Volume 18, Number 62 (1-2016)                   jha 2016, 18(62): 42-51 | Back to browse issues page


XML Persian Abstract Print


Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:

Langarizadeh M, Ghazi Saeedi M, Karam Niay Far M, Hoseinpour M. Predicting Premature Birth in Pregnant Women via Assisted Reproductive Technologies using Neural Network. jha. 2016; 18 (62) :42-51
URL: http://jha.iums.ac.ir/article-1-1787-en.html

1- Assistant Professor of Department of Health Information Management, School of Health Management and Information Sciences, Iran University of Medical Sciences, Tehran, Iran
2- Assistant professor of Department of Health Information Management, School of Allied medical sciences, Tehran University of Medical Sciences, Tehran, Iran
3- Obstetrician & Gynecologist, Sarem Cell Researches Center, Sarem Hospital, Tehran, Iran , drkaramnia@yahoo.com
4- MSc. of Medical Informatics, School of Allied medical sciences, Tehran University of Medical Sciences, Tehran, Iran
Abstract:   (811 Views)

Introduction: Nowadays, assisted reproductive technologies are widely used to treat infertility in couples. Studies indicate that the rate of premature birth after using Assisted Reproductive Technologies has been increased as compared to normal pregnancies. The purpose of our study was predicting premature birth in pregnant women via Assisted Reproductive Technologies using artificial neural networks.

Methods: In this retrospective study, initially 45 variables were identified as effective factors for prediction of premature birth in pregnant women via Assisted Reproductive Technologies and data of 130 women were extracted using clinical records in Sarem hospital in Tehran from 1998 to 2014 in October and November, 2014. The most important variables were identified as effective variables using feature selection algorithm and decision tree in SPSS Clementine. Multi-Layer Perceptron network was designed to predict the premature birth in Matlab software. Confusion matrix was used for evaluation in order to calculate accuracy, sensitivity and specificity.

Results: We identified fifteen effective features using feature selection algorithm and decision tree as inputs of the neural networks. Multi-Layer Perceptron network was designed and evaluated. The accuracy, sensitivity and specificity of the test data were 87.2%, 80.0% and 88.2%, respectively and for the total data were 95.4%, 95.0% and 95.5%, respectively.

Conclusion: According to the results, designed neural network for predicting premature birth in pregnant women via Assisted Reproductive Technologies can be helpful in prevention of premature birth complications.

Full-Text [PDF 857 kb]   (563 Downloads)    
Type of Study: Applicable | Subject: General
Received: 2015/01/27 | Accepted: 2015/11/9 | Published: 2015/11/9

Send email to the article author


© 2015 All Rights Reserved | Journal of Health Administration

Designed & Developed by : Yektaweb