Key features of the GeoSentinel database and their modelling benefits for outbreak detection
Feature . | Description . | Modelling benefits . |
---|---|---|
Reliability | Use of validated diagnostic testing leads to high-quality data. | Reliable output depends on valid input. Misinformation on social media,15 increasing the need for trusted sources. |
Scalability | Large amount of global, historical data (since 1995) and a growing network. | Labelled data allow for supervised learning.10, Sustained growth allows for powerful deep anomaly detection methods.16 |
Dimensionality | Variety of data collected (e.g. patient demographic information, travel history, reason for travel). | Potential for multivariate monitoring methods,7 spatial–temporal monitoring7 and high-risk subgroup identification.1 |
Timeliness | Sites incentivized to promptly enter records, leading to rapid alerts.11,12 | Automated monitoring of travel-related health data in real-time or at high frequency enables efficient insights and alerts. |
Feature . | Description . | Modelling benefits . |
---|---|---|
Reliability | Use of validated diagnostic testing leads to high-quality data. | Reliable output depends on valid input. Misinformation on social media,15 increasing the need for trusted sources. |
Scalability | Large amount of global, historical data (since 1995) and a growing network. | Labelled data allow for supervised learning.10, Sustained growth allows for powerful deep anomaly detection methods.16 |
Dimensionality | Variety of data collected (e.g. patient demographic information, travel history, reason for travel). | Potential for multivariate monitoring methods,7 spatial–temporal monitoring7 and high-risk subgroup identification.1 |
Timeliness | Sites incentivized to promptly enter records, leading to rapid alerts.11,12 | Automated monitoring of travel-related health data in real-time or at high frequency enables efficient insights and alerts. |
Key features of the GeoSentinel database and their modelling benefits for outbreak detection
Feature . | Description . | Modelling benefits . |
---|---|---|
Reliability | Use of validated diagnostic testing leads to high-quality data. | Reliable output depends on valid input. Misinformation on social media,15 increasing the need for trusted sources. |
Scalability | Large amount of global, historical data (since 1995) and a growing network. | Labelled data allow for supervised learning.10, Sustained growth allows for powerful deep anomaly detection methods.16 |
Dimensionality | Variety of data collected (e.g. patient demographic information, travel history, reason for travel). | Potential for multivariate monitoring methods,7 spatial–temporal monitoring7 and high-risk subgroup identification.1 |
Timeliness | Sites incentivized to promptly enter records, leading to rapid alerts.11,12 | Automated monitoring of travel-related health data in real-time or at high frequency enables efficient insights and alerts. |
Feature . | Description . | Modelling benefits . |
---|---|---|
Reliability | Use of validated diagnostic testing leads to high-quality data. | Reliable output depends on valid input. Misinformation on social media,15 increasing the need for trusted sources. |
Scalability | Large amount of global, historical data (since 1995) and a growing network. | Labelled data allow for supervised learning.10, Sustained growth allows for powerful deep anomaly detection methods.16 |
Dimensionality | Variety of data collected (e.g. patient demographic information, travel history, reason for travel). | Potential for multivariate monitoring methods,7 spatial–temporal monitoring7 and high-risk subgroup identification.1 |
Timeliness | Sites incentivized to promptly enter records, leading to rapid alerts.11,12 | Automated monitoring of travel-related health data in real-time or at high frequency enables efficient insights and alerts. |
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.