Logistic regression scoring index for detection of interstitial lung disease
Document Type
Article
Publication Date
1-1-2019
Abstract
The decision for a lung transplant requires the knowledge of the severity of interstitial lung disease (ILD). A manual scoring sheet was developed for discriminating ILD cases from non-ILD cases based on eleven indicators selected by the radiologist. The manual scoring sheet requires the radiologists to score the severity of general ILD from visual inspection on the changes of lung tissue using the high resolution computed tomography images. This paper investigates the use of logistic regression in developing a scoring index, ϕ (x), to investigate the presence of ILD. The result shows that the scoring index, ϕ (x), is robust and has discriminatory potentials. The threshold (c, d) having values (0.7, 0.24) allowed correct classifications of 80% of ILD cases and 61% of non-ILD cases in a total sample of 134 cases. This study suggests that ϕ (x) may be used as a simple and practical scoring index in initial investigations for ILD detection. © 2006-2019 Asian Research Publishing Network (ARPN).
Keywords
logistic regression, scoring index, interstitial lung disease, detection, confidence probability
Divisions
MathematicalSciences
Funders
Ministry of Education Malaysia,Universiti Teknologi Malaysia,University of Malaya
Publication Title
ARPN Journal of Engineering and Applied Sciences
Volume
14
Issue
5
Publisher
Asian Research Publishing Network