Spatiotemporal AED Location Optimization

Timothy C.Y. Chan, Christopher L.F. Sun, Derya Demirtas, Laurie J. Morrison, Steven C. Brooks

Research output: Contribution to journalMeeting AbstractAcademic

Abstract

Background: Mathematical optimization can be used to plan future AED placement to maximize out-of-hospital cardiac arrest (OHCA) coverage. Many public access AEDs are placed in locations without 24/7 access. AED coverage can be overestimated unless temporal availability is considered.
Objective: To develop a new spatiotemporal AED location optimization model that accounts for both spatial and temporal information.

Methods: We identified all atraumatic public-location OHCAs occurring in Toronto, Canada from Jan. 2006 – Aug. 2014. We gathered location and operating hours data for 4898 buildings that were used as potential sites for AED placement. We extended a previously published spatial optimization model, which identifies locations to place AEDs that maximize the number of historical OHCAs occurring within 100 m of an AED. The new spatiotemporal model finds AED locations that maximize the number of OHCAs occurring within 100 m of an available AED, considering when the OHCAs occurred (“actual coverage”). We then compared the spatial and spatiotemporal models on actual coverage of out-of-sample OHCAs using 10-fold cross validation. Statistical analysis was performed using McNemar’s test.
Results: We identified 2440 atraumatic public-location OHCAs. AED locations chosen by the spatiotemporal model outperformed those chosen by the spatial model by 26.1% in actual coverage (p<0.001). The figure shows coverage improvement at all times of day: daytime (11.2%), evening (37.4%), and night (292.3%). Equivalently, 40.2% fewer AEDs are needed when using the spatiotemporal model to reach the same level of actual coverage provided by AEDs located according to the spatial model.
Conclusion: Spatiotemporal optimization can maximize actual OHCA coverage by accounting for AED availability when identifying future AED locations. The largest gains occurred during the evening and night, which is when the largest coverage losses were experienced by Toronto’s existing AEDs.
Original languageEnglish
Article numberA16492
JournalCirculation
Volume132
Issue numberSuppl. 3
Publication statusPublished - 2015
EventResuscitation Science Symposium 2015 - Orange County Convention Center, Orlando, United States
Duration: 7 Nov 20159 Nov 2015

Fingerprint

Out-of-Hospital Cardiac Arrest
Canada

Cite this

Chan, T. C. Y., Sun, C. L. F., Demirtas, D., Morrison, L. J., & Brooks, S. C. (2015). Spatiotemporal AED Location Optimization. Circulation, 132(Suppl. 3), [A16492].
Chan, Timothy C.Y. ; Sun, Christopher L.F. ; Demirtas, Derya ; Morrison, Laurie J. ; Brooks, Steven C. / Spatiotemporal AED Location Optimization. In: Circulation. 2015 ; Vol. 132, No. Suppl. 3.
@article{47e01943959d435bac4b0e17a0560e47,
title = "Spatiotemporal AED Location Optimization",
abstract = "Background: Mathematical optimization can be used to plan future AED placement to maximize out-of-hospital cardiac arrest (OHCA) coverage. Many public access AEDs are placed in locations without 24/7 access. AED coverage can be overestimated unless temporal availability is considered.Objective: To develop a new spatiotemporal AED location optimization model that accounts for both spatial and temporal information.Methods: We identified all atraumatic public-location OHCAs occurring in Toronto, Canada from Jan. 2006 – Aug. 2014. We gathered location and operating hours data for 4898 buildings that were used as potential sites for AED placement. We extended a previously published spatial optimization model, which identifies locations to place AEDs that maximize the number of historical OHCAs occurring within 100 m of an AED. The new spatiotemporal model finds AED locations that maximize the number of OHCAs occurring within 100 m of an available AED, considering when the OHCAs occurred (“actual coverage”). We then compared the spatial and spatiotemporal models on actual coverage of out-of-sample OHCAs using 10-fold cross validation. Statistical analysis was performed using McNemar’s test.Results: We identified 2440 atraumatic public-location OHCAs. AED locations chosen by the spatiotemporal model outperformed those chosen by the spatial model by 26.1{\%} in actual coverage (p<0.001). The figure shows coverage improvement at all times of day: daytime (11.2{\%}), evening (37.4{\%}), and night (292.3{\%}). Equivalently, 40.2{\%} fewer AEDs are needed when using the spatiotemporal model to reach the same level of actual coverage provided by AEDs located according to the spatial model.Conclusion: Spatiotemporal optimization can maximize actual OHCA coverage by accounting for AED availability when identifying future AED locations. The largest gains occurred during the evening and night, which is when the largest coverage losses were experienced by Toronto’s existing AEDs.",
author = "Chan, {Timothy C.Y.} and Sun, {Christopher L.F.} and Derya Demirtas and Morrison, {Laurie J.} and Brooks, {Steven C.}",
note = "Abstracts From the American Heart Association's 2015 Scientific Sessions and Resuscitation Science Symposium",
year = "2015",
language = "English",
volume = "132",
journal = "Circulation",
issn = "0009-7322",
publisher = "Lippincott Williams and Wilkins",
number = "Suppl. 3",

}

Chan, TCY, Sun, CLF, Demirtas, D, Morrison, LJ & Brooks, SC 2015, 'Spatiotemporal AED Location Optimization' Circulation, vol. 132, no. Suppl. 3, A16492.

