Epidemic Modelingwith Discrete Space Scheduled Walkers

Epidemic Modelingwith Discrete Space Scheduled Walkers

The Utility of Agent Based Models: Epidemics, Epizootics, and Healthcare Friday, April 3 2:00 3:00 p.m. Engineering Senate Chambers Rm: E3-262 VW-Superbug Coughing Pigs MRSA-Superbug Bob McLeod ECE Dept. Seminar Series Wearing a Suit for the Auspicious Occasion Internet Innovation Center ALL WELCOME! The Utility of Agent Based Models:

Applications to Epidemics, Epizootics, Healthcare, Preparedness Planning, etc. Opportunities for Research Robert D. McLeod [email protected] Professor ECE University of Manitoba Internet Innovation Centre (IIC) Dept. Electrical and Computer Engineering University of Manitoba IIC, April 3. 2009 Internet Innovation Center Overview Part One: Agent Based Models (ABM) Introduction

Motivation: Interest in modeling complex systems Part Two: Examples of ABM Utility Epidemic modeling: Discrete Space Scheduled Walker Epizootic modeling: Patient Access and Emergency Department Waiting Time Reduction Nosocomial Infections (rhymes with polynomial) Part Three: Extensions and ECE Opportunities

Summary/Discussion Interspersed with pop science references and questions Internet Innovation Center 3 Overview Goals (Future) : Develop high utility ABM simulators Wrt epidemics: Preparedness, recovery, mitigation, policy Wrt healthcare: Patient access, nosocomial infections Goals (Present): Garner Interest toward grant applications

Looking for $20K as matching funds (sources or leads) In general: Preaching the ABM Gospel and its Utility: Hallelujah Internet Innovation Center 4 Part 1: Book Reviews/Motivation World Without Us: Alan Weisman Pandemonium: Andrew Nikiforuk

The Numerati: Stephen Baker Super Crunchers: Ian Ayres The Tipping Point, Outliers: Malcolm Gladwell The Black Swan, Fooled by Randomness: Nassim Taleb The Man Who Knew Too Much: Alan Turing and the Invention of the Computer: David Leavitt Internet Innovation Center

5 Part 1: Agent Based Modeling Long time interest: Complex Systems and Modeling Research resulted from a Programming Challenge Make the equations as simple as possible, but not simpler, Albert Einstein ABM is computational modeling essentially devoid of governing equations

ABMs are pure mathematics. Is that a G.H. Hardy reference? No, its a G. Boole reference. Internet Innovation Center 6 Making models more useful You can observe a lot by watching: Yogi Berra Prediction is very difficult, especially about the future: Niels Bohr In the country of the blind, the one-eyed man is King: Desiderius Erasmus Internet Innovation Center

How?: Data Mining and Statistical Inferencing using an ABM engine Refs: Wikipedia 7 Part 2: Agent Based Modeling Utility App1: A nice attribute about ABMs in general is that they are ideal idea communication vehicles App2: Epidemic modeling - DSSW Model

Epizootic modeling Extension to ABMs that have not been exploited App3: Modeling an Emergency Department Further demonstration of ABM utility Waiting times and nosocomial infections Internet Innovation Center 8 App1: Initial Specification for

Epidemic Modeling Basis idea: Data mine people-people interactions. (Often Disparate Sources) Topology: Data mined from maps Behaviour: Data mined from demographics Models based on real network topologies and scheduled walkers. The goal of the research is to shed light on the problems with very complicated phenomena through data-driven modeling and simulation and statistical inference. Internet Innovation Center

9 The Model Data mining is a common theme in modern information technology: Analytical methods may not exist or are complex. Data does exists and can be extracted. Statistical methods can easily deal with the vast amount of data that is available (or becoming so). Our work here is an attempt to help promote data-driven epidemic simulation and modeling:

Where data is available we demonstrate its utility, where unavailable we demonstrate how it would be utilized. Unavailable refers to practical or political limitations. Internet Innovation Center 10 Where: Topological Data Sources Google Earth with Overlays Google Maps Correct by construction small world topologies Internet Innovation Center 11

