Dimension reduction-based penalized logistic regression for cancer classification using microarray data

The use of penalized logistic regression for cancer classification using microarray expression data is presented. Two dimension reduction methods are respectively combined with the penalized logistic regression so that both the classification accuracy and computational speed are enhanced. Two other machine-learning methods, support vector machines and least-squares regression, have been chosen for comparison. It is shown that our methods have achieved at least equal or better results. They also have the advantage that the output probability can be explicitly given and the regression coefficients are easier to interpret. Several other aspects, such as the selection of penalty parameters and components, pertinent to the application of our methods for cancer classification are also discussed.     

Tobacco and myocardial infarction: is snuff less dangerous than cigarettes?

OBJECTIVE--To estimate the risk of myocardial infarction in snuff users, cigarette smokers, and non-tobacco users in northern Sweden, where using snuff is traditional. DESIGN--Case-control study. SETTING--Northern Sweden. SUBJECTS--All 35-64 year old men who had had a first myocardial infarction and a population based sample of 35-64 year old men who had not had an infarction in the same geographical area. MAIN OUTCOME MEASURE--Tobacco consumption (regular snuff dipping, regular cigarette smoking, non-tobacco use) and risk of acute myocardial infarction. RESULTS--59 of 585 (10%) patients who had a first myocardial infarction and 87 of 589 (15%) randomly selected men without myocardial infarction were non-smokers who used snuff daily. The age adjusted odds ratio for myocardial infarction was 0.89 (95% confidence interval 0.62 to 1.29) for exposure to snuff and 1.87 (1.40 to 2.48) for cigarette smoking compared with non-tobacco users, showing an increased risk in smokers but not in snuff dippers. Regular cigarette smokers had a significantly higher risk of myocardial infarction than regular snuff dippers (age adjusted odds ratio 2.09; 1.39 to 3.15). Smoking, but not snuff dipping, predicted myocardial infarction in a multiple logistic regression model that included age and level of education. CONCLUSIONS--In middle aged men snuff dipping is associated with a lower risk of myocardial infarction than cigarette smoking.     

Third generation oral contraceptives and risk of myocardial infarction: an international case-control study. Transnational Research Group on Oral Contraceptives and the Health of Young Women.

OBJECTIVE--To test whether use of combined oral contraceptives containing third generation progestogens is associated with altered risk of myocardial infarction. DESIGN--Matched case-control study. SETTING--16 centres in Austria, France, Germany, Switzerland, and the United Kingdom. SUBJECTS--Cases were 153 women aged 16-44 with a myocardial infarction event. Controls were 498 women (at least 3 controls per case) unaffected by myocardial infarction who were matched with their corresponding case for age and for hospital or community setting within four months of the index infarction. MAIN OUTCOME MEASURES--Odds ratios derived with stratified analyses and unconditional logistic regression to adjust for potential confounding variables. RESULTS--The estimated odds ratio for myocardial infarction of third compared with second generation oral contraceptives among all 651 study subjects was 0.36 (95% confidence interval 0.1 to 1.2) (P = 0.11). The odds ratio for the United Kingdom and Germany alone was 0.45 (0.1 to 1.8) (P = 0.26). Other odds ratios for the five countries were 3.1 (1.5 to 6.3) (P = 0.003) for use of second generation products v no current use and 1.1 (0.4 to 3.4) (P = 0.9) for use of third generation products v no current use. Among the confounding variables the independent contribution of smoking (for which adjustment was made in the above estimates) proved to be important (10.1 (5.7 to 17.9), P < 0.001). CONCLUSION--An odds ratio of 0.45 with wide confidence intervals shows that third generation oral contraceptives compared with second generation products are associated with a reduced risk of myocardial infarction or with no difference. This finding from an interim analysis should be interpreted with extreme caution. However, the excess risk of venous thromboembolism associated with the use of third generation products may be balanced by the reduced risk of myocardial infarction associated with the same products.     

Regression trees for predicting mortality in patients with cardiovascular disease: What improvement is achieved by using ensemble‐based methods?