Spatiotemporal AED Location Optimization. / Chan, Timothy C.Y.; Sun, Christopher L.F.; Demirtas, Derya; Morrison, Laurie J.; Brooks, Steven C.

In: Circulation, Vol. 132, No. Suppl. 3, A16492, 2015.

Research output: Contribution to journalMeeting AbstractAcademic

TY - JOUR

T1 - Spatiotemporal AED Location Optimization

AU - Chan, Timothy C.Y.

AU - Sun, Christopher L.F.

AU - Demirtas, Derya

AU - Morrison, Laurie J.

AU - Brooks, Steven C.

N1 - Abstracts From the American Heart Association's 2015 Scientific Sessions and Resuscitation Science Symposium

PY - 2015

Y1 - 2015

N2 - Background: Mathematical optimization can be used to plan future AED placement to maximize out-of-hospital cardiac arrest (OHCA) coverage. Many public access AEDs are placed in locations without 24/7 access. AED coverage can be overestimated unless temporal availability is considered.Objective: To develop a new spatiotemporal AED location optimization model that accounts for both spatial and temporal information.Methods: We identified all atraumatic public-location OHCAs occurring in Toronto, Canada from Jan. 2006 – Aug. 2014. We gathered location and operating hours data for 4898 buildings that were used as potential sites for AED placement. We extended a previously published spatial optimization model, which identifies locations to place AEDs that maximize the number of historical OHCAs occurring within 100 m of an AED. The new spatiotemporal model finds AED locations that maximize the number of OHCAs occurring within 100 m of an available AED, considering when the OHCAs occurred (“actual coverage”). We then compared the spatial and spatiotemporal models on actual coverage of out-of-sample OHCAs using 10-fold cross validation. Statistical analysis was performed using McNemar’s test.Results: We identified 2440 atraumatic public-location OHCAs. AED locations chosen by the spatiotemporal model outperformed those chosen by the spatial model by 26.1% in actual coverage (p<0.001). The figure shows coverage improvement at all times of day: daytime (11.2%), evening (37.4%), and night (292.3%). Equivalently, 40.2% fewer AEDs are needed when using the spatiotemporal model to reach the same level of actual coverage provided by AEDs located according to the spatial model.Conclusion: Spatiotemporal optimization can maximize actual OHCA coverage by accounting for AED availability when identifying future AED locations. The largest gains occurred during the evening and night, which is when the largest coverage losses were experienced by Toronto’s existing AEDs.

AB - Background: Mathematical optimization can be used to plan future AED placement to maximize out-of-hospital cardiac arrest (OHCA) coverage. Many public access AEDs are placed in locations without 24/7 access. AED coverage can be overestimated unless temporal availability is considered.Objective: To develop a new spatiotemporal AED location optimization model that accounts for both spatial and temporal information.Methods: We identified all atraumatic public-location OHCAs occurring in Toronto, Canada from Jan. 2006 – Aug. 2014. We gathered location and operating hours data for 4898 buildings that were used as potential sites for AED placement. We extended a previously published spatial optimization model, which identifies locations to place AEDs that maximize the number of historical OHCAs occurring within 100 m of an AED. The new spatiotemporal model finds AED locations that maximize the number of OHCAs occurring within 100 m of an available AED, considering when the OHCAs occurred (“actual coverage”). We then compared the spatial and spatiotemporal models on actual coverage of out-of-sample OHCAs using 10-fold cross validation. Statistical analysis was performed using McNemar’s test.Results: We identified 2440 atraumatic public-location OHCAs. AED locations chosen by the spatiotemporal model outperformed those chosen by the spatial model by 26.1% in actual coverage (p<0.001). The figure shows coverage improvement at all times of day: daytime (11.2%), evening (37.4%), and night (292.3%). Equivalently, 40.2% fewer AEDs are needed when using the spatiotemporal model to reach the same level of actual coverage provided by AEDs located according to the spatial model.Conclusion: Spatiotemporal optimization can maximize actual OHCA coverage by accounting for AED availability when identifying future AED locations. The largest gains occurred during the evening and night, which is when the largest coverage losses were experienced by Toronto’s existing AEDs.

M3 - Meeting Abstract

VL - 132

JO - Circulation

JF - Circulation

SN - 0009-7322

IS - Suppl. 3

M1 - A16492

ER -

Chan TCY, Sun CLF, Demirtas D, Morrison LJ, Brooks SC. Spatiotemporal AED Location Optimization. Circulation. 2015;132(Suppl. 3). A16492.