To Introduce Artificial Intelligence (AI) and Machine Learning in Cardiotocography (CTG) Interpretation to Improve Clinical Use
Conditions
Brief summary
The project leaders plan to create a clinical decision support (CDS) system by programming a self-learning software to analyze the cardiotocography (CTG) traces in the - already existing - database from the maternity department of the Inselspital Berne. The project leaders will process and analyze all clinical outcomes of the estimated 10000-15000 eligible patient records. CSEM will design, develop, and validate several AI architectures with the intend to create the CDS system. The AI would learn to assist on this task by training machine learning (ML) algorithms. The main purpose of the AI-CDS will be to determine the best fetal extraction moment during labor, based on a self-learning approach, as a superhuman support tool for obstetricians in decision making during labor.
Interventions
None listed
Sponsors
CSEM Centre Suisse d'Electronique et de Microtechnique SA - Recherche et Developpement
Insel Gruppe AG, University Hospital Bern
Study design
Observational model
COHORT
Time perspective
RETROSPECTIVE
Eligibility
Sex/Gender
FEMALE
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* CTG-registrations of patients with singleton pregnancies during labour from 01.01.2006 to 31.12.2019 * Gestational age ≥ 24+0 weeks * Age ≥ 18 years * Written informed consent
Exclusion criteria
* Documented refusal * Multiple pregnancies * CTG-registrations of planned caesarean sections
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Superior prediction of fetal morbidity through the self-learning CDS system than if performed by obstetricians alone, especially in regards to specificity. | 3 months |
Countries
Switzerland
Contacts
Primary ContactAnda Radan
Backup ContactKarin Strahm
Outcome results
None listed