In biomedical research, the logistic regression model is the most commonly used method for predicting the probability of a binary outcome. While many clinical researchers have expressed an enthusiasm for regression trees, this method may have limited accuracy for predicting health outcomes. We aimed to evaluate the improvement that is achieved by using ensemble‐based methods, including bootstrap aggregation (bagging) of regression trees, random forests, and boosted regression trees. We analyzed 30‐day mortality in two large cohorts of patients hospitalized with either acute myocardial infarction (N = 16,230) or congestive heart failure (N = 15,848) in two distinct eras (1999–2001 and 2004–2005). We found that both the in‐sample and out‐of‐sample prediction of ensemble methods offered substantial improvement in predicting cardiovascular mortality compared to conventional regression trees. However, conventional logistic regression models that incorporated restricted cubic smoothing splines had even better performance. We conclude that ensemble methods from the data mining and machine learning literature increase the predictive performance of regression trees, but may not lead to clear advantages over conventional logistic regression models for predicting short‐term mortality in population‐based samples of subjects with cardiovascular disease.     

Reduced risk of death at 28 days in patients taking a beta blocker before admission to hospital with myocardial infarction.

OBJECTIVE--To see whether patients taking an oral beta blocker at the time of admission to hospital with myocardial infarction have a reduced risk of death at 28 days. DESIGN--Retrospective analysis of data collected on patients admitted over four years. SETTING--Community based study. PATIENTS--2430 Consecutive patients living in the Perth statistical division admitted to hospital with myocardial infarction during 1984-7. MAIN OUTCOME MEASURE--Survival at 28 days among patients taking a beta blocker at onset of myocardial infarction. RESULTS--Patients were grouped into those who were and were not taking a beta blocker at the time of admission. Though patients taking a beta blocker were older and more likely to have a history of myocardial infarction, angina, or hypertension, the overall mortality at 28 days was similar in the two groups. A logistic regression model used to adjust for factors predictive of cardiac death at 28 days confirmed that patients taking a beta blocker at the time of admission had a significantly reduced risk of death (relative risk 0.50; 95% confidence interval 0.34 to 0.76). Though the incidence of fatal ventricular fibrillation was similar in the two groups, mean peak creatine kinase activity was significantly lower in the beta blocker group. CONCLUSIONS--These data support the value of long term use of beta blockers in patients at risk of myocardial infarction. They suggest that patients taking these agents before admission to hospital with myocardial infarction have a significant survival advantage at 28 days, which may be due to a reduction in infarct size.     

Regularized logistic regression and multiobjective variable selection for classifying MEG data

This paper addresses the question of maximizing classifier accuracy for classifying task-related mental activity from Magnetoencelophalography (MEG) data. We propose the use of different sources of information and introduce an automatic channel selection procedure. To determine an informative set of channels, our approach combines a variety of machine learning algorithms: feature subset selection methods, classifiers based on regularized logistic regression, information fusion, and multiobjective optimization based on probabilistic modeling of the search space. The experimental results show that our proposal is able to improve classification accuracy compared to approaches whose classifiers use only one type of MEG information or for which the set of channels is fixed a priori.     

纤维蛋白原与冠心病PCI介入治疗围术期心肌梗死的相关性

目的:探讨纤维蛋白原与冠心病介入治疗围术期心肌梗死的相关性。方法:2013年1月到2015年1月,选择在我院进行诊治的冠心病患者92例,都给予PCI介入手术治疗,在手术前后进行纤维蛋白原与心功能的测定,对围术期心肌梗死发生情况与临床资料进行调查与分析。结果:所有患者都介入手术治疗成功,术后LVESVI与LVEDVI值都明显低于术前(P0.05),而术后LVEF值明显高于术前(P0.05);术后患者的血浆纤维蛋白原值为3.66±0.42 g/L,明显低于术前的7.45±0.56 g/L(P0.05)。围手术期发生心肌梗死8例,发生率为8.7%。Spearman秩相关分析法结果显示心肌梗死发病与血浆纤维蛋白原、LVESVI、LVEDVI、LVEF值都存在明显相关性(P0.05),多元Logistic回归分析结果显示纤维蛋白原、LVESVI、LVEDVI、LVEF、年龄为导致冠心病围术期心肌梗死的主要危险因素(P0.05)。结论:介入手术治疗冠心病具有很好的效果,但是围术期心肌梗死的发生率比较高,纤维蛋白原能有效反应病变状况,在心肌梗死的发生发展中起着关键性作用。     

