A Neural Network Model Analysis to Identify Victims of Intimate Partner Violence

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

The objective of this study was to determine if a neural network model can identify victims of intimate partner violence (IPV). A custom neural network model was constructed and trained using the 1995 ED databases at Truman Medical Center of all female visits. The input vector developed was an array of 100 binary elements containing, in coded form, the patient's age, day of week, primary diagnosis (excluding 995.81), disposition, race, time, and E-code. The trained network was then presented with a series of 19,830 female patients from the 1996 ED database to determine if it could discriminate cases from control subjects. The neural network identified 231 of 297 known IPV victims (sensitivity 78%) in the 1996 database. It also categorized 2234 false-positive patients out of 19,533 IPV-negative patients (specificity 89%). A computer-based neural network model, when supplied with information commonly available in the ED medical record, can identify victims of IPV.

Original languageEnglish (US)
Pages (from-to)87-89
Number of pages3
JournalAmerican Journal of Emergency Medicine
Volume22
Issue number2
DOIs
StatePublished - Mar 2004

Fingerprint

Neural Networks (Computer)
Databases
Medical Records
Intimate Partner Violence

Keywords

  • Intimate partner
  • Neural network
  • Violence

ASJC Scopus subject areas

  • Emergency Medicine

Cite this

A Neural Network Model Analysis to Identify Victims of Intimate Partner Violence. / Sprecher, Armand G.; Muelleman, Robert Leo; Wadman, Michael Charles.

In: American Journal of Emergency Medicine, Vol. 22, No. 2, 03.2004, p. 87-89.

Research output: Contribution to journalArticle

@article{21fd15c18b88426c82bbd4099e1f703c,
title = "A Neural Network Model Analysis to Identify Victims of Intimate Partner Violence",
abstract = "The objective of this study was to determine if a neural network model can identify victims of intimate partner violence (IPV). A custom neural network model was constructed and trained using the 1995 ED databases at Truman Medical Center of all female visits. The input vector developed was an array of 100 binary elements containing, in coded form, the patient's age, day of week, primary diagnosis (excluding 995.81), disposition, race, time, and E-code. The trained network was then presented with a series of 19,830 female patients from the 1996 ED database to determine if it could discriminate cases from control subjects. The neural network identified 231 of 297 known IPV victims (sensitivity 78{\%}) in the 1996 database. It also categorized 2234 false-positive patients out of 19,533 IPV-negative patients (specificity 89{\%}). A computer-based neural network model, when supplied with information commonly available in the ED medical record, can identify victims of IPV.",
keywords = "Intimate partner, Neural network, Violence",
author = "Sprecher, {Armand G.} and Muelleman, {Robert Leo} and Wadman, {Michael Charles}",
year = "2004",
month = "3",
doi = "10.1016/j.ajem.2003.12.005",
language = "English (US)",
volume = "22",
pages = "87--89",
journal = "The American journal of emergency medicine",
issn = "0735-6757",
publisher = "W.B. Saunders Ltd",
number = "2",

}

TY - JOUR

T1 - A Neural Network Model Analysis to Identify Victims of Intimate Partner Violence

AU - Sprecher, Armand G.

AU - Muelleman, Robert Leo

AU - Wadman, Michael Charles

PY - 2004/3

Y1 - 2004/3

N2 - The objective of this study was to determine if a neural network model can identify victims of intimate partner violence (IPV). A custom neural network model was constructed and trained using the 1995 ED databases at Truman Medical Center of all female visits. The input vector developed was an array of 100 binary elements containing, in coded form, the patient's age, day of week, primary diagnosis (excluding 995.81), disposition, race, time, and E-code. The trained network was then presented with a series of 19,830 female patients from the 1996 ED database to determine if it could discriminate cases from control subjects. The neural network identified 231 of 297 known IPV victims (sensitivity 78%) in the 1996 database. It also categorized 2234 false-positive patients out of 19,533 IPV-negative patients (specificity 89%). A computer-based neural network model, when supplied with information commonly available in the ED medical record, can identify victims of IPV.

AB - The objective of this study was to determine if a neural network model can identify victims of intimate partner violence (IPV). A custom neural network model was constructed and trained using the 1995 ED databases at Truman Medical Center of all female visits. The input vector developed was an array of 100 binary elements containing, in coded form, the patient's age, day of week, primary diagnosis (excluding 995.81), disposition, race, time, and E-code. The trained network was then presented with a series of 19,830 female patients from the 1996 ED database to determine if it could discriminate cases from control subjects. The neural network identified 231 of 297 known IPV victims (sensitivity 78%) in the 1996 database. It also categorized 2234 false-positive patients out of 19,533 IPV-negative patients (specificity 89%). A computer-based neural network model, when supplied with information commonly available in the ED medical record, can identify victims of IPV.

KW - Intimate partner

KW - Neural network

KW - Violence

UR - http://www.scopus.com/inward/record.url?scp=1642300824&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1642300824&partnerID=8YFLogxK

U2 - 10.1016/j.ajem.2003.12.005

DO - 10.1016/j.ajem.2003.12.005

M3 - Article

C2 - 15011219

AN - SCOPUS:1642300824

VL - 22

SP - 87

EP - 89

JO - The American journal of emergency medicine

JF - The American journal of emergency medicine

SN - 0735-6757

IS - 2

ER -