Who and When Of similar importance to location (where), is the agents (who) are being infected. This data is generally technically available but may be practically unavailable. An agents schedule (when) is also of critical importance. This data is more typically inferred rather than explicitly available, but as we are primarily creatures of habit reasonable assumptions can be made. Internet Innovation Center 12

What The what here is typically a disease, either bacterial or viral, communicated with an associated probability of contraction when in contact with an infectious agent. Example 1 of stochastic behaviour: Example 2 of stochastic behaviour: Modified schedule when ill: Low mobility when sick or getting sick. (Nota Bene: the agent decides to stay home)

Weighted random schedule. (Dont feel like going to work today) Example of contact: Physical touch, third party (door knob), cough. Internet Innovation Center 13 Implementation Based on the model our underlying simulation model is that of a Discrete-Space Scheduled Walker (DSSW). In contrast to other models that are based on random or Brownian walkers on artificial topologies.

We capture the most important aspects of real-people networks, incorporating (correct by construction) notions such as small world networks, scale free networks. it is what it is Internet Innovation Center (nota bene) 14 What if I live here I take this bus I work here Internet Innovation Center 15

City of Winnipeg, population: 635,869 The User Interface to DSSW Parameters for simulation are set up in a number of files and the user can step or loop through the simulation at any given rate. During the simulation, a number of plots and statistics are collected and logged to a web server where the user can then further analyze the simulation run. Internet Innovation Center 16 Analysis

Some data that is available on the corresponding web server Internet Innovation Center 17 Seasonal Variations Seasonal variations are well known and provide fairly well labeled data for comparison Comparison allows for a tuning of parameters to more closely reflect actual data collected for a particular disease

Internet Innovation Center 18 Mutations tipping point Seasonal Variation A mutation to a deadlier strain or a sudden variation in the mode of transmission (e.g. virus shift or drift, bioterrorism) Other uses would be in helping to evaluate the extent of inoculations (Herd immunity) or policies. This will allow for epidemiologists to partially close the loop when evaluating policy. (ABM utility, ref. CDC) Internet Innovation Center 19

App1: DSSW Summary Introduced an ABM for epidemic modeling. Basic characteristic of the model is to extract and combine real topographic and demographic data. Data mining and scheduled walkers. Model using real data is feasible Results in better characterization of epidemic dynamics

Further work will focus on refining the model, and validating the afore-mentioned conjecture. Complementary to equation based approaches Internet Innovation Center 20 App2: ABM Potential for Epizootics Epizootics: outbreak of disease affecting many animals Agent based modeling of epizootics. Domestic, feral, and/or natural ABBOTSFORD, B.C. - The H5 avian influenza virus has been confirmed on a commercial turkey farm in British Columbia's Fraser Valley, and as many as 60,000 birds will be euthanized, the

Canadian Food Inspection Agency said Saturday. January 24/09 Timely if nothing else Internet Innovation Center 21 ABM Potential for epizootics Nicely constrained problem: Many Intensive Livestock Production Operations are nearly Farrow to Fork Best chances of ABM demonstrated utility Cattle, swine and poultry (BTW: 1918 was a swine flu) e.g. A pork producer should be interested in the potential of an ABM as a tool in modeling a swine production environment. Extendable beyond a single farm to an entire region including transport and processing. Allow CFIA to Model: Bio-security measures

Figure 3 Internet Innovation Center 22 Similar ABMs for Poultry Broiler grow-out intensive unit production. Similar epizootic concerns Manmade pathogen reservoir Similar problems in other

monocultures Internet Innovation Center 23 Mobility and Infection Longevity Percent dead 100% Mobility/Longevity Impact Substantive shift in the Percolation Threshold (decrease) Percolation threshold is like a tipping point Mobility has a big effect: The mobility threshold for disease is a critical percolation phenomenon for an epizootic 5% 42% Population Density

