Predicting the duration of the first stage of spontaneous labor using a neural network

Lawrence D Devoe, Susan Samuel, Paula Prescott, Bruce A. Work

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

To create a neural network that predicts the length of the first stage of term labor. Two hundred patients with gestations ≥36 weeks, in spontaneous active labor are the study group: 159 for training and 41 for testing; 4 training set patients had second-stage cesarean section for obstructed labor. The network is designed with Brainmaker MacIntosh 1.0 (California Scientific Software). Inputs are uterine activity, estimated fetal weight, position, station, and gestational age; maternal parity, age, height, weight, membrane status, and cervical dilatation. Actual first stages are regressed on those predicted by the network or by a standard partogram set. Differences between actual first stage lengths and those predicted by the neural network or partogram are compared with t-tests; while the proportions of first stages accurately predicted within 1 or 2 h are compared for both methods with chi-square tests. The network trained in 4 h (1388 runs) to a 0.15 tolerance. The network predictions have significantly higher correlation (r = 0.88) than do standard partograms (r = 0.35) with actual first stage durations. Mean differences between predicted and actual first stages are significantly lower for network output than with partograms; these differences increased with first stages exceeding 3 h; 100% of trained network values are within 2 h of actual first stage length. The network performs similarly for a new set of 41 previously unseen labors. This neural network predicts the length of the first stage of spontaneous labor and uses inputs readily available to obstetricians. It outperforms typical partograms for estimating this important feature of normal labor. Future application for intrapartum prognosis could be based on this successful design.

Original languageEnglish (US)
Pages (from-to)256-261
Number of pages6
JournalJournal of Maternal-Fetal and Neonatal Medicine
Volume5
Issue number5
DOIs
StatePublished - Jan 1 1996

Fingerprint

First Labor Stage
Fetal Weight
Maternal Age
Chi-Square Distribution
Parity
Cesarean Section
Gestational Age
Software
Weights and Measures
Pregnancy
Membranes

Keywords

  • Labor
  • Neural network
  • Partogram

ASJC Scopus subject areas

  • Pediatrics, Perinatology, and Child Health
  • Obstetrics and Gynecology

Cite this

Predicting the duration of the first stage of spontaneous labor using a neural network. / Devoe, Lawrence D; Samuel, Susan; Prescott, Paula; Work, Bruce A.

In: Journal of Maternal-Fetal and Neonatal Medicine, Vol. 5, No. 5, 01.01.1996, p. 256-261.

Research output: Contribution to journalArticle

@article{b82078d37be24f64becb06dbdc6b87e4,
title = "Predicting the duration of the first stage of spontaneous labor using a neural network",
abstract = "To create a neural network that predicts the length of the first stage of term labor. Two hundred patients with gestations ≥36 weeks, in spontaneous active labor are the study group: 159 for training and 41 for testing; 4 training set patients had second-stage cesarean section for obstructed labor. The network is designed with Brainmaker MacIntosh 1.0 (California Scientific Software). Inputs are uterine activity, estimated fetal weight, position, station, and gestational age; maternal parity, age, height, weight, membrane status, and cervical dilatation. Actual first stages are regressed on those predicted by the network or by a standard partogram set. Differences between actual first stage lengths and those predicted by the neural network or partogram are compared with t-tests; while the proportions of first stages accurately predicted within 1 or 2 h are compared for both methods with chi-square tests. The network trained in 4 h (1388 runs) to a 0.15 tolerance. The network predictions have significantly higher correlation (r = 0.88) than do standard partograms (r = 0.35) with actual first stage durations. Mean differences between predicted and actual first stages are significantly lower for network output than with partograms; these differences increased with first stages exceeding 3 h; 100{\%} of trained network values are within 2 h of actual first stage length. The network performs similarly for a new set of 41 previously unseen labors. This neural network predicts the length of the first stage of spontaneous labor and uses inputs readily available to obstetricians. It outperforms typical partograms for estimating this important feature of normal labor. Future application for intrapartum prognosis could be based on this successful design.",
keywords = "Labor, Neural network, Partogram",
author = "Devoe, {Lawrence D} and Susan Samuel and Paula Prescott and Work, {Bruce A.}",
year = "1996",
month = "1",
day = "1",
doi = "10.3109/14767059609025431",
language = "English (US)",
volume = "5",
pages = "256--261",
journal = "Journal of Maternal-Fetal and Neonatal Medicine",
issn = "1476-7058",
publisher = "Informa Healthcare",
number = "5",

