypes extend back to Kraepelin; however, in marked contrast to the study of depression subtypes, validation of mania subtypes by multivariate statistical methods has seldom been attempted. We applied Grade of Membership (GOM) analysis to the rated clinical features of 327 inpatients with DSM-III-R mania diagnoses. GOM is a type of latent structure multivariate analysis, which differs from others of this type in making no a priori distributional assumptions about groupings. We obtained 5 GOM Pure Types with good face validity. The major Kraepelinian forms of "hypomania," "acute mania," "delusional mania," and "depressive or anxious mania" were validated. The major new finding is of two mixed mania presentations, each with marked lability of mood. The first of these displayed a dominant mood of severe depression with labile periods of pressured, irritable hostility and paranoia, and the complete absence of euphoria or humor. The second mixed mania Pure Type displayed a true, incongruous mixture of affects: periods of classical manic symptoms with euphoria, elation, humor, grandiosity, psychosis, and psychomotor activation, switching frequently to moderately depressed mood with pressured anxiety and irritability. This multivariate analysis validated classical clinical descriptions of the major subtypes of mania. Two distinct forms of mixed manic episodes were identified. DSM-III-R criteria did not reliably identify either of these two natural groups of mixed bipolar patients. As occurs in depression, this clinical heterogeneity of mania may influence response to drug treatments.

Adolescent↗

Cytosolic free calcium in platelets: relationships to blood pressure and indices of systemic calcium metabolism.

Relationships between cytosolic free calcium ([Ca2+]i) in platelets, indices of systemic calcium metabolism and blood pressure were examined in 86 subjects; 29 patients with untreated and 29 patients with treated essential hypertension, six patients with borderline hypertension and 22 healthy reference subjects. In order to analyse interactions between the variables, multivariate statistical analyses were employed. The patients with untreated hypertension had higher [Ca2+]i values in non-activated platelets (P = 0.04) and lower levels of plasma ionized calcium (P = 0.02) than the reference subjects. In multivariate models analysing platelet [Ca2+]i mean blood pressure (MBP), plasma ionized calcium, serum parathyroid hormone (PTH) and body mass index (BMI), the relationship between platelet [Ca2+]i and blood pressure was attenuated (P = 0.13), whereas the inverse relationships between plasma ionized calcium and MBP (P = 0.01) and between platelet [Ca2+]i and serum PTH (P = 0.06) seen in univariate analyses persisted. According to the multivariate models the [Ca2+]i value explained only 5% of the MBP variability. Thus, the data from this investigation do not support a close relationship between basal platelet [Ca2+]i and blood pressure. The inverse relationship between plasma ionized calcium and blood pressure, independent of platelet [Ca2+]i and serum PTH, suggests a direct interaction between plasma ionized calcium and blood pressure regulation.

Adolescent↗

[The stepwise statistical analysis of free behavior in rats].

Logically justified sequential application of a number of statistical methods makes it possible to describe comprehensively free behavior of animals. In particular, only in the free behavior experiments it was shown that some behavioral indices of August rats were less variable than those of Wistar rats. Moreover, the results of statistical analysis suggest that August rats are less resistant to stress than Wistar. Additional studies had to be conducted earlier in order to draw the analogous conclusions. On the basis of application of correlation and multivariate statistical analysis it was shown that the horizontal motor activity can be the most important characteristic describing the free behavior due to its maximal informativity.

Animals↗

Expression of p53 and bcl-2 in primary locally confined renal cell carcinomas: no evidence for prognostic significance.

p53 tumor suppressor gene mutations are present in a wide variety of human cancers, and the bcl-2 gene product is considered to prevent apoptosis. However, the significance of these gene products for the aggressiveness of the tumor and correspondingly for the prognosis of patients with renal cell carcinoma (RCC) is unclear. The expression of p53 and bcl-2 gene products was studied immunohistochemically using formalin-fixed paraffinembedded tumor samples of 31 locally confined RCC of patients treated with radical nephrectomy. The significance of these 2 parameters, in addition to tumor stage and malignancy grade, was tested with regard to survival and time of no recurrence using Kaplan-Meier-plots by the log rank test or Tarone's test and the Cox multiple hazard regression analysis (mean follow-up 5.4 years). Only 5 of the 31 RCCs stained positively for p53 and only 2 showed positive bcl-2 staining of tumor cells. Tumor stage (P < 0.002) and malignancy grade (P < 0.007) were statistically significant prognostic parameters for both survival and disease-free period by univariate analysis. In contrast, the detection of either p53 (P > 0.67) or bcl-2 gene product (P > 0.28) had no prognostic impact. Also in the multivariate statistical analysis, neither of the 2 parameters i.e. p53 and bcl-2 expression significantly improved the prognostic impact of the conventional prognosticators stage and grade, if applied in addition. The expression of p53 and bcl-2 seems unimportant as a prognostic factor in locally confined RCCs.