Initial Heparin Therapy in Acute Myocardial Infarction

The administration of heparin during the first 48 hours following acute myocardial infarction is widely practised. Heparin treatment is also recommended for acute coronary insufficiency on the grounds that it may prevent development of an impending myocardial infarction. These measures had been accepted without support of a controlled clinical trial. By random selection, 101 patients hospitalized with a provisional diagnosis of acute myocardial infarction received heparin (100 mg. intravenously every eight hours for 48 hours) and 105 patients were assigned to a control group. Both groups of patients received bishydroxycoumarin (Dicumarol). The mortality in the heparin series was 30% and in the control group, 28%. A significantly large number of the heparin-treated patients developed clinical and laboratory proof of recent myocardial infarction. It is concluded that early intermittent intravenous heparin treatment does not lower the mortality in patients with acute myocardial infarction nor does it prevent impending myocardial infarction in patients with acute coronary insufficiency.     

An application of conditional logistic regression and multifactor dimensionality reduction for detecting gene-gene Interactions on risk of myocardial infarction: The importance of model validation

Background  

To examine interactions among the angiotensin converting enzyme (ACE) insertion/deletion, plasminogen activator inhibitor-1 (PAI-1) 4G/5G, and tissue plasminogen activator (t-PA) insertion/deletion gene polymorphisms on risk of myocardial infarction using data from 343 matched case-control pairs from the Physicians Health Study. We examined the data using both conditional logistic regression and the multifactor dimensionality reduction (MDR) method. One advantage of the MDR method is that it provides an internal prediction error for validation. We summarize our use of this internal prediction error for model validation.     

Leukemia Prediction Using Sparse Logistic Regression

We describe a supervised prediction method for diagnosis of acute myeloid leukemia (AML) from patient samples based on flow cytometry measurements. We use a data driven approach with machine learning methods to train a computational model that takes in flow cytometry measurements from a single patient and gives a confidence score of the patient being AML-positive. Our solution is based on an regularized logistic regression model that aggregates AML test statistics calculated from individual test tubes with different cell populations and fluorescent markers. The model construction is entirely data driven and no prior biological knowledge is used. The described solution scored a 100% classification accuracy in the DREAM6/FlowCAP2 Molecular Classification of Acute Myeloid Leukaemia Challenge against a golden standard consisting of 20 AML-positive and 160 healthy patients. Here we perform a more extensive validation of the prediction model performance and further improve and simplify our original method showing that statistically equal results can be obtained by using simple average marker intensities as features in the logistic regression model. In addition to the logistic regression based model, we also present other classification models and compare their performance quantitatively. The key benefit in our prediction method compared to other solutions with similar performance is that our model only uses a small fraction of the flow cytometry measurements making our solution highly economical.     

Acute Myocardial Infarction: Clinical Application of Technetium 99m Stannous Pyrophosphate Infarct Scintigraphy

Acute myocardial infarction is being recognized as a spectrum of clinical subsets. This appreciation has been brought about to a large degree by the development of several new tools that can be applied clinically to aid in evaluation of patients with acute infarction, and in some cases to provide short and long-term prognostic information. In the realm of noninvasive methods, several tests utilizing radiopharmaceuticals and scintillation cameras have emerged and are rapidly becoming reliable diagnostic parameters in patients with coronary disease and infarction. Technetium 99m (stannous) pyrophosphate (TcPYP) scintigraphy, one of the first of these techniques to find clinical use, has been shown to be an accurate indicator of acute transmural myocardial infarction and provides added sensitivity and specificity to the diagnosis. Increased diagnostic accuracy, the dimension of visible localization and the potential for infarct sizing promise physicians better understanding of a patient''s clinical presentation and a more rational approach to management.     