Internet Innovation Center 24 Percolation with mobility. Our study was a very preliminary attempt to use ABMs for ILPO (Swam or particle type ABMs) Although crude, clearly illustrates the impact of mobility on disease spread Provides design feedback on ILPOs w/o mobility with mobility Disease Spread Internet Innovation Center 25 Epizootic Research Grant: Epilogue

Pork Board Proposal Rejected: Comments These models are great except... they are models. In practice I fail to see how this model gets us any closer to solving disease problems. Perhaps not modeling diseases will get us closer All but $2,000 of the $50,000 is for support of grad students. Next time Ill add an Ford F-150 (farm truck) Science in your Papers, Science Fiction in Proposals Internet Innovation Center 26 App3: ABMs for Patient Access

Models for reducing Emergency Department waiting times and improving patient diversion policies. Useful for closing the loop when evaluating policy decisions Staffing, resources, patient diversion Agent based simulation of an Emergency Department Models patient flow through the modeling of individuals (patients, doctors, service agents (registration, triage))

Examination rooms, waiting rooms (topology) Internet Innovation Center 27 Emergency Department Scenario Basic ED setting with data collection resources illustrated. i.e. Empirical data collected here could be used in the ED and patient diversion simulator. An ED ABM will Integrate with

whatever technology becomes available (ECE Benefit) E.g. Modification of patient arrival and treatment times. Provide initial conditions for simulation Internet Innovation Center 28 Metropolitan Multiple ED Scenario Integrated telecom backbone for a regional health authority. Data backhauled to a

central server (CORE) for processing, simulation, and policy optimization. Illustrates use of simulation enhanced patient diversion policy. e.g. Ambulances and walk in patients. NHS Trend: No diversion, better triage. Internet Innovation Center 29 Simulation Proof of Concept Visual Simulation Suite Screenshot Object oriented (OO), open-source, visual simulator to analyze and forecast emergency

department waiting times. (Nota Bene) EDs can be instantiated with various resources, patient loads and associated triage levels Internet Innovation Center 30 Simulation Scenarios City wide scenarios Two EDs with two doctors, two EDs with three doctors, two EDs with four doctors.

Effect of different staffing levels is compared when there is no patient diversion Similar basic scenario is used to compare patient diversion models. Internet Innovation Center 31 Simulation Scenario (Patient Diversion) Patient diversion modeled using Random Early Detection (RED) algorithm from Telecommunication Network Engineering. Random RED: patients diverted to random ED

Requires local ED information only Guided RED: patients probabilistically sent to EDs with fewer patients waiting Requires city wide communication and coordination Internet Innovation Center 32 Simulations and results Varying the number of Doctors, no patient diversion Two Doctors

Queue Lengths: For fewer doctors queue lengths are longer. Three Doctors Four Doctors Internet Innovation Center 33 Simulations and results Varying redirection policy, averaged across all EDs Queue Length: Scenario with the information sharing experiences the

shortest queues without additional resource allocation No diversion Diversion to random ED Probabilistic diversion to less busy ED Internet Innovation Center 34 Demonstration: Video on YouTube: http://www.youtube.com/watch?v=_6-Hk-_1MJ8&

Extensions: Machine Learning for Policy and Resource Provisioning Use the model as a starting environment for modeling the spread of an infectious disease within a Hospital. (Extends Agent) Anecdotal evidence for modeling to improve policy One of these is not a taxi Internet Innovation Center 35

Making models more useful Agree All models are wrong but some models are useful. George E.P. Box, Statistician Truth is ever to be found in the simplicity, and not in the multiplicity and confusion of things. Sir Isaac Newton Perhaps truth can actually be found in the multiplicity and confusion of things! Us Ref: Wikipedia Internet Innovation Center 36

Part 3: Possible Extensions and data Mining Opportunities At present DSSW epidemic ABM appears mainly well suited to egalitarian type diseases Who agnostic disease Here we present a few extensions and opportunities Extensions of utility to secondary/tertiary interest groups Manitoba Hydro, CFPA, Manitoba EMO, Public Safety, etc. Preparedness planning, mitigation and recovery Internet Innovation Center 37

