Using Akaike information criterion and minimum mean square error mode in compensating for ultrasonographic errors for estimation of fetal weight by new operators

Yueh Chin Cheng, Yu Hsien Chiu, Hsien Chang Wang, Fong Ming Chang, Kao Chi Chung, Chiung Hsin Chang, Kuo Sheng Cheng

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Objectives: The accuracy of ultrasound (US) measurements is operator dependent. In order to decrease the operator-dependent errors in estimated fetal weight (EFW), a model selection analysis was undertaken to select significant compensation weighting factors on ultrasonographic parameters to support artificial neural network (ANN), and thus to enhance the accuracy of fetal weight estimation. Materials and Methods: In total, 2127 singletons were examined by prenatal US within 3 days before delivery for ANN development, and another 100 cases were selected from new operators for evaluation. First, correlation analysis was used to analyze the differences between the prenatal and postnatal parameters. Second, Akaike information criterion (AIC) was used to determine the number of database partition and optimal weightings for compensating the input parameters of the ANN model. Finally, minimum mean squared error (MMSE) mode was utilized to determine the optimal EFW. Results: EFW of the proposed compensation model using AIC and MMSE showed mean absolute percent error of 5.1 ± 3.1% and mean absolute error of 158.9 ± 96.2 g. When comparing the accuracy of EFW, our model using AIC and MMSE was superior to those conventional EFW formulas (all p < 0.05). Conclusion: We proved that performing the parameter compensation (by AIC) and model compensations (by MMSE) for the ANN model can improve EFW accuracy. Our AIC-MMSE model of EFW will contribute to the improvement of accuracy when adding new US datasets measured by new operators.

Original languageEnglish
Pages (from-to)46-52
Number of pages7
JournalTaiwanese Journal of Obstetrics and Gynecology
Volume52
Issue number1
DOIs
Publication statusPublished - 2013 Mar 1

Fingerprint

Fetal Weight
Neural Networks (Computer)
Databases

All Science Journal Classification (ASJC) codes

  • Obstetrics and Gynaecology

Cite this

@article{ed535b7f27284d329b1c48b5fbac653a,
title = "Using Akaike information criterion and minimum mean square error mode in compensating for ultrasonographic errors for estimation of fetal weight by new operators",
abstract = "Objectives: The accuracy of ultrasound (US) measurements is operator dependent. In order to decrease the operator-dependent errors in estimated fetal weight (EFW), a model selection analysis was undertaken to select significant compensation weighting factors on ultrasonographic parameters to support artificial neural network (ANN), and thus to enhance the accuracy of fetal weight estimation. Materials and Methods: In total, 2127 singletons were examined by prenatal US within 3 days before delivery for ANN development, and another 100 cases were selected from new operators for evaluation. First, correlation analysis was used to analyze the differences between the prenatal and postnatal parameters. Second, Akaike information criterion (AIC) was used to determine the number of database partition and optimal weightings for compensating the input parameters of the ANN model. Finally, minimum mean squared error (MMSE) mode was utilized to determine the optimal EFW. Results: EFW of the proposed compensation model using AIC and MMSE showed mean absolute percent error of 5.1 ± 3.1{\%} and mean absolute error of 158.9 ± 96.2 g. When comparing the accuracy of EFW, our model using AIC and MMSE was superior to those conventional EFW formulas (all p < 0.05). Conclusion: We proved that performing the parameter compensation (by AIC) and model compensations (by MMSE) for the ANN model can improve EFW accuracy. Our AIC-MMSE model of EFW will contribute to the improvement of accuracy when adding new US datasets measured by new operators.",
author = "Cheng, {Yueh Chin} and Chiu, {Yu Hsien} and Wang, {Hsien Chang} and Chang, {Fong Ming} and Chung, {Kao Chi} and Chang, {Chiung Hsin} and Cheng, {Kuo Sheng}",
year = "2013",
month = "3",
day = "1",
doi = "10.1016/j.tjog.2013.01.008",
language = "English",
volume = "52",
pages = "46--52",
journal = "Taiwanese Journal of Obstetrics and Gynecology",
issn = "1028-4559",
publisher = "Elsevier Taiwan LLC",
number = "1",