}

TY - JOUR

T1 - Predicting the duration of the first stage of spontaneous labor using a neural network

AU - Devoe, Lawrence D

AU - Samuel, Susan

AU - Prescott, Paula

AU - Work, Bruce A.

PY - 1996/1/1

Y1 - 1996/1/1

N2 - To create a neural network that predicts the length of the first stage of term labor. Two hundred patients with gestations ≥36 weeks, in spontaneous active labor are the study group: 159 for training and 41 for testing; 4 training set patients had second-stage cesarean section for obstructed labor. The network is designed with Brainmaker MacIntosh 1.0 (California Scientific Software). Inputs are uterine activity, estimated fetal weight, position, station, and gestational age; maternal parity, age, height, weight, membrane status, and cervical dilatation. Actual first stages are regressed on those predicted by the network or by a standard partogram set. Differences between actual first stage lengths and those predicted by the neural network or partogram are compared with t-tests; while the proportions of first stages accurately predicted within 1 or 2 h are compared for both methods with chi-square tests. The network trained in 4 h (1388 runs) to a 0.15 tolerance. The network predictions have significantly higher correlation (r = 0.88) than do standard partograms (r = 0.35) with actual first stage durations. Mean differences between predicted and actual first stages are significantly lower for network output than with partograms; these differences increased with first stages exceeding 3 h; 100% of trained network values are within 2 h of actual first stage length. The network performs similarly for a new set of 41 previously unseen labors. This neural network predicts the length of the first stage of spontaneous labor and uses inputs readily available to obstetricians. It outperforms typical partograms for estimating this important feature of normal labor. Future application for intrapartum prognosis could be based on this successful design.

AB - To create a neural network that predicts the length of the first stage of term labor. Two hundred patients with gestations ≥36 weeks, in spontaneous active labor are the study group: 159 for training and 41 for testing; 4 training set patients had second-stage cesarean section for obstructed labor. The network is designed with Brainmaker MacIntosh 1.0 (California Scientific Software). Inputs are uterine activity, estimated fetal weight, position, station, and gestational age; maternal parity, age, height, weight, membrane status, and cervical dilatation. Actual first stages are regressed on those predicted by the network or by a standard partogram set. Differences between actual first stage lengths and those predicted by the neural network or partogram are compared with t-tests; while the proportions of first stages accurately predicted within 1 or 2 h are compared for both methods with chi-square tests. The network trained in 4 h (1388 runs) to a 0.15 tolerance. The network predictions have significantly higher correlation (r = 0.88) than do standard partograms (r = 0.35) with actual first stage durations. Mean differences between predicted and actual first stages are significantly lower for network output than with partograms; these differences increased with first stages exceeding 3 h; 100% of trained network values are within 2 h of actual first stage length. The network performs similarly for a new set of 41 previously unseen labors. This neural network predicts the length of the first stage of spontaneous labor and uses inputs readily available to obstetricians. It outperforms typical partograms for estimating this important feature of normal labor. Future application for intrapartum prognosis could be based on this successful design.

KW - Labor

KW - Neural network

KW - Partogram

UR - http://www.scopus.com/inward/record.url?scp=0029989348&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029989348&partnerID=8YFLogxK

U2 - 10.3109/14767059609025431

DO - 10.3109/14767059609025431

M3 - Article

C2 - 8930796

AN - SCOPUS:0029989348

VL - 5

SP - 256

EP - 261

JO - Journal of Maternal-Fetal and Neonatal Medicine

JF - Journal of Maternal-Fetal and Neonatal Medicine

SN - 1476-7058

IS - 5

ER -