Parallel Multiclass Logistic Regression for Classifying Large Scale Image Datasets

Few studies have reported the relationship between knee pain and hypercholesterolemia in the elderly population with osteoarthritis (OA), independent of other variables. The aim of this study was to reveal the association between knee pain and metabolic diseases including hypercholesterolemia using a large-scale cohort. A cross-sectional study was conducted using data from the Korea National Health and the Nutrition Examination Survey (KNHANES-V, VI-1; 2010–2013). Among the subjects aged ≥60 years, 7438 subjects (weighted number estimate = 35,524,307) who replied knee pain item and performed the simple radiographs of knee were enrolled. Using multivariable ordinal logistic regression analysis, variables affecting knee pain were identified, and the odds ratio (OR) was calculated. Of the 35,524,307 subjects, 10,630,836 (29.9%) subjects experienced knee pain. Overall, 20,290,421 subjects (56.3%) had radiographic OA, and 8,119,372 (40.0%) of them complained of knee pain. Multivariable ordinal logistic regression analysis showed that among the metabolic diseases, only hypercholesterolemia was positively correlated with knee pain in the OA group (OR 1.24; 95% Confidence Interval 1.02–1.52, p = 0.033). There were no metabolic diseases correlated with knee pain in the non-OA group. This large-scale study revealed that in the elderly, hypercholesterolemia was positively associated with knee pain independent of body mass index and other metabolic diseases in the OA group, but not in the non-OA group. These results will help in understanding the nature of arthritic pain, and may support the need for exploring the longitudinal associations.

Download Full-text

45 Application of Machine Learning Models to Thermal Burn Patient Outcome Predictions in the Aftermath of a Nuclear Event

Journal of Burn Care & Research ◽

10.1093/jbcr/irab032.049 ◽

2021 ◽

Vol 42 (Supplement_1) ◽

pp. S33-S34

Author(s):

Morgan A Taylor ◽

Randy D Kearns ◽

Jeffrey E Carter ◽

Mark H Ebell ◽

Curt A Harris

Keyword(s):

Machine Learning ◽

Logistic Regression ◽

Length Of Stay ◽

Regression Models ◽

Large Scale ◽

Prediction Models ◽

Burn Patients ◽

Thermal Burn ◽

Logistic Regression Models ◽

Burn Patient

Abstract Introduction A nuclear disaster would generate an unprecedented volume of thermal burn patients from the explosion and subsequent mass fires (Figure 1). Prediction models characterizing outcomes for these patients may better equip healthcare providers and other responders to manage large scale nuclear events. Logistic regression models have traditionally been employed to develop prediction scores for mortality of all burn patients. However, other healthcare disciplines have increasingly transitioned to machine learning (ML) models, which are automatically generated and continually improved, potentially increasing predictive accuracy. Preliminary research suggests ML models can predict burn patient mortality more accurately than commonly used prediction scores. The purpose of this study is to examine the efficacy of various ML methods in assessing thermal burn patient mortality and length of stay in burn centers. Methods This retrospective study identified patients with fire/flame burn etiologies in the National Burn Repository between the years 2009 – 2018. Patients were randomly partitioned into a 67%/33% split for training and validation. A random forest model (RF) and an artificial neural network (ANN) were then constructed for each outcome, mortality and length of stay. These models were then compared to logistic regression models and previously developed prediction tools with similar outcomes using a combination of classification and regression metrics. Results During the study period, 82,404 burn patients with a thermal etiology were identified in the analysis. The ANN models will likely tend to overfit the data, which can be resolved by ending the model training early or adding additional regularization parameters. Further exploration of the advantages and limitations of these models is forthcoming as metric analyses become available. Conclusions In this proof-of-concept study, we anticipate that at least one ML model will predict the targeted outcomes of thermal burn patient mortality and length of stay as judged by the fidelity with which it matches the logistic regression analysis. These advancements can then help disaster preparedness programs consider resource limitations during catastrophic incidents resulting in burn injuries.

Download Full-text

Logistic Regression Bias Correction for Large Scale Data with Rare Events

Advanced Data Mining and Applications - Lecture Notes in Computer Science ◽

10.1007/978-3-642-53917-6_12 ◽

2013 ◽

pp. 133-144 ◽

Cited By ~ 1

Author(s):

Zhen Qiu ◽

Hongyan Li ◽

Hanchen Su ◽

Gaoyan Ou ◽

Tengjiao Wang

Keyword(s):

Logistic Regression ◽

Bias Correction ◽

Large Scale ◽

Rare Events ◽

Large Scale Data ◽

Scale Data

Download Full-text

The Impact of Extraversion and Sensation Seeking on Tourist Role

Social Behavior and Personality An International Journal ◽

10.2224/sbp.2015.43.1.75 ◽

2015 ◽

Vol 43 (1) ◽

pp. 75-84 ◽

Cited By ~ 6

Author(s):

Cheng-Yu Li ◽

Shiao-Yuan Lu ◽

Bi-Kun Tsai ◽

Keh-Yuan Yu

Keyword(s):

