Event Types

[jones139]

The database is based around the concept of Events, where an event is either a genuine seizure (which may or may not have been detected by OpenSeizureDetector), or a false alarm. A number of data points are stored that cover the period before and after the event to allow the event to be analysed to improve the reliability of the seizure detector algorithms.
The OpenSeizureDetector Android App will automatically create and upload Events for data that have generated WARNING, ALARM or FALL notifications from OpenSeizureDetector, or where the user has added a seizure manually using the 'Report Seizure' button in the app.
It is important that users review the events that they have generated to classify them so that developers can use the data to improve the detection system.
It is proposed that each Event is given a Type and a Sub-Type to make it as easy as possible for users to select the most appropriate classification. A free form text field will also be provided for the user to provide additional information.
There will need to be some iteration to determine the most useful types and sub-types based on user and developer feedback. The initial proposal for discussion is:
Event Types

1. Seizure
2. Fall
3. Other Medical Issue
4. False Alarm

Event Sub-Types

• Seizure: Tonic-Clonic
• Seizure: Other (please describe in notes)
• Fall: Uncontrolled (subject unconscious)
• Fall: Controlled (subject attempted to save him/herself)
• Fall: Stumble (not a complete fall)
• Fall: Other (please describe in notes)
• Other Medical Issue: Fever/shaking
• Other Medical Issue: Cold / shivering
• Other Medical Issue: Other (please describe in notes)
• False Alarm: Brushing Teeth
• False Alarm: Brushing Hair
• False Alarm: Typing
• False Alarm: .......
• False Alarm: Other (please describe in notes)

Please suggest changes and additional false alarm categories.