Encephalopathy
Conditions
Keywords
Full Term Newborn, EEG, HIE
Brief summary
The project aims at designing a machine learning solution able to recognize characteristics signals patterns of brain damages in full term babies born within a context of Hypoxic Ischemic Encephalopathy (HIE)
Detailed description
Retrospective study based on a digital EEG signal library intending to design, train and test an efficient AI solution for hypothermia protocol start indications. The output of the Project is to make available to pediatric resuscitation units an adequate tool to guide them in the decision of hypothermia protocol start in a general context of neurophysiologist competence scarcity. EEG signal that would allow the algorithm design will be based on several parameters of the conventional EEG and not only on signal amplitude
Interventions
OTHERDecision of hypothermia
Decision of hypothermia thanks to EEG signal
Sponsors
Assistance Publique - Hôpitaux de Paris
Study design
Observational model
COHORT
Time perspective
RETROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
No minimum to 2 Days
Healthy volunteers
No
Inclusion criteria
* Full term (\> 36 weeks) * HIE context * EEG recording before 6 hours of life
Exclusion criteria
-Opposition of parental authority holders of a patient born after 2015
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Decision of hypothermia protocol start | 6 months | Use of EEG signal in order to develop an algorithm |
Countries
France
Outcome results
None listed