Skip to content

Validation and evaluation of the EUREKA AI system for real-time support in surgery based on a European patient cohort

Validation and evaluation of the EUREKA AI system for real-time support in surgery based on a European patient cohort - EUREKA

Status
Recruiting
Phases
Unknown
Study type
Observational
Source
DRKS
Registry ID
DRKS00037980
Enrollment
150
Registered
2025-09-25
Start date
2025-11-06
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

Elective operations of the following entities: colorectal surgery, operation of the upper GI tract, liver surgery, pancreatic surgery

Interventions

Group 1: Retrospective analysis of video recordings of surgical procedures performed on prospectively enrolled patients with indications for elective, minimally invasive visceral surgery. The EUREKA A

Sponsors

Universitätsklinikum Ulm, Klinik für Allgemein- und Viszeralchirurgie
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to No maximum

Inclusion criteria

Inclusion criteria: Body Mass Index (BMI) = 18 and = 50 kg/m2 Indication for minimally invasive surgery

Exclusion criteria

Exclusion criteria: ASA = 4 Pregnant or breastfeeding patients Mental impairment that prevent recognition of nature, kind and extent of the trial

Design outcomes

Primary

MeasureTime frame
1. Qualitative assessment (sensitivity and specificity of the results) 2. quantitative evaluation of the intraoperatively recorded videos. Re 1. Physicians evaluate the sensitivity and specificity of the results based on a subjective, visual assessment of the prospectively recorded surgical videos. They put subjective scores to evaluate sensitivity, false positives and visual support effectiveness. The structures identified by the EUREKA AI system are validated by the surgeons. Re 2. Expert surgeons identify the area of the anatomical target structure in the prospectively recorded surgical videos. Based on that, quantitative evaluation metrics are calculated to evaluate the degree of match between the correct area and the detected area recognised by the AI. The candidates of the evaluation metrics include Recall, Precision, IoU and Dice index. All analyses take place after the operation.

Secondary

MeasureTime frame
1. Recording of measurement stability (i.e. disturbances, errors in the AI system) 2. Reliability in repeat tests All analyses take place after the operation.

Countries

Germany

Contacts

Public ContactNuh N. Rahbari

Universitätsklinikum Ulm, Klinik für Allgemein- und Viszeralchirurgie

Nuh.Rahbari@uniklinik-ulm.de+4973150053501

Outcome results

None listed

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