Data Mining Comment: Data Mining is the process of processing large amounts of data and picking out relevant information. (wiki defn: common notion) Here data mining is 2 phase. Mining what to mine Mining the what Data Fusion: combine data from multiple sources Internet Innovation Center Mine what to mine Data Mining

Data Fusion 38 DSSW Extensions: Hierarchy Incorporate Hierarchy Intracity and Intercity Basic modality remains: data-driven models of discrete space- and time- walkers, (mined). Cities are largely autonomous Allows for the problem to remain tractable and allow for efficient modes of computation (parallelism can be exploited). Internet Innovation Center 39

Extensions: Extracting Patterns of Behaviour Patterns of behavior can be taken from tracking technologies that are in place albeit not mined for use in epidemic modeling. E.g. Financial Transaction Profiling Usually mined to detect fraud E.g. Cell phone tracking, where are you services By default the service provider already knows where you are, even more so with GPS Potential Obstacle: Privacy Internet Innovation Center 40 Related Research: Extracting Patterns of Behaviour (Benefit of being an ECE)

Consumer wireless electronics: MAC snooping and tracking. (non obvious data source) Bluetooth headsets (ingress and egress of signalized arterials) Similar protocols for WiFi Device-enabled Kiosks and vending machines Security cameras and systems with person detection Monitoring for behaviour patterns those of illegal activities and terrorist threats Internet Innovation Center 41 Related Research: Extracting Patterns of Behaviour (Benefit of being an ECE) http://gigapan.org/viewGigapanFullscreen.php?auth=033

ef14483ee899496648c2b4b06233c Internet Innovation Center 42 Related Research: Extracting Patterns of Behaviour from Demographics Clickable(minable) neighborhood demographic information: http://www.toronto.ca/demographics/profiles_map_and_index.htm Internet Innovation Center 43 Related Research: Extracting Patterns of Behaviour continued Tracking subway ridership.

Token data mining of ridership Their Objective: Bioterrorism impact Mining online transportation information systems Helsinki public transport Input to ABM Internet Innovation Center 44 Related Research: Real-time Helsinki Public Transport Information Internet Innovation Center 45 Related Research:

Ubiquitous Vehicle Tracking Cameras Modeling Arterials for traffic flow. ITS data useful for epidemic modeling Similar data is available for air traffic. Ref: http://www.edmontontrafficcam.com/cams.php Internet Innovation Center 46 Related Research: Extracting Patterns of Behaviour (Economic Impact)

Economic Impact: Costs to implement policy. (ref: Brookings) Specifically, the economic impact of restricting air travel as a policy in controlling a flu pandemic. Models global air travel and estimates impact and cost associated with travel restrictions. E.g. 95% travel restriction required before significantly impairing disease spread Not a surprise (also they removed edges not vertices, cf. percolation) Internet Innovation Center 47 Related Research: Extracting Patterns of Behaviour (Economic Impact) Internet Innovation Center 48

Related Research: Googles Flu trends Researchers "found that certain search terms are good indicators of flu activity. Google Flu Trends uses aggregated Google search data to estimate flu activity in your state up to two weeks faster than traditional systems" such as data collected by CDC. Internet Innovation Center 49

Related Opportunity: Googles Gmail Google mail (gmail) provides an example of data mining to extract coarse spatial behaviour patterns. gmail, web/mail server has a reasonable estimate of your activity status (busy, available, idle, offline, etc.). In addition to status, your web browser's IP address also provides coarse-grained information of where you are logged in. If I access gmail from a mobile device, this is also known to various degrees. Eric Schmidt, CEO of Google, said, "From a

technological perspective, it is the beginning." Internet Innovation Center 50 Other sources of information/ concern Occasional/periodic mass gatherings E.g. Special event that may perturb a global simulation E.g. The Hajj Largest mass pilgrimage in the world. 2007 an estimated 2-3 million people participated. Conditions are difficult and thus offers opportunity for a large scale disease such as influenza to take hold. These people then disperse to their home countries, many via public transport, and could easily influence the spread and outbreak of the disease.

