Predicting Covid-19 pandemic waves with biologically and behaviorally informed universal differential equations
© 2024 The Author(s)..
During the COVID-19 pandemic, it became clear that pandemic waves and population responses were locked in a mutual feedback loop in a classic example of a coupled behavior-disease system. We demonstrate for the first time that universal differential equation (UDE) models are able to extract this interplay from data. We develop a UDE model for COVID-19 and test its ability to make predictions of second pandemic waves. We find that UDEs are capable of learning coupled behavior-disease dynamics and predicting second waves in a variety of populations, provided they are supplied with learning biases describing simple assumptions about disease transmission and population response. Though not yet suitable for deployment as a policy-guiding tool, our results demonstrate potential benefits, drawbacks, and useful techniques when applying universal differential equations to coupled systems.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2024 |
---|---|
Erschienen: |
2024 |
Enthalten in: |
Zur Gesamtaufnahme - volume:10 |
---|---|
Enthalten in: |
Heliyon - 10(2024), 4 vom: 29. Feb., Seite e25363 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Kuwahara, Bruce [VerfasserIn] |
---|
Links: |
---|
Themen: |
Compartmental model |
---|
Anmerkungen: |
Date Revised 20.02.2024 published: Electronic-eCollection Citation Status PubMed-not-MEDLINE |
---|
doi: |
10.1016/j.heliyon.2024.e25363 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM368602761 |
---|
LEADER | 01000caa a22002652 4500 | ||
---|---|---|---|
001 | NLM368602761 | ||
003 | DE-627 | ||
005 | 20240220232322.0 | ||
007 | cr uuu---uuuuu | ||
008 | 240219s2024 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1016/j.heliyon.2024.e25363 |2 doi | |
028 | 5 | 2 | |a pubmed24n1300.xml |
035 | |a (DE-627)NLM368602761 | ||
035 | |a (NLM)38370214 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Kuwahara, Bruce |e verfasserin |4 aut | |
245 | 1 | 0 | |a Predicting Covid-19 pandemic waves with biologically and behaviorally informed universal differential equations |
264 | 1 | |c 2024 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Revised 20.02.2024 | ||
500 | |a published: Electronic-eCollection | ||
500 | |a Citation Status PubMed-not-MEDLINE | ||
520 | |a © 2024 The Author(s). | ||
520 | |a During the COVID-19 pandemic, it became clear that pandemic waves and population responses were locked in a mutual feedback loop in a classic example of a coupled behavior-disease system. We demonstrate for the first time that universal differential equation (UDE) models are able to extract this interplay from data. We develop a UDE model for COVID-19 and test its ability to make predictions of second pandemic waves. We find that UDEs are capable of learning coupled behavior-disease dynamics and predicting second waves in a variety of populations, provided they are supplied with learning biases describing simple assumptions about disease transmission and population response. Though not yet suitable for deployment as a policy-guiding tool, our results demonstrate potential benefits, drawbacks, and useful techniques when applying universal differential equations to coupled systems | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a Compartmental model | |
650 | 4 | |a Covid-19 | |
650 | 4 | |a Machine learning | |
650 | 4 | |a Physics informed neural networks | |
700 | 1 | |a Bauch, Chris T |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t Heliyon |d 2015 |g 10(2024), 4 vom: 29. Feb., Seite e25363 |w (DE-627)NLM255913095 |x 2405-8440 |7 nnns |
773 | 1 | 8 | |g volume:10 |g year:2024 |g number:4 |g day:29 |g month:02 |g pages:e25363 |
856 | 4 | 0 | |u http://dx.doi.org/10.1016/j.heliyon.2024.e25363 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 10 |j 2024 |e 4 |b 29 |c 02 |h e25363 |