Search or add a thesis

Advanced Search (Beta)
Home > Prevalence and Risk Factors Associated With Preterm Birth in Pakistan: Meta-Analysis and Data Mining Approach

Prevalence and Risk Factors Associated With Preterm Birth in Pakistan: Meta-Analysis and Data Mining Approach

Thesis Info

Access Option

External Link

Author

Hanif, Asif

Program

PhD

Institute

Hajvery University

City

Lahore

Province

Punjab

Country

Pakistan

Thesis Completing Year

2018

Thesis Completion Status

Completed

Subject

Statistics

Language

English

Link

http://prr.hec.gov.pk/jspui/bitstream/123456789/14577/1/asif%20hanif%20statitics%202018%20hajvery%20uni%20lhr%20prr.docx

Added

2021-02-17 19:49:13

Modified

2024-03-24 20:25:49

ARI ID

1676727005065

Similar


Objectives: The current study was planned with 4 main objectives; the first objective was to find prevalence of PTB in Pakistan (by taking primary data) and to find the global pooled prevalence (using Meta-analysis), 2nd objective was to explore risk of PTB in presence of different risk factors like, 3rd objective was to compare number of risk factors by scoring / indexing of factors contributing to PTB and 4th objective was to see combined effect of all risk factors of different domains in terms of number of risk factors. Methodology: Data collected from departments of Gynecology of four different hospitals of major provinces of Pakistan and secondary data of published reaches was utilized for Meta-analysis. A total of 1,691 females were taken in current study with the help of consultant gynecologists or duty staff nurse to note their gestational age after delivery to find prevalence of preterm birth in Pakistan and 46 studies (n = 6,788,033) were used to find pooled prevalence of preterm birth using Meta-analysis technique. To find risk factors of preterm birth two groups were made i.e. preterm birth (served as cases) and full term birth (served as controls) taken from those 1,691 females. The advanced data mining techniques like Binary Logistic Regression (LR), Artificial Neural Networks (ANNs) and Support Vector Machine (SVM) were used to find best predictive accuracy of preterm birth in presence of different factors. Lastly scoring/indexing of different domains was done to know the combined impact of number of factors for preterm birth. Results: According to results of this study prevalence of preterm birth was found as 21.64% females and overall pooled prevalence of preterm birth was found as 11.3%. Using different data mining techniques the heights sensitivity and specificity were provided by Support Vector Machine (SVM) preceded by Artificial Neural Networks (ANNs) and Logistic Regression. Interestingly highest area under receiver operator characteristic curve (ROC) was calculated as 96.95% using ANNs, the area under curve of ROC for Logistic Regression was 93.89% and of SVM was 96.81%. Conclusion: Based on the findings a high prevalence of preterm birth was found in Pakistan and globally using Meta-analysis. Many modifiable and controllable risk factors like body mass index, different curable diseases or conditions during pregnancy and fetal conditions were found.
Loading...
Loading...

Similar News

Loading...

Similar Articles

Loading...

Similar Article Headings

Loading...