Inference methods for the conditional logistic regression model with longitudinal data

This paper considers inference methods for case-control logistic regression in longitudinal setups. The motivation is provided by an analysis of plains bison spatial location as a function of habitat heterogeneity. The sampling is done according to a longitudinal matched case-control design in which, at certain time points, exactly one case, the actual location of an animal, is matched to a number of controls, the alternative locations that could have been reached. We develop inference methods for the conditional logistic regression model in this setup, which can be formulated within a generalized estimating equation (GEE) framework. This permits the use of statistical techniques developed for GEE-based inference, such as robust variance estimators and model selection criteria adapted for non-independent data. The performance of the methods is investigated in a simulation study and illustrated with the bison data analysis.     

眼科以咽部瘙痒为首发症状的心肌梗死1 例- 附文献复习

目的:提高异位疼痛为首发症状心肌梗死的临床警惕性与确诊率。方法:通过我院不典型心肌梗死病例1例就诊过程进行回顾性追溯分析,进行相关文献复习,总结提高诊断准确性,减少误诊的经验方法。结果:因诊断及时,患者经积极的抢救治疗,症状改善,治愈出院。结论:提高对异位疼痛为首发症状心肌梗死的警惕,拓宽诊断思维,全面体检,尤其注意心电图:动态心电图、心肌酶谱的监测,及时确诊,以尽量减少不典型心肌梗死的误诊。     

D- 二聚体在急性心肌梗死诊断中的应用研究

目的:探讨D-二聚体在急性心肌梗死诊断中的应用价值,为提高急性心肌梗死的诊断准确性提供有益的参考.方法:以我院2010年1月至2011年1月收治的急性心肌梗死患者34例为实验组,以我院同期健康体检的34例为对照组,对两组的基础水平以及实验组溶栓治疗各时间点的D-二聚体水平进行检测分析.结果:在基础状态下实验组的D-二聚体为(1.59±0.31)mg/L,显著高于对照组(P＜0.05).实验组患者在溶栓治疗6h后的D-二聚体水平为(3.48± 1.40)mg/L,显著高于溶栓治疗前(P＜0.05).其余各时间点的测量值差异无统计学意义(P＞0.05).结论:D-二聚体检测有利于急性心肌梗死患者的诊断,对临床治疗和预后有较好的指导意义,是较好的心肌梗死诊断分子标志物.在发现D-二聚体水平异常时,应结合临床症状、心电图、胸部X光片的结果进行综合判断,确保诊断的可靠性.     

The association between phenomena on the Sun, geomagnetic activity, meteorological variables, and cardiovascular characteristic of patients with myocardial infarction

It has been found that solar and geomagnetic activity affects the cardiovascular system. Some evidence has been reported on the increase in the rate of myocardial infarction, stroke and myocardial infarction related deaths during geomagnetic storms. We investigated the association between cardiovascular characteristics of patients, admitted for myocardial infarction with ST elevation (STEMI), and geomagnetic activity (GMA), solar proton events (SPE), solar flares, and meteorological variables during admission. The data of 1,979 patients hospitalized at the Hospital of Lithuanian University of Health Sciences (Kaunas) were analyzed. We evaluated the association between environmental variables and patient’s characteristics by multivariate logistic regression, controlling patient’s gender and age. Two days after geomagnetic storms the risk of STEMI was over 1.5 times increased in patients who had a medical history of myocardial infarction, stable angina, renal or pulmonary diseases. The dose–response association between GMA level and STEMI risk for patients with renal diseases in history was observed. Two days after SPE the risk of STEMI in patients with stable angina in anamnesis was increased over 1.5 times, adjusting by GMA level. The SPE were associated with an increase of risk for patients with renal diseases in history. This study confirms the strongest effect of phenomena in the Sun in high risk patients.     

Effect of TNF-α and IL-1β genetic variants on the development of myocardial infarction in Turkish population

Tumor necrosis factor-alpha (TNF-α) and interleukin 1 beta (IL-1β) genetic variants which resulting in TNF-α and IL-1 overproduction may increase susceptibility to autoimmune diseases such as atherosclerosis. We have studied the association of TNF-α G308A and IL-1β (+3953) C/T polymorphism with myocardial infarction in Turkish population. 143 patients with myocardial infarction and 213 age-matched healthy controls were included in the study. In univariant analysis, the frequencies of IL-1β, TNF-α genotype or allele, and haplotype of C:A and T:A were significantly elevated in patients when compared with those of controls. GA genotype of TNF-α, T allele of IL-1β and A of TNF-α allele seem to be risk factors for myocardial infarction. In contrast, CC genotype of IL-1β and GG genotype of TNF-α have protective effect against myocardial infarction. In multivariate logistic regression analysis, TNF-α A allele, gender and smoking were associated with myocardial infarction. In conclusion, we can state that TNF-α A allele might be associated with myocardial infarction.     

