Skip to content

Real-time Computer Aided Detection of Barrett's Neoplasia

Real-time Computer Aided Detection of Barrett's Neoplasia: a Pilot Study

Status
Completed
Phases
Unknown
Study type
Observational
Source
ClinicalTrials.gov
Registry ID
NCT05628441
Enrollment
30
Registered
2022-11-28
Start date
2022-05-16
Completion date
2022-10-13
Last updated
2022-11-28

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

Conditions

Barrett Adenocarcinoma, Barrett Esophagus

Keywords

Barrett's Esophagus, Computer Aided Detection

Brief summary

In the real-time pilot study, the performance of a recently developed Computer Aided Detection (CADe) system will be investigated in 30 patients during real-time live endoscopic examination. Three expert endoscopists in the field of Barrett's esophagus and endoscopic resection techniques will perform the inclusions following a standardized data acquisition protocol. In total, 15 patients with a visible neoplastic lesion in their Barrett's esophagus and 15 patients without any visible abnormalities in their Barrett's esophagus will be included in this study. Outcomes: to investigate feasibility of the use of the CADe system in the endoscopy suite, to investigate the performance of the CADe system when used by endoscopists.

Detailed description

Patients in this study will undergo a scheduled surveillance endoscopy (according to current guidelines; n = 15) or are referred to a tertiary referral center for endoscopic resection of a possible neoplastic area in their known Barrett's esophagus (n = 15). Patients will be consciously sedated (standard care) before start of the procedure. During the endoscopic procedure, the endoscopist follows the standard protocol after extensive cleaning of the Barrett's segment: first, a standardized pullback video will be recorded, followed by a careful inspection of the Barrett segment every 2 centimeters. Per 2cm, a video of 10 seconds and 3 overview images are being recorded. During the imaging protocol, the CADe system presents its prediction on a second screen, next to the main endoscopy monitor. After the imaging protocol, standard clinical care is being resumed.

Interventions

OTHERComputer Aided Detection system for Barrett's neoplasia

CADe system running in the back, not directly interfering with patient care

Sponsors

Academisch Medisch Centrum - Universiteit van Amsterdam (AMC-UvA)
Lead SponsorOTHER

Study design

Observational model
COHORT
Time perspective
PROSPECTIVE

Eligibility

Sex/Gender
ALL
Age
18 Years to 99 Years
Healthy volunteers
Yes

Inclusion criteria

* Age \> 18 years; * Patients under surveillance for their diagnosed non-dysplastic Barrett's esophagus or patients refered for endoscopic treatment because of proven high-grade dysplasia or adenocarcinoma; * A circumferential Barrett length of at least 2cm

Exclusion criteria

* Medical history of prior surgical or endoscopical treatment for neoplasia in their esophagus * Reflux esophagitis \> grade 2 (LA classification) * Inability to undergo endoscopic treatment and/or biopsies

Design outcomes

Primary

MeasureTime frameDescription
Combined performance of endoscopist + CADe system in terms of per-patient sensitivity and specificity5 monthsTo investigate the combined performance between endoscopist and CADe system on a per-patient level.

Secondary

MeasureTime frameDescription
Stand-alone performance of CADe system in terms of per-patient sensitivity and specificity5 monthsTo investigate the performance of the CADe system on new data on a per-patient level.

Countries

Netherlands

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: Feb 8, 2026