Adult↗

A chemometric approach to the detection of milk adulteration based on protein profiles determined by capillary electrophoresis.

The general objective of this study was to utilize chemometrics in the interpretation of capillary electrophoresis milk protein profiles, for the detection of pasteurized milk adulteration with rehydrated milk powder or a rehydrated dairy-based milk substitute. The specific objectives were 1) to collect quantitative data on major casein and whey proteins in authentic and adulterated milks in a single CE analysis; and 2) to apply a pattern recognition procedure, Soft Independent Modeling of Class Analogies (SIMCA), on collected CE protein data, for the development of a statistical model useful in the detection of pasteurized milk adulteration. Authentic samples were fresh milk collected from various farms over a period of six months. Adulterated samples were authentic fresh milk partially or totally substituted with rehydrated milk powder or a rehydrated commercial milk substitute at different levels. Quantitative protein data obtained by capillary free zone electrophoresis for beta-lactoglobulin, alpha-lactalbumin, beta-casein, and alpha-casein of 86 samples, authentic and adulterated samples, were used as a training set to build a SIMCA multivariate statistical model. The detection of sample outliers was useful for the elimination of unusual samples and optimization of the multivariate model. From the 35 commercial pasteurized milks tested, which were treated as unknowns, a total of 14 samples (40%) were not assigned to the authentic or fresh milk group, meaning that these samples had some type of adulteration at the levels included in the training set (> 15%). Decision-making on detecting adulteration of unknown commercial pasteurized milk samples was eased since predictions were based on statistical probabilities.

Animals↗

Predictors of transcutaneous oxygen tension in the lower limbs of diabetic subjects.

We examined neuropathy, ankle pressure index (API), and other factors as predictors of transcutaneous oxygen (TcPO2) in the lower limbs of 657 diabetic subjects. Eligible subjects underwent a clinical assessment that included three standard measures of autonomic neuropathy. TcPO2 measurements were performed at 37 degrees C and 44 degrees C at four lower limb locations. Associations between potential predictors and TcPO2 were tested using univariate and multivariate statistics. Mean TcPO2 at any site did not differ by presence of autonomic neuropathy at either temperature, except for a significantly lower value at 44 degrees C below the knee (56.5 versus 59.2 mmHg, p = 0.021). In multivariate analysis, autonomic neuropathy was significantly and independently related to leg 44 degrees C TcPO2 only (coefficient = -2.6734, p = 0.0182). Much stronger associations were seen between TcPO2 and age, ankle blood pressure, and relative body weight on the plantar foot; and between API, glycosylated haemoglobin, ankle blood pressure, and pedal oedema on the dorsal foot and leg. We conclude that factors related to lower limb TcPO2 vary depending on measurement site. Autonomic neuropathy is not an important determinant of TcPO2 in the feet of diabetic subjects. Although several predictors of TcPO2 were identified, most of the variance of this measurement remains unexplained.

Algorithms↗

Prognostic factors in early invasive carcinoma of the uterine cervix. A clinical, histopathologic, and statistical analysis of 343 cases.

Three hundred forty-three cases of early invasive carcinoma of the uterine cervix were analyzed. Depth of infiltration and lateral extension of the tumors varied between 0.2 and 9.0 mm and 0.4 and 17.2 mm and were found to be of no prognostic importance. A multivariate statistical analysis revealed several risk factors: an epithelialized portio at diagnosis, suspected or evident tumor invasion in capillary-like spaces, absence of tumor-free margins at conization, and treatment with conization or simple hysterectomy, especially in the presence of the above-mentioned factors. Tumors of a large cell keratinizing type were prognostically favorable. Confluent growth was not associated with increased risk of recurrence or death from recurrence. Cervical biopsies proved to be inadequate as diagnostic material. Recurrences were almost always local and developed late. In absence of risk factors conization is sufficient treatment. In the presence of risk factors treatment should be the same as that performed for frankly invasive Stage IB carcinomas.

Carcinoma↗

Analysis of outcome and complications in 400 cases of microvascular head and neck reconstruction.

