A Novel Statistical Approach to Investigate Wisconsin Breast Cancer Data by Generalized Linear Mixed Model Approach in Cancer Epidemiology in Medicine DOI Open Access
Neslihan İyit, Neriman Akdam

Published: Oct. 11, 2023

The main aim of this study is to predict whether the type breast cancer “benign” or “malignant” by classical generalized linear model (GLM) approach and an extended family GLM called mixed (GLMM) for binomially distributed response variable with binary link functions. In study, advanced statistical modeling based on GLMM traditional various functions proposed investigate relationships between “malignant benign diagnosis BC in patients” “nine attributes” 699 diagnosed patients. This also focuses significance accurate classification patients studies medicine WBC dataset. superiority over especially belonging dataset emphasized field medicine. Also importance power IC performance metrics as goodness-of-fit test statistics are strongly inferences from “best” fitted model. findings, best among approaches determined under “logit” function “id” random effect most statistically significant odds occurance being 7.9104, 5.6888, 5.6643, 4.9842, 4.1212, 2.0679, 1.8755, 1.3970 times more than every one-unit increase quantities “clump thickness”, “bland chromatin”, “mitoses”, “bare nuclei”, “cell shape”, “marginal adhesion”, “epithelial cell size”, respectively.

Language: Английский

The global effects of digestive system cancers worldwide on the COVID-19 pandemic by negative binomial (mixed) regression models DOI Creative Commons

Neslihan İyit

Journal of Radiation Research and Applied Sciences, Journal Year: 2024, Volume and Issue: 17(3), P. 100944 - 100944

Published: May 8, 2024

In this study, the main goal is to determine statistically significant relationships between biggest epidemic problem of last years as COVID-19 pandemic and digestive system cancers belonging 168 countries worldwide using generalized linear mixed model (GLMM) its special case (GLM) approaches, obtain global inferences that will shed light on pandemic. For goal, response variable "total cases per 100,000 people" until January 14, 2022. The explanatory variables are total number people suffering from colon rectum, stomach, lip oral cavity, esophageal, nasopharynx 2019, respectively. negative binomial (NB) regression in GLM iteratively reweighted least squares (IRLS) algorithm NB GLMM with "countries" taken "random effects" Adaptive Gauss-Hermite Quadrature (AGHQ) approximation method at 1, 2, 10, 20 quadrature points used for modelling systems cancer data. by information criteria, under log-link function random effects AGHQ point detected most appropriate

Language: Английский

Citations

1
A Novel Statistical Approach to Investigate Wisconsin Breast Cancer Data by Generalized Linear Mixed Model Approach in Cancer Epidemiology in Medicine DOI Open Access
Neslihan İyit, Neriman Akdam

Published: Oct. 11, 2023

The main aim of this study is to predict whether the type breast cancer “benign” or “malignant” by classical generalized linear model (GLM) approach and an extended family GLM called mixed (GLMM) for binomially distributed response variable with binary link functions. In study, advanced statistical modeling based on GLMM traditional various functions proposed investigate relationships between “malignant benign diagnosis BC in patients” “nine attributes” 699 diagnosed patients. This also focuses significance accurate classification patients studies medicine WBC dataset. superiority over especially belonging dataset emphasized field medicine. Also importance power IC performance metrics as goodness-of-fit test statistics are strongly inferences from “best” fitted model. findings, best among approaches determined under “logit” function “id” random effect most statistically significant odds occurance being 7.9104, 5.6888, 5.6643, 4.9842, 4.1212, 2.0679, 1.8755, 1.3970 times more than every one-unit increase quantities “clump thickness”, “bland chromatin”, “mitoses”, “bare nuclei”, “cell shape”, “marginal adhesion”, “epithelial cell size”, respectively.

Language: Английский

Citations

1