Skip to content

Improving Safety and Efficacy of Endoscopic Procedures – A Deep Machine Learning based Depth of Sedation Monitor

Improving Safety and Efficacy of Endoscopic Procedures – A Deep Machine Learning based Depth of Sedation Monitor

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
DRKS
Registry ID
DRKS00016605
Enrollment
1000
Registered
2019-05-07
Start date
2019-06-04
Completion date
Unknown
Last updated
2026-06-01

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

GI and bronchial endoscopy

Interventions

Group 1: During endoscopic procedures vital signs plus a 2-lead frontotemporal EEG will be recorded in real-time as well as clinical sedation depth according to validated scales and state of conscious

Sponsors

Universitätsklinikum Halle (Saale)
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to No maximum

Inclusion criteria

Inclusion criteria: 1) Adult males or females capable of giving consent. 2) Planned procedure is of elective nature and in endoscopy Department. 3) Anticipated Duration of procedure is > 20 minutes. 4) Sedation Regimen is NAPS.

Exclusion criteria

Exclusion criteria: 1) Impaired Hearing. 2) Cognitive (GCS =14 Points) or acute psychiatric alteration. 3) Patients with structural brain disease, cerebral metastasis or known cerebral epilepsy. Also, mental alteration due to newly prescribed medication. 4) Sedation with substances different than outlined in protocol. 5) Allergy or intolerance to Propofol and it's components.

Design outcomes

Primary

MeasureTime frame
There is a shared primary endpoint: A primary endpoint is the prognostic accuracy of the statistical model to separate consciousness from unconsciousness, whereby - vital signs and processed EEG-parameters may be included in the model, - the concrete set of input variables will be based on preliminary analysis of study data, - the prognostic accuracy of the model is defined by its test characteristics (Sn, Sp, PPV, NPV) and its AUC of the Reciever-Operating-Curve, - the model will be validated on a sufficiently large cohort. A primary endpoint is the prognostic accuracy in differentiating different levels of sedation according to the MOAA/S-Scale, whereby the points mentioned above also apply.

Secondary

MeasureTime frame
Secondary endpoints are the prognostic accuracies of single processed EEG-Parameters in regard to state of consciousness and clinical sedation depth as outlined in the primary endpoint. These include • EEG-parameters of the frequency domain (e.g. TP, MF, SEF90, SEF95, PPF, WSMF and power in a-, ß-, d-, ?- and ?-frequency bands), • EEG-parameters of the time-frequency domain (e.g. bispectrum, bicoherence, phase synchronization) and • measures of complexity as well as non-linear parameters (e.g. ApEn, PeEn, Order Recurrence Rate and Order Phase Coupling).

Countries

Germany

Contacts

Public ContactJakob Garbe

Klinik für Innere Medizin I, Universitätsklinikum Halle

jakob.garbe@uk-halle.de0345 557 2928

Outcome results

None listed

Source: DRKS (via WHO ICTRP) · Data processed: Jun 11, 2026