Machine Learning for Drug Overdose Surveillance

Daniel Neill, William Herlands

Research output: Contribution to journalArticle

Abstract

We describe two recently proposed machine learning approaches for discovering emerging trends in fatal accidental drug overdoses. The Gaussian Process Subset Scan (Herlands, McFowland, Wilson, & Neill, 2017) enables early detection of emerging patterns in spatio-temporal data, accounting for both the complex, correlated nature of the data and the fact that detecting subtle patterns requires integration of information across multiple spatial areas and multiple time steps. We apply this approach to 17 years of county-aggregated data for monthly opioid overdose deaths in the New York City metropolitan area, showing clear advantages in the utility of discovered patterns as compared to typical anomaly detection approaches. To detect and characterize emerging overdose patterns that differentially affect a subpopulation of the data, including geographic, demographic, and behavioral patterns (e.g., which combinations of drugs are involved), we apply the Multidimensional Tensor Scan (Neill, 2017) to 8 years of case-level overdose data from Allegheny County, Pennsylvania. We discover previously unidentified overdose patterns which reveal unusual demographic clusters, show impacts of drug legislation, and demonstrate potential for early detection and targeted intervention. These approaches to early detection of overdose patterns can inform prevention and response efforts, as well as understanding the effects of policy changes.

Original languageEnglish (US)
Pages (from-to)8-14
Number of pages7
JournalJournal of Technology in Human Services
Volume36
Issue number1
DOIs
StatePublished - Jan 2 2018

Fingerprint

Drug Overdose
Tensors
Learning systems
surveillance
Demography
Drug Legislation
drug
Drug Combinations
Opioid Analgesics
learning
agglomeration area
legislation
Machine Learning
death
trend

Keywords

  • Disease surveillance
  • machine learning
  • opioids
  • subset scan

ASJC Scopus subject areas

  • Health(social science)
  • Social Sciences (miscellaneous)
  • Social Sciences(all)
  • Computer Networks and Communications

Cite this

Machine Learning for Drug Overdose Surveillance. / Neill, Daniel; Herlands, William.

In: Journal of Technology in Human Services, Vol. 36, No. 1, 02.01.2018, p. 8-14.

Research output: Contribution to journalArticle

@article{475714c0849d4499a98be10f3a5a22f4,
title = "Machine Learning for Drug Overdose Surveillance",
abstract = "We describe two recently proposed machine learning approaches for discovering emerging trends in fatal accidental drug overdoses. The Gaussian Process Subset Scan (Herlands, McFowland, Wilson, & Neill, 2017) enables early detection of emerging patterns in spatio-temporal data, accounting for both the complex, correlated nature of the data and the fact that detecting subtle patterns requires integration of information across multiple spatial areas and multiple time steps. We apply this approach to 17 years of county-aggregated data for monthly opioid overdose deaths in the New York City metropolitan area, showing clear advantages in the utility of discovered patterns as compared to typical anomaly detection approaches. To detect and characterize emerging overdose patterns that differentially affect a subpopulation of the data, including geographic, demographic, and behavioral patterns (e.g., which combinations of drugs are involved), we apply the Multidimensional Tensor Scan (Neill, 2017) to 8 years of case-level overdose data from Allegheny County, Pennsylvania. We discover previously unidentified overdose patterns which reveal unusual demographic clusters, show impacts of drug legislation, and demonstrate potential for early detection and targeted intervention. These approaches to early detection of overdose patterns can inform prevention and response efforts, as well as understanding the effects of policy changes.",
keywords = "Disease surveillance, machine learning, opioids, subset scan",
author = "Daniel Neill and William Herlands",
year = "2018",
month = "1",
day = "2",
doi = "10.1080/15228835.2017.1416511",
language = "English (US)",
volume = "36",
pages = "8--14",
journal = "Journal of Technology in Human Services",
issn = "1522-8835",
publisher = "Routledge",
number = "1",

}

TY - JOUR

T1 - Machine Learning for Drug Overdose Surveillance

AU - Neill, Daniel

AU - Herlands, William

PY - 2018/1/2

Y1 - 2018/1/2

N2 - We describe two recently proposed machine learning approaches for discovering emerging trends in fatal accidental drug overdoses. The Gaussian Process Subset Scan (Herlands, McFowland, Wilson, & Neill, 2017) enables early detection of emerging patterns in spatio-temporal data, accounting for both the complex, correlated nature of the data and the fact that detecting subtle patterns requires integration of information across multiple spatial areas and multiple time steps. We apply this approach to 17 years of county-aggregated data for monthly opioid overdose deaths in the New York City metropolitan area, showing clear advantages in the utility of discovered patterns as compared to typical anomaly detection approaches. To detect and characterize emerging overdose patterns that differentially affect a subpopulation of the data, including geographic, demographic, and behavioral patterns (e.g., which combinations of drugs are involved), we apply the Multidimensional Tensor Scan (Neill, 2017) to 8 years of case-level overdose data from Allegheny County, Pennsylvania. We discover previously unidentified overdose patterns which reveal unusual demographic clusters, show impacts of drug legislation, and demonstrate potential for early detection and targeted intervention. These approaches to early detection of overdose patterns can inform prevention and response efforts, as well as understanding the effects of policy changes.

AB - We describe two recently proposed machine learning approaches for discovering emerging trends in fatal accidental drug overdoses. The Gaussian Process Subset Scan (Herlands, McFowland, Wilson, & Neill, 2017) enables early detection of emerging patterns in spatio-temporal data, accounting for both the complex, correlated nature of the data and the fact that detecting subtle patterns requires integration of information across multiple spatial areas and multiple time steps. We apply this approach to 17 years of county-aggregated data for monthly opioid overdose deaths in the New York City metropolitan area, showing clear advantages in the utility of discovered patterns as compared to typical anomaly detection approaches. To detect and characterize emerging overdose patterns that differentially affect a subpopulation of the data, including geographic, demographic, and behavioral patterns (e.g., which combinations of drugs are involved), we apply the Multidimensional Tensor Scan (Neill, 2017) to 8 years of case-level overdose data from Allegheny County, Pennsylvania. We discover previously unidentified overdose patterns which reveal unusual demographic clusters, show impacts of drug legislation, and demonstrate potential for early detection and targeted intervention. These approaches to early detection of overdose patterns can inform prevention and response efforts, as well as understanding the effects of policy changes.

KW - Disease surveillance

KW - machine learning

KW - opioids

KW - subset scan

UR - http://www.scopus.com/inward/record.url?scp=85041099006&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041099006&partnerID=8YFLogxK

U2 - 10.1080/15228835.2017.1416511

DO - 10.1080/15228835.2017.1416511

M3 - Article

AN - SCOPUS:85041099006

VL - 36

SP - 8

EP - 14

JO - Journal of Technology in Human Services

JF - Journal of Technology in Human Services

SN - 1522-8835

IS - 1

ER -