Logistic Regression ◽

Regression Analysis ◽

Logistic Regression Analysis ◽

Sensation Seeking ◽

Large Scale ◽

Personality Variables ◽

Tourism Marketing ◽

Large Scale Survey ◽

And Behaviors ◽

The Impact

In recent years, personality variables, such as extraversion and sensation seeking, have been used to investigate tourist preferences and behaviors. For this study, we classified tourist roles into three types: the familiarized mass tourist, the organized mass tourist, and the independent tourist. We investigated the impact of extraversion and sensation seeking on tourist roles in a large-scale survey of Taiwanese citizens (N = 1,249) aged 20 years and older. Using logistic regression analysis, the results indicated that sensation seeking was a significant predictor of tourist role, but extraversion was not. Compared to familiarized mass tourists, people who are sensation-seeking are more likely to become independent tourists rather than organized mass tourists. We provide suggestions for tourism marketing.

Download Full-text

Data-Driven Decision Making and Proactive Citizen–Scientist Communication: A Cross-Sectional Study on COVID-19 Vaccination Adherence

Vaccines ◽

10.3390/vaccines9121384 ◽

2021 ◽

Vol 9 (12) ◽

pp. 1384

Author(s):

Emil Syundyukov ◽

Martins Mednis ◽

Linda Zaharenko ◽

Eva Pildegovica ◽

Ieva Danovska ◽

...

Keyword(s):

Public Health ◽

Logistic Regression ◽

Exploratory Study ◽

Large Scale ◽

Severe Disease ◽

Data Driven ◽

Personal Factors ◽

Cross Sectional ◽

Digital Platform ◽

Factors Associated

Due to the severe impact of COVID-19 on public health, rollout of the vaccines must be large-scale. Current solutions are not intended to promote an active collaboration between communities and public health researchers. We aimed to develop a digital platform for communication between scientists and the general population, and to use it for an exploratory study on factors associated with vaccination readiness. The digital platform was developed in Latvia and was equipped with dynamic consent management. During a period of six weeks 467 participants were enrolled in the population-based cross-sectional exploratory study using this platform. We assessed demographics, COVID-19-related behavioral and personal factors, and reasons for vaccination. Logistic regression models adjusted for the level of education, anxiety, factors affecting the motivation to vaccinate, and risk of infection/severe disease were built to investigate their association with vaccination readiness. In the fully adjusted multiple logistic regression model, factors associated with vaccination readiness were anxiety (odds ratio, OR = 3.09 [95% confidence interval 1.88; 5.09]), feelings of social responsibility (OR = 1.61 [1.16; 2.22]), and trust in pharmaceutical companies (OR = 1.53 [1.03; 2.27]). The assessment of a large number of participants in a six-week period show the potential of a digital platform to create a data-driven dialogue on vaccination readiness.

Download Full-text

FAKTOR-FAKTOR YANG MEMPENGARUHI PEROLEHAN KREDIT OLEH PENGUSAHA MIKRO

Jurnal Ilmiah Ekonomi Bisnis ◽

10.35972/jieb.v4i3.237 ◽

2018 ◽

Vol 4 (3) ◽

Author(s):

Abdul Azis Safii ◽

Tri Suwarno

Keyword(s):

Logistic Regression ◽

Regression Model ◽

Large Scale ◽

Logistic Regression Model ◽

Banking Sector ◽

Binary Logistic Regression ◽

Binary Logistic Regression Model ◽

Access To Credit ◽

The Poor ◽

Micro Enterprises

Abstract: The number of micro-entrepreneurs and the dominant number of micro enterprises compared to medium and large-scale enterprises in Indonesia are not balanced by the provision of access to credit and venture capital for micro businesses. This resulted in a micro-sector sector identical to the poor being vulnerable to exploitation by moneylenders who exploit the difficulties of micro entrepreneurs accessing credit from the banking sector. This study examines the factors that determine the accessibility of credit by micro entrepreneur in Bojonegoro regency. A total sum of 270 micro entrepreneurs who have applied for banking loan were sampled from the study area. With an binary logistic regression model the research resulting that education, skill on entrepreneur, and monthly net profits generated by the microenterprise are significant in determining the accessibility of microcredit. Keywords: micro entrepreneur, microcredit, credit accessibility Abstrak: Perkembangan jumlah pengusaha mikro serta dominannya jumlah usaha mikro dibandingkan dengan usaha menengah dan usaha besar di Indonesia, tidak diimbingi dengan penyediaan akses kredit dan modal usaha bagi para pelaku usaha mikro. Hal tersebut mengakibatkan sektor usaha mikro yang identik dengan masyarakat miskin rentan dieksploitasi oleh rentenir yang memanfaatkan sulitnya para pengusaha mikro mengakses kredit dari sektor perbankan. Penelitian ini menggunakan data primer yang di ambil langsung dari pengusaha mikro dengan teknik kuesioner. Analisis data dengan metode binary logistic regression mendapatkan hasil variabel yang berpengaruh signifikan terhadap akses kredit para pengusaha mikro adalah variabel usia pengusaha, laba bersih usaha tiap bulan, dan jumlah karyawan yang di pekerjakan. Kata kunci : usaha mikro, microcredit, akses kredit