Internet Innovation Center 51 Mass Gatherings: Hajj Tawaf, circumambulation of the Kabah Mosque at Kabah Internet Innovation Center 52 Related Research: Extracting Patterns of Behaviour (RFID tracking/RTLS)

Although not explicit, patterns of behavior and interactions of agents can be extracted at critical institutions such as hospitals, through the use of RFID tracking. As RFID sensor networks move from inventory to enhanced applications, data collected from RFID tracking at clinics and hospitals can be envisioned as an input to an ABM. (similar to WiFi Campus tracking) Internet Innovation Center 53 Preparedness, planning and mitigation Preparedness planning: A massive undertaking but one in which an ABM city model could be useful in providing

planners with policies and expectation how goods and services could be provisioned in the event of a catastrophe. This aspect can be catastrophe agnostic Simple investigations as to how long food/fuel/medical supplies would last and could be distributed will be modeled. Internet Innovation Center 54 Preparedness, planning and mitigation Provisioning of resources extempore will lead to an

aggravated and worsening disaster. Models can become an effective tool for any city. Specific model to their region Allowing for provisioning not only of supplies but for inoculation services as well as temporary hospital and/or mortuary facilities. Internet Innovation Center 55 Preparedness, planning and mitigation Power generation: Remote maintained by healthy individuals: Stakeholders

Hydro/Electric Easily Isolated: Transportation Stakeholders: EMO Food production/provisions: Local Stakeholders: Producers(Distributors) Water Supply: Remote, EMO Result: Pandemic Lag if Prepared Internet Innovation Center 56 Multiple Hospital Model Patient Diversion : Future Work Incorporate empirical data mined from ITS sources such as Google/Globis real-time traffic to estimate delays the ambulance would experience enroute

Internet Innovation Center 57 Summary Presented our Agent Based Modeling approach to high utility simulation. Emphasis on data mining of spatial topologies and agent behavior patterns Presented several indirect data sources Non-obvious connection to epidemic modeling Good for ECEs though Presented potential extensions: Utility of ABMs Epidemics, Epizootics, ED Wait times, Nosocomials Opportunities in preparedness planning, mitigation Internet Innovation Center

58 Ideally one would like to model everything: (someday will) Threats: epidemic natural or bio-terrorist. (In progress) Model impact of policy, provisioning (PHAC) Model Food Supply: Intensive unit production facilities through from birth to slaughter. Model Food and Fuel Supply and Distribution:

Guidelines for stock provisioning. (CPMA) Model infrastructure: Transportation, water, power. Model impact of policy (Amenable to ABMs) Assess interest in moving forward, from tertiary groups. Internet Innovation Center 59 Exploring research opportunities Being devoid of equations, agent based models allow for a tradeoffs between specificity and utility. We would like to be part of a larger modeling effort and want to explore that possibility. Extend models beyond epidemics to related areas of direct interest to Manitoba.

Assessing interest to provide some degree of matching funds to apply for a MITACS seed grant. May 2009. Total matching funds we are targeting is 20K, providing 70K of funding if successful. Leverage other efforts: Possible with some traction here Internet Innovation Center 60 Dissemination efforts: Epi-at-home.com: Future home of Epidemic ABM open source project (DSSW)

Bio-inference.ca: Future home of ABM and data mining opportunities (non obvious sources) Epizootic, patient access, preparedness planning Facebook group: Pandemic Awareness Day Exploring social networks as an information tool A non invasive information portal (140+ members) A growing number of papers/proposals/talks.

Internet Innovation Center 61 A mathematicians apology: ABMs are facing considerable resistance from the Mathematical community. ABMs will never be useful Recent quote from a famous and influential mathematician. Two Mistakes: His null hypothesis can only be rejected .

I think there is a world market for maybe five computers T.J. Watson (IBM 1943) Even brilliant people make mistakes. Internet Innovation Center 62 Miscellany: Developing ABM Codes Product development often under a water fall model. Errors introduced early and not contained or corrected amplify costs, time and money.