OBJECTIVE: To determine the incidence and causes of perioperative complications in patients who undergo microvascular free flap procedures for reconstruction of the head and neck. SETTING: Academic tertiary care medical center. PATIENTS AND METHODS: A total of 400 consecutive microvascular free flap procedures were performed for reconstruction of the head and neck, with 95% of the defects arising after the treatment of malignancies. Flap donor sites included radial forearm (n = 183), fibula (n = 145), rectus abdominis (n = 38), subscapular system (n = 28), iliac crest (n = 5), and a jejunal flap. Patient-related characteristics (age; sex; diagnosis; comorbidity level; tumor stage; defect site; primary vs secondary reconstruction; and history of surgery, radiation therapy, or chemotherapy) and the incidence of perioperative complications were recorded prospectively over a 7-year period. RESULTS: The perioperative mortality was 1.3%. Overall, perioperative complications occurred in 36.1% of all cases. Free flaps proved to be extremely reliable, with a 0.8% incidence of free flap failure and a 3% incidence of partial flap necrosis. Perioperative medical complications occurred in 20.5% of cases, with pulmonary, cardiac, and infectious complications predominating. Multivariate statistical analysis showed significant relationships between the incidence of perioperative complications and preoperative comorbidity level as indicated by American Society of Anesthesiologists (ASA) status (P =.02). CONCLUSIONS: The present study confirms that free flaps are extremely reliable in achieving successful reconstruction of the head and neck. The incidence of perioperative complications is related to preoperative comorbidity level.

Adolescent↗

Head and facial anthropometry of mixed-race US Army male soldiers for military design and sizing: a pilot study.

In the United States, the biologically admixed population is increasing. Such demographic changes may affect the distribution of anthropometric characteristics, which are incorporated into the design of equipment and clothing for the US Army and other large organizations. The purpose of this study was to examine multivariate craniofacial anthropometric distributions between biologically admixed male populations and single racial groups of Black and White males. Multivariate statistical results suggested that nose breadth and lip length were different between Blacks and Whites. Such differences may be considered for adjustments to respirators and chemical-biological protective masks. However, based on this pilot study, multivariate anthropometric distributions of admixed individuals were within the distributions of single racial groups. Based on the sample reported, sizing and designing for the admixed groups are not necessary if anthropometric distributions of single racial groups comprising admixed groups are known.

Adult↗

Intelligent real-time performance monitoring and quality prediction for batch/fed-batch cultivations.

Supervision of batch bioprocess operations in real-time during the progress of a batch run offers many advantages over end-of-batch quality control. Multivariate statistical techniques such as multiway partial least squares (MPLS) provide an efficient modeling and supervision framework. A new type of MPLS modeling technique that is especially suitable for online real-time process monitoring and the multivariate monitoring charts are presented. This online process monitoring technique is also extended to include predictions of end-of-batch quality measurements during the progress of a batch run. Process monitoring, quality estimation and fault diagnosis activities are automated and supervised by embedding them into a real-time knowledge-based system (RTKBS). Interpretation of multivariate charts is also automated through a generic rule-base for efficient alarm handling. The integrated RTKBS and the implementation of MPLS-based process monitoring and quality control are illustrated using a fed-batch penicillin production benchmark process simulator.

Algorithms↗

Mitotic counts in follicular lymphomas.

Increased mitotic activity is associated with significantly shorter patient survival in some of the subtypes of diffuse non-Hodgkin's lymphomas. This study on 105 cases of follicular lymphoma was undertaken to determine the clinical significance of mitotic activity in follicular lymphomas. For each case, two pathologists independently counted mitotic figures in 20 random high-power fields. The difference of the average mitotic counts over 20 high-power fields for the two pathologists showed a Gaussian distribution with a median difference of -0.26 counts per high-power field and a standard error of 0.10 per cent. In 94 cases (91 per cent), the difference was less than two mitotic counts per high-power field, indicating good interobserver agreement. There was a statistically significant difference in mitotic counts between subtypes of follicular lymphoma as well as a gradient among subtypes, with the lowest mitotic activity in the poorly differentiated lymphocytic subtype and highest in the undifferentiated subtype. A multivariate statistical analysis of clinical and pathologic variables showed that mitotic figures were not of prognostic significance.

Humans↗

Very low-dose adjuvant chemotherapy in steroid receptor negative stage I breast cancer patients. Austrian Breast Cancer Study Group.

A randomised clinical trial was performed to test whether or not low-dose chemotherapy lasting only 35 days improves the outcome of breast cancer patients with stage I disease and negative oestrogen and progesterone receptors (ER-, PgR-). Between 1984 and 1990, 277 stage I breast cancer patients with tumours negative for both oestrogen and progesterone receptors were randomised to receive either low-dose short-term chemotherapy or no chemotherapy. Chemotherapy consisted of one cycle of doxorubicin, vincristin (AV) and one cycle of cyclophosphamide, methotrexate, fluorouracil (CMF). Patients were stratified for tumour stage, type of surgery, menopausal status and participating centre. Results were analysed both by univariate and multivariate statistical. After a median length of follow-up of 84 months, disease-free (DFS) and overall survival (OS) did not differ significantly between patients having received adjuvant chemotherapy and the control group. Uni- and multivariate analysis did not show any significant prognostic or therapy related factor. A low-dose short-term adjuvant chemotherapy is insufficient to improve the prognosis of patients with breast cancer stage I with ER-, PgR-tumours.

Adult↗