}

TY - JOUR

T1 - Using Akaike information criterion and minimum mean square error mode in compensating for ultrasonographic errors for estimation of fetal weight by new operators

AU - Cheng, Yueh Chin

AU - Chiu, Yu Hsien

AU - Wang, Hsien Chang

AU - Chang, Fong Ming

AU - Chung, Kao Chi

AU - Chang, Chiung Hsin

AU - Cheng, Kuo Sheng

PY - 2013/3/1

Y1 - 2013/3/1

N2 - Objectives: The accuracy of ultrasound (US) measurements is operator dependent. In order to decrease the operator-dependent errors in estimated fetal weight (EFW), a model selection analysis was undertaken to select significant compensation weighting factors on ultrasonographic parameters to support artificial neural network (ANN), and thus to enhance the accuracy of fetal weight estimation. Materials and Methods: In total, 2127 singletons were examined by prenatal US within 3 days before delivery for ANN development, and another 100 cases were selected from new operators for evaluation. First, correlation analysis was used to analyze the differences between the prenatal and postnatal parameters. Second, Akaike information criterion (AIC) was used to determine the number of database partition and optimal weightings for compensating the input parameters of the ANN model. Finally, minimum mean squared error (MMSE) mode was utilized to determine the optimal EFW. Results: EFW of the proposed compensation model using AIC and MMSE showed mean absolute percent error of 5.1 ± 3.1% and mean absolute error of 158.9 ± 96.2 g. When comparing the accuracy of EFW, our model using AIC and MMSE was superior to those conventional EFW formulas (all p < 0.05). Conclusion: We proved that performing the parameter compensation (by AIC) and model compensations (by MMSE) for the ANN model can improve EFW accuracy. Our AIC-MMSE model of EFW will contribute to the improvement of accuracy when adding new US datasets measured by new operators.

AB - Objectives: The accuracy of ultrasound (US) measurements is operator dependent. In order to decrease the operator-dependent errors in estimated fetal weight (EFW), a model selection analysis was undertaken to select significant compensation weighting factors on ultrasonographic parameters to support artificial neural network (ANN), and thus to enhance the accuracy of fetal weight estimation. Materials and Methods: In total, 2127 singletons were examined by prenatal US within 3 days before delivery for ANN development, and another 100 cases were selected from new operators for evaluation. First, correlation analysis was used to analyze the differences between the prenatal and postnatal parameters. Second, Akaike information criterion (AIC) was used to determine the number of database partition and optimal weightings for compensating the input parameters of the ANN model. Finally, minimum mean squared error (MMSE) mode was utilized to determine the optimal EFW. Results: EFW of the proposed compensation model using AIC and MMSE showed mean absolute percent error of 5.1 ± 3.1% and mean absolute error of 158.9 ± 96.2 g. When comparing the accuracy of EFW, our model using AIC and MMSE was superior to those conventional EFW formulas (all p < 0.05). Conclusion: We proved that performing the parameter compensation (by AIC) and model compensations (by MMSE) for the ANN model can improve EFW accuracy. Our AIC-MMSE model of EFW will contribute to the improvement of accuracy when adding new US datasets measured by new operators.

UR - http://www.scopus.com/inward/record.url?scp=84878013118&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878013118&partnerID=8YFLogxK

U2 - 10.1016/j.tjog.2013.01.008

DO - 10.1016/j.tjog.2013.01.008

M3 - Article

C2 - 23548217

AN - SCOPUS:84878013118

VL - 52

SP - 46

EP - 52

JO - Taiwanese Journal of Obstetrics and Gynecology

JF - Taiwanese Journal of Obstetrics and Gynecology

SN - 1028-4559

IS - 1

ER -