Lessons learned in knowledge translation. The specification should be as close to executable as possible. (Any modern telecom protocol) Verification is more easily undertaken with an ABM during the development cycle. These are byproducts of an ABM Internet Innovation Center 63 Miscellany: ABMs vs. IBMs

Individual based models (IBMs) Emphasis on the agent as the individual (person) Agent based models (ABMs) Agents can be people, animals, or inanimate objects (e.g. piece of medical equipment in a hospital) This generalization of the notion of agent constitutes an implementation advantage, from an objectoriented paradigm Analogy: Executable Specification

Internet Innovation Center 64 IIC Contact: U of M ABM initiatives Bob McLeod Professor ECE University of Manitoba Internet Innovation Center E3-416 EITC University of Manitoba Winnipeg, Manitoba R3T 5V6 Acknowledgements: Too many to list. Named throughout talk. Healthcare ABM Research: Marek Laskowski

Email: [email protected] http://www.iic.umanitoba.ca Internet Innovation Center 65 Pandemic Awareness Day: A Facebook Group that Invites You to Join Us! Seasonal Variation of Influenza or Facebook Ad? 140 100 Members as of Mar.16/09

100% Percent infected Mobility/Longevity Impact Substantive shift in the Percolation Threshold (decrease) Percolation threshold is like a tipping point 5% 42% Internet Innovation Center Population Density 67

Recently Viewed Presentations

  • Notes Section 1.3

    Notes Section 1.3

    Geometry Notes Section 4-2 What you'll learn. . . . How to apply the angle sum theorem How to recognize exterior angles and apply the exterior angle theorem Vocabulary Exterior angle of a triangle Remote interior angles Third angle theorem...
  • International Monetary Fund September 16, 2016 The Long

    International Monetary Fund September 16, 2016 The Long

    Sources: CEIC Data Company Ltd.; and IMF staff estimates. Rebalancing process starts as stimulus is wound down. 2011 marked a big shift in policies. In Figure 1 investment peaks in 2011.
  • Praktyka w Binz 2009

    Praktyka w Binz 2009

    Dni wolne podczas naszych praktyk to żadne leniuchowanie. Razem postanowiliśmy zobaczyć wszystkie najciekawsze miejsca znajdujące się w pobliżu. Toteż każdy dzień wolny upływał nam na podróżach w niezwykle komfortowych i tanich niemieckich pociągach. Odwiedziliśmy między innymi takie miasta jak: Bergen...
  • Sensory Systems - Western Washington University

    Sensory Systems - Western Washington University

    Sensory Systems 1. False. My 6th sense is telling me that this is a function of how fast the information is received by the brain. The difference in perception between a strong smell and a weak smell is due to...
  • Jacks gone to the dogs in Alaska February

    Jacks gone to the dogs in Alaska February

    Arial Times New Roman Wingdings Symbol Tahoma Project Overview (Standard) Microsoft Word Document Microsoft Clip Gallery Microsoft Graph 2000 Chart Microsoft Graph Chart Jack's gone to the dogs in Alaska February 25, 2005 Slide 2 Analyses of Lattice Squares Analyses...
  • Once you have logged in - you will need to MODIFY <click> your registration. This is used to change your contact information, add new workshops, or un-register from workshops you will not be able to attend.
  • The Italian Behavioral Risk Factor Surveillance System ...

    The Italian Behavioral Risk Factor Surveillance System ...

    A sample of at least 275 persons per LHU are interviewed every month by previously trained ASL personnel using a standardized questionnaire with questions adapted from other surveys or surveillance systems like ISTAT on national level and BRFSS, Cindi, Steps...
  • 22nd Alzheimer Europe Conference Vienna 2012 What is

    22nd Alzheimer Europe Conference Vienna 2012 What is

    The Dementia Adviser Service Implementation Group, including Dawn Brooker, Kathy Dale, David Hitchen, Kumbi Mandinyenya, Carole Edwards and Bernie Coope. All participants in the evaluation in particular the people living with dementia and their families and supporters. Association for Dementia...