Sparse learning and stability selection for predicting MCI to AD conversion using baseline ADNI data

ABSTRACT: BACKGROUND: Patients with Mild Cognitive Impairment (MCI) are at high risk of progression to Alzheimer's dementia. Identifying MCI individuals with high likelihood of conversion to dementia and the associated biosignatures has recently received increasing attention in AD research. Different biosignatures for AD (neuroimaging, demographic, genetic and cognitive measures) may contain complementary information for diagnosis and prognosis of AD. METHODS: We have conducted a comprehensive study using a large number of samples from the Alzheimer's Disease Neuroimaging Initiative (ADNI) to test the power of integrating various baseline data for predicting the conversion from MCI to probable AD and identifying a small subset of biosignatures for the prediction and assess the relative importance of different modalities in predicting MCI to AD conversion. We have employed sparse logistic regression with stability selection for the integration and selection of potential predictors. Our study differs from many of the other ones in three important respects: (1) we use a large cohort of MCI samples that are unbiased with respect to age or education status between case and controls (2) we integrate and test various types of baseline data available in ADNI including MRI, demographic, genetic and cognitive measures and (3) we apply sparse logistic regression with stability selection to ADNI data for robust feature selection. RESULTS: We have used 319 MCI subjects from ADNI that had MRI measurements at the baseline and passed quality control, including 177 MCI Non-converters and 142 MCI Converters. Conversion was considered over the course of a 4-year follow-up period. A combination of 15 features (predictors) including those from MRI scans, APOE genotyping, and cognitive measures achieves the best prediction with an AUC score of 0.8587. These results also demonstrate the effectiveness of stability selection for feature selection in the context of sparse logistic regression.     

THE Q WAVE IN L III AND aVF OF THE ELECTROCARDIOGRAM—A Vectorcardiographic Analysis With the Use of the Frank System

Seventy-three consecutive patients with a Q wave in Lead III and aVF in the electrocardiogram were studied. Vectorcardiograms were recorded with the use of the Frank system.In 32 cases the ECG''s were compatible with the diagnosis of an inferior myocardial infarction based on a Q wave in Lead III and/or aVF greater than 0.04 second duration and greater than 25 per cent of the amplitude of the R wave. In this group, there were 16 patients with coronary disease and the VCG confirmed the electrocardiographic diagnosis of an infarction in 14 cases. In 13 of the other 16 cases without history of coronary disease the VCG did not suggest the presence of an infarction.In all 17 cases with questionable electrocardiographic diagnosis of an inferior infarction, and without history of coronary disease, the VCG denied the presence of an infarction. In 18 cases with small Q III or Q aVF the VCG''s were within normal limits. In two cases with normal Q III and Q aVF the VCG''s did not detect the presence of an infarction in both cases.The vectorcardiographic diagnosis of an inferior myocardial infarction was based on the superior orientation (at or above 360 degrees) of the 10, 20, 25 and 30-msec vectors in the frontal plane, superior displacement of the maximum QRS vector and clockwise rotation. In the left sagittal plane the 10, 20, 25 and 30-msec vectors were oriented at or above 180 degrees with the loop rotating counterclockwise.The data presented suggest that vectorcardiography is a useful adjunct to electrocardiography in the diagnosis of an inferior myocardial infarction.     

The morphofunctional state of the neurons in the nodose ganglia of the normal vagus nerves and in experimental myocardial infarct

Basing on a complex approach with the use of general morphological and morphometrical methods, a morphofunctional characteristics of the nodose ganglia neurons of the nervus vagus has been presented in 203 dogs, both normal and at an experimental myocardial infarction. The neurons in question react to the myocardial infarction with a complex of changes in the intracellular structure parameters and their chemical markers. Their manifestation degree is higher and lasts longer, when the myocardial infarction happens at day time.