Neural network prediction of nonstress test results

How often should we perform nonstress tests?

Lawrence D Devoe, Ernest Carlton, Paula Prescott

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

OBJECTIVE: Our purpose was to predict outcomes and optimal intervals for nonstress tests of term gravid women with neural networks. STUDY DESIGN: We studied 100 normal term patients whose 30-minute nonstress tests, performed on 5 consecutive days, were computer analyzed for the following elements: fetal heart rate baseline, variability, signal loss, accelerations (>15 beats/min), and decelerations. The training set used 65 patients; the testing, 35 patients. Nonstress test data (days 1 to 4) were inputs; day 5 data were training patterns. Networks for each nonstress tet element used Brainmaker Macintosh 1.0 (California Scientific Software, Nevada City, Calif.) trained to 0.12 tolerance. Actual fetal heart rate elements and their daily differences were compared with predictions by the networks and multiple regressions. RESULTS: There was little difference between networks using daily or alternate-day inputs for predicting test performance on day 5; networks using test intervals >2 days could not be trained to tolerance. Long-term fetal heart rate variation was the nonstress test element best predicted. Daily differences networks provided better prediction of all day 5 data than did actual daily values networks or multiple regression formulas. CONCLUSIONS: Baseline long-term fetal heart rate variability seems to be the most predictable fetal heart rate element over time and should merit more consideration in overall fetal testing. Fetal heart rate elements are not easily predicted by any method for intervals longer than 2 days. Using longer test intervals might run a greater risk for unanticipated changes in nonstress test outcomes, even when fetal condition is normal.

Original languageEnglish (US)
Pages (from-to)1128-1131
Number of pages4
JournalAmerican Journal of Obstetrics and Gynecology
Volume173
Issue number4
DOIs
StatePublished - Jan 1 1995

Fingerprint

Fetal Heart Rate
Deceleration
Software

Keywords

  • Neural networks
  • fetal heart rate
  • nonstress test

ASJC Scopus subject areas

  • Obstetrics and Gynecology
  • Medicine(all)

Cite this

Neural network prediction of nonstress test results : How often should we perform nonstress tests? / Devoe, Lawrence D; Carlton, Ernest; Prescott, Paula.

In: American Journal of Obstetrics and Gynecology, Vol. 173, No. 4, 01.01.1995, p. 1128-1131.

Research output: Contribution to journalArticle

@article{4ed97f2a44c54c1db5b0a46e23a018d7,
title = "Neural network prediction of nonstress test results: How often should we perform nonstress tests?",
abstract = "OBJECTIVE: Our purpose was to predict outcomes and optimal intervals for nonstress tests of term gravid women with neural networks. STUDY DESIGN: We studied 100 normal term patients whose 30-minute nonstress tests, performed on 5 consecutive days, were computer analyzed for the following elements: fetal heart rate baseline, variability, signal loss, accelerations (>15 beats/min), and decelerations. The training set used 65 patients; the testing, 35 patients. Nonstress test data (days 1 to 4) were inputs; day 5 data were training patterns. Networks for each nonstress tet element used Brainmaker Macintosh 1.0 (California Scientific Software, Nevada City, Calif.) trained to 0.12 tolerance. Actual fetal heart rate elements and their daily differences were compared with predictions by the networks and multiple regressions. RESULTS: There was little difference between networks using daily or alternate-day inputs for predicting test performance on day 5; networks using test intervals >2 days could not be trained to tolerance. Long-term fetal heart rate variation was the nonstress test element best predicted. Daily differences networks provided better prediction of all day 5 data than did actual daily values networks or multiple regression formulas. CONCLUSIONS: Baseline long-term fetal heart rate variability seems to be the most predictable fetal heart rate element over time and should merit more consideration in overall fetal testing. Fetal heart rate elements are not easily predicted by any method for intervals longer than 2 days. Using longer test intervals might run a greater risk for unanticipated changes in nonstress test outcomes, even when fetal condition is normal.",
keywords = "Neural networks, fetal heart rate, nonstress test",
author = "Devoe, {Lawrence D} and Ernest Carlton and Paula Prescott",
year = "1995",
month = "1",
day = "1",
doi = "10.1016/0002-9378(95)91338-6",
language = "English (US)",
volume = "173",
pages = "1128--1131",
journal = "American Journal of Obstetrics and Gynecology",
issn = "0002-9378",
publisher = "Mosby Inc.",
number = "4",

}

TY - JOUR

T1 - Neural network prediction of nonstress test results

T2 - How often should we perform nonstress tests?