Download Full-text

Abstract 15796: The Inverse Correlation Between Lifetime Occupational Radiation Exposure Duration and Prevalence of Atrial Arrhythmia

Circulation ◽

10.1161/circ.142.suppl_3.15796 ◽

2020 ◽

Vol 142 (Suppl_3) ◽

Author(s):

Rithika Thirumal ◽

Catherine Vanchiere ◽

Ruchi Bhandari ◽

Sania Jiwani ◽

Ronald Horswell ◽

...

Keyword(s):

Logistic Regression ◽

Radiation Exposure ◽

Large Scale ◽

Inverse Correlation ◽

Cardiovascular Complications ◽

Occupational Radiation Exposure ◽

Electronic Survey ◽

Increased Risk ◽

Traditional Risk Factors ◽

Occupational Radiation

Introduction: Fluoroscopy assisted procedures have increased occupational radiation exposure among Cardiologists. Radiation has been linked to cardiovascular complications but its effects on cardiac rhythm has not been extensively explored. Hypothesis: We hypothesized that radiation exposure is associated with increased risk of atrial arrhythmias (AA) despite appropriate leaded body coverage. Methods: Demographic, social, occupational, and medical history was collected from board-certified cardiologists via an electronic survey. Bivariate and multivariable logistic regression analyses were performed. Results: We received 1478 responses from cardiologists; 85.4% were males, 79% were White and 66.1% were ≤65 yrs of age. 35.6% of respondents were interventional cardiologists and 16.4% were electrophysiologists, and of those, 92.2% wore lead apparel during all times of radiation exposure. Cardiologists >50 yrs of age, with >10,000 hours of occupational radiation exposure, had a significantly lower prevalence of AA compared to those with ≤10,000 hours of radiation exposure (11.1% vs 16.7%, p =0.019). A multivariate logistic regression was performed and among cardiologists >50 years of age, exposure to >10,000 radiation hours was significantly associated with lower likelihood of AA, after adjusting for age, sex, DM, HTN and OSA (adjusted OR 0.57; 95% CI 0.38 - 0.85, p =0.007). Traditional risk factors such as age, sex, HTN, DM and OSA were more prevalent in those with AA and cataracts, a well-established complication of radiation exposure in cardiologists, was more prevalent in those exposed to >10,000 radiation hours compared to those exposed to ≤10,000 radiation hours, validating the dependent (AA) and independent variables (radiation exposure), respectively. Conclusions: Radiation exposure in Cardiologists with appropriate lead apparel is inversely related to AA. Large scale prospective studies are needed to validate our findings.

Download Full-text

Prediction of incident myocardial infarction using machine learning applied to harmonized electronic health record data

BMC Medical Informatics and Decision Making ◽

10.1186/s12911-020-01268-x ◽

2020 ◽

Vol 20 (1) ◽

Author(s):

Divneet Mandair ◽

Premanand Tiwari ◽

Steven Simon ◽

Kathryn L. Colborn ◽

Michael A. Rosenberg

Keyword(s):

Neural Network ◽

Machine Learning ◽

Risk Factors ◽

Myocardial Infarction ◽

Logistic Regression ◽

Large Scale ◽

Deep Neural Network ◽

Low Frequency ◽

Common Data Model ◽

Electronic Health Record Data

Abstract Background With cardiovascular disease increasing, substantial research has focused on the development of prediction tools. We compare deep learning and machine learning models to a baseline logistic regression using only ‘known’ risk factors in predicting incident myocardial infarction (MI) from harmonized EHR data. Methods Large-scale case-control study with outcome of 6-month incident MI, conducted using the top 800, from an initial 52 k procedures, diagnoses, and medications within the UCHealth system, harmonized to the Observational Medical Outcomes Partnership common data model, performed on 2.27 million patients. We compared several over- and under- sampling techniques to address the imbalance in the dataset. We compared regularized logistics regression, random forest, boosted gradient machines, and shallow and deep neural networks. A baseline model for comparison was a logistic regression using a limited set of ‘known’ risk factors for MI. Hyper-parameters were identified using 10-fold cross-validation. Results Twenty thousand Five hundred and ninety-one patients were diagnosed with MI compared with 2.25 million who did not. A deep neural network with random undersampling provided superior classification compared with other methods. However, the benefit of the deep neural network was only moderate, showing an F1 Score of 0.092 and AUC of 0.835, compared to a logistic regression model using only ‘known’ risk factors. Calibration for all models was poor despite adequate discrimination, due to overfitting from low frequency of the event of interest. Conclusions Our study suggests that DNN may not offer substantial benefit when trained on harmonized data, compared to traditional methods using established risk factors for MI.

Download Full-text