Measuring Strategies and Challenges for Capturing Infant State Behavior

2.50
Hdl Handle:
http://hdl.handle.net/10755/150205
Type:
Presentation
Title:
Measuring Strategies and Challenges for Capturing Infant State Behavior
Abstract:
Measuring Strategies and Challenges for Capturing Infant State Behavior
Conference Sponsor:Sigma Theta Tau International
Conference Year:2001
Conference Date:November 10 - 14, 2001
Author:Barbosa, Gail, ScD
P.I. Institution Name:Medical University of South Carolina
Title:Assistant Professor
Objectives: To describe 1) development of a portable, non-invasive home sleep monitoring system for determination of infant sleep states, and 2) validation of rule-based, computer algorithm for categorization of infant sleep state. Design: Comparative study of sleep data collection and infant sleep state classification methods. Sample: Convenience sample of healthy full term male and female infants, ages 4-12 weeks of age, classified as fussy infants. Setting: Sleep data are collected in the infant’s home environment over a four day period using the piezo transducer sleep mattress and 483 Holter SSR Digicorder Recorder System. Continuous sleep data are digitally recorded on removable hard cards, then downloaded and compressed for transfer to a web site for retrieval and analysis by data specialists. Concept: the sleep collection system is lightweight, safe, small, and relatively tamper free. The portable system can be placed in any flat area where the newborn sleeps. The system records data for four days and relies on 3 alkaline batteries, requiring no electrical connection. The data are compressed and downloaded for analysis. Infant sleep state can be visually interpreted and hand scored. A computerized, rule-based system is currently under testing using an algorithm. Accuracy of the rule-based computer system in determining infant sleep state is under development, validated with hand scored interpretation and classification of infant sleep state and then compared to a computer analysis of infant sleep state, derived from rule based software. Measures: Validity of the DigiCorder and mattress system was established by comparing minute by minute human observer recordings (considered the gold standard) to the visual print out of data captured by the sleep monitoring system. The assignment of infant sleep state using the computer based algorithm was compared to hand scored classification. Agreement was analyzed using the Kappa statistic, which adjusts for agreement based on chance alone. Findings Kappa agreement of sleep data between the human observer and sleep mattress Digicorder system is 0.84. To date, kappa agreement between hand scored classification of infant sleep state and computer generated categorization of sleep state is 0.70. Disagreement is largely related to episodes of periodic infant breathing, during which the computer misclassifies infant state. Conclusions: Use of the adapted system significantly improves efficiency and convenience of collecting sleep data in the home setting. The computer rule-based system, under development, demonstrates promise in accurate and efficient classification of infant sleep state. The information gained in pilot work will be used to refine the rule based software in accurately determining infant state. Implications: The measurement of infant sleep in the home setting presents challenges that find solutions in the innovative use of technology. The development of this system may have application to the measure of other physiologic data measured in the home environment.
Repository Posting Date:
26-Oct-2011
Date of Publication:
10-Nov-2001
Sponsors:
Sigma Theta Tau International

Full metadata record

DC FieldValue Language
dc.typePresentationen_GB
dc.titleMeasuring Strategies and Challenges for Capturing Infant State Behavioren_GB
dc.identifier.urihttp://hdl.handle.net/10755/150205-
dc.description.abstract<table><tr><td colspan="2" class="item-title">Measuring Strategies and Challenges for Capturing Infant State Behavior</td></tr><tr class="item-sponsor"><td class="label">Conference Sponsor:</td><td class="value">Sigma Theta Tau International</td></tr><tr class="item-year"><td class="label">Conference Year:</td><td class="value">2001</td></tr><tr class="item-conference-date"><td class="label">Conference Date:</td><td class="value">November 10 - 14, 2001</td></tr><tr class="item-author"><td class="label">Author:</td><td class="value">Barbosa, Gail, ScD</td></tr><tr class="item-institute"><td class="label">P.I. Institution Name:</td><td class="value">Medical University of South Carolina</td></tr><tr class="item-author-title"><td class="label">Title:</td><td class="value">Assistant Professor</td></tr><tr class="item-email"><td class="label">Email:</td><td class="value">barbosag@musc.edu</td></tr><tr><td colspan="2" class="item-abstract">Objectives: To describe 1) development of a portable, non-invasive home sleep monitoring system for determination of infant sleep states, and 2) validation of rule-based, computer algorithm for categorization of infant sleep state. Design: Comparative study of sleep data collection and infant sleep state classification methods. Sample: Convenience sample of healthy full term male and female infants, ages 4-12 weeks of age, classified as fussy infants. Setting: Sleep data are collected in the infant&rsquo;s home environment over a four day period using the piezo transducer sleep mattress and 483 Holter SSR Digicorder Recorder System. Continuous sleep data are digitally recorded on removable hard cards, then downloaded and compressed for transfer to a web site for retrieval and analysis by data specialists. Concept: the sleep collection system is lightweight, safe, small, and relatively tamper free. The portable system can be placed in any flat area where the newborn sleeps. The system records data for four days and relies on 3 alkaline batteries, requiring no electrical connection. The data are compressed and downloaded for analysis. Infant sleep state can be visually interpreted and hand scored. A computerized, rule-based system is currently under testing using an algorithm. Accuracy of the rule-based computer system in determining infant sleep state is under development, validated with hand scored interpretation and classification of infant sleep state and then compared to a computer analysis of infant sleep state, derived from rule based software. Measures: Validity of the DigiCorder and mattress system was established by comparing minute by minute human observer recordings (considered the gold standard) to the visual print out of data captured by the sleep monitoring system. The assignment of infant sleep state using the computer based algorithm was compared to hand scored classification. Agreement was analyzed using the Kappa statistic, which adjusts for agreement based on chance alone. Findings Kappa agreement of sleep data between the human observer and sleep mattress Digicorder system is 0.84. To date, kappa agreement between hand scored classification of infant sleep state and computer generated categorization of sleep state is 0.70. Disagreement is largely related to episodes of periodic infant breathing, during which the computer misclassifies infant state. Conclusions: Use of the adapted system significantly improves efficiency and convenience of collecting sleep data in the home setting. The computer rule-based system, under development, demonstrates promise in accurate and efficient classification of infant sleep state. The information gained in pilot work will be used to refine the rule based software in accurately determining infant state. Implications: The measurement of infant sleep in the home setting presents challenges that find solutions in the innovative use of technology. The development of this system may have application to the measure of other physiologic data measured in the home environment.</td></tr></table>en_GB
dc.date.available2011-10-26T10:18:54Z-
dc.date.issued2001-11-10en_GB
dc.date.accessioned2011-10-26T10:18:54Z-
dc.description.sponsorshipSigma Theta Tau Internationalen_GB
All Items in this repository are protected by copyright, with all rights reserved, unless otherwise indicated.