AU - Devoe, Lawrence D

AU - Carlton, Ernest

AU - Prescott, Paula

PY - 1995/1/1

Y1 - 1995/1/1

N2 - OBJECTIVE: Our purpose was to predict outcomes and optimal intervals for nonstress tests of term gravid women with neural networks. STUDY DESIGN: We studied 100 normal term patients whose 30-minute nonstress tests, performed on 5 consecutive days, were computer analyzed for the following elements: fetal heart rate baseline, variability, signal loss, accelerations (>15 beats/min), and decelerations. The training set used 65 patients; the testing, 35 patients. Nonstress test data (days 1 to 4) were inputs; day 5 data were training patterns. Networks for each nonstress tet element used Brainmaker Macintosh 1.0 (California Scientific Software, Nevada City, Calif.) trained to 0.12 tolerance. Actual fetal heart rate elements and their daily differences were compared with predictions by the networks and multiple regressions. RESULTS: There was little difference between networks using daily or alternate-day inputs for predicting test performance on day 5; networks using test intervals >2 days could not be trained to tolerance. Long-term fetal heart rate variation was the nonstress test element best predicted. Daily differences networks provided better prediction of all day 5 data than did actual daily values networks or multiple regression formulas. CONCLUSIONS: Baseline long-term fetal heart rate variability seems to be the most predictable fetal heart rate element over time and should merit more consideration in overall fetal testing. Fetal heart rate elements are not easily predicted by any method for intervals longer than 2 days. Using longer test intervals might run a greater risk for unanticipated changes in nonstress test outcomes, even when fetal condition is normal.

AB - OBJECTIVE: Our purpose was to predict outcomes and optimal intervals for nonstress tests of term gravid women with neural networks. STUDY DESIGN: We studied 100 normal term patients whose 30-minute nonstress tests, performed on 5 consecutive days, were computer analyzed for the following elements: fetal heart rate baseline, variability, signal loss, accelerations (>15 beats/min), and decelerations. The training set used 65 patients; the testing, 35 patients. Nonstress test data (days 1 to 4) were inputs; day 5 data were training patterns. Networks for each nonstress tet element used Brainmaker Macintosh 1.0 (California Scientific Software, Nevada City, Calif.) trained to 0.12 tolerance. Actual fetal heart rate elements and their daily differences were compared with predictions by the networks and multiple regressions. RESULTS: There was little difference between networks using daily or alternate-day inputs for predicting test performance on day 5; networks using test intervals >2 days could not be trained to tolerance. Long-term fetal heart rate variation was the nonstress test element best predicted. Daily differences networks provided better prediction of all day 5 data than did actual daily values networks or multiple regression formulas. CONCLUSIONS: Baseline long-term fetal heart rate variability seems to be the most predictable fetal heart rate element over time and should merit more consideration in overall fetal testing. Fetal heart rate elements are not easily predicted by any method for intervals longer than 2 days. Using longer test intervals might run a greater risk for unanticipated changes in nonstress test outcomes, even when fetal condition is normal.

KW - Neural networks

KW - fetal heart rate

KW - nonstress test

UR - http://www.scopus.com/inward/record.url?scp=0028843837&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028843837&partnerID=8YFLogxK

U2 - 10.1016/0002-9378(95)91338-6

DO - 10.1016/0002-9378(95)91338-6

M3 - Article

VL - 173

SP - 1128

EP - 1131

JO - American Journal of Obstetrics and Gynecology

JF - American Journal of Obstetrics and Gynecology

SN - 0002-9378

IS - 4

ER -