CarefulAI
CEOs Feed
  • About Us
  • AI Assurance
  • AI Model Design
  • Motives and Models
  • Model Feedback

Service User / Clinical Validation of the SISDA Phrase Library

SISDA aims to enable clinicians to be alerted to suicidal ideation language in text submitted the public, using NHS forms and Tools.

Background on SISDA

For background on SISDA please see its model label under 'Motives and Models'

Background on the SISDA data set derived from NLP and User Group Research

Signal Source: text input by a member of the public in a NHS Form / Tool
Signal Type: a phrase
What a Signal Triggers: a person being engaged more quickly by a mental health clinician
The Intent to be Validated: suicide ideation

Validators have the options of labelling  each signal as one of the following:
  • To be Validated
  • Evidence of possible Intent
  • Campaigns relating to the Intent
  • Flippant Reference to the Intent
  • Information or Support for the Intent
  • Memorial or Condolence for the Intent
  • Reporting of the Intent
  • None of the above

For example:
Picture
The purpose of validating the SISDA library in this manner is to inform those people who may wish to deploy SISDA to gather evidence for their DCB0129/160 documentation

To register your interest in validating SISDA within a specific user group please make an enquiry via 'Contact us' ​

London Office

What Our Clients Are Saying

''I love your mantra.  AI is Artificial Ignorance until it is validated by people for people.' NHSE Director

'You deliver technically complex subjects beautifully' NHSE Clinician

'Federated AI powered by a mobile phone, creating insights that are useful to the user, which they then control .  Brilliant.' NHSE Clinician

Picture