Real-Time AI During Pancreatoscopy for Detection of Pancreatic Neoplasia
Trial statusNot yet recruiting
Trial phaseNot listed
Trial typeObservational
Biological sexAll
Age18+
SponsorInstituto Ecuatoriano de Enfermedades Digestivas
About this trial
This prospective pilot study will evaluate the diagnostic performance of a previously validated artificial intelligence (AI) model when applied in real time during digital per-oral pancreatoscopy (POPS). The study will include adults undergoing clinically indicated pancreatoscopy for suspected or known intraductal papillary mucinous neoplasm (IPMN), indeterminate pancreatic-duct abnormalities, or preoperative assessment of IPMN extent.
During the procedure, the endoscopist will first record a visual assessment while the AI system is hidden. The AI overlay will then be activated during the same pancreatoscopy examination, and its findings will be recorded independently. AI and endoscopist assessments will be compared with a prespecified reference standard based on surgical histopathology when available or tissue sampling and clinical/imaging follow-up when surgery is not performed.
The primary objective is to estimate the sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of real-time AI for identifying high-grade dysplasia or invasive carcinoma. The study is designed as a pilot to assess feasibility and generate preliminary diagnostic-accuracy estimates for future confirmatory research.
Eligibility criteria
This trial does not accept healthy volunteersQualifiers
Adults aged 18 years or older.
Patients with a clinical indication for digital per-oral pancreatoscopy (POPS) as part of their diagnostic evaluation or preoperative assessment.
Ability to undergo digital POPS according to the treating team's clinical assessment.
Ability to provide written informed consent.
Disqualifiers
Acute pancreatitis within 2 weeks before the planned pancreatoscopy.
Hemodynamic instability or clinical condition precluding safe pancreatoscopy.
ASA physical status IV or V when the treating team determines that the procedure cannot be safely performed.
Uncorrectable coagulopathy or other contraindication to pancreatoscopy and/or tissue sampling.
Trial population
The study population will consist of adults aged 18 years or older undergoing clinically indicated digital per-oral pancreatoscopy (POPS) at the Instituto Ecuatoriano de Enfermedades Digestivas (IECED). Participants will be patients evaluated for suspected or known intraductal papillary mucinous neoplasm (IPMN), indeterminate pancreatic duct abnormalities, including strictures or filling defects, or for preoperative assessment and mapping of IPMN extent. Approximately 60 evaluable participants will be included. All participants will undergo the same prospective diagnostic assessment sequence, consisting of an initial endoscopist visual assessment followed by real-time application of the AIWorks-Cholangioscopy artificial intelligence model during the same pancreatoscopy examination. Participants will subsequently be evaluated against the prespecified reference standard based on surgical histopathology when available or tissue sampling and clinical/imaging follow-up.
Trial design
Design model
Cohort
Time perspective
Prospective
Treatments tested in this trial
AIWorks-Cholangioscopy Real-Time Artificial Intelligence Model
Diagnostic testA previously validated artificial intelligence model developed for digital cholangioscopy will be applied in real time to digital per-oral pancreatoscopy video without modification of its model weights. The system provides real-time visual detection and localization of suspected pancreatic duct abnormalities during pancreatoscopy. The AI assessment will be performed after the endoscopist has completed and locked the initial visual assessment. AI findings will be recorded as an index diagnostic test and will not independently determine tissue sampling, treatment, surgery, or other clinical management.
Treatment groups
60 Participants
are divided into 1 treatment groupGroup A: Digital POPS1 intervention
Trial outcomes
Primary outcomes
1
Patient-level diagnostic accuracy of real-time artificial intelligence for high-grade dysplasia or invasive carcinoma
Diagnostic accuracy of the real-time AI model for identifying high-grade dysplasia or invasive carcinoma at the patient level during digital per-oral pancreatoscopy. AI findings will be classified as positive or negative according to the prespecified diagnostic threshold and compared with the reference standard. The primary analysis will report sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy, each with corresponding 95% confidence intervals.
Time frame
From baseline to 6 monthsSecondary outcomes
1
Endoscopist diagnostic accuracy for high-grade dysplasia or invasive carcinoma
Diagnostic performance of the endoscopist's initial visual assessment, performed before activation of the AI overlay, for identification of high-grade dysplasia or invasive carcinoma at the patient level. Sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy will be calculated using the prespecified reference standard.
Time frame
From baseline to 6 months2
Agreement between artificial intelligence and endoscopist assessments
Concordance and discordance between the AI assessment and the endoscopist's pre-AI visual assessment for identification of suspected neoplastic lesions and high-grade dysplasia or invasive carcinoma. Agreement will be summarized using paired proportions and, when appropriate, Cohen's kappa coefficient.
Time frame
During Index procedure3
Diagnostic accuracy for IPMN epithelium
Diagnostic performance of the real-time AI model for identification of IPMN epithelium, classified as IPMN epithelium present versus absent, using the prespecified reference standard.
Time frame
From baseline to 6 months follow up4
Segment-level detection of abnormal pancreatic duct areas
Ability of the AI model to identify and localize abnormal pancreatic duct segments during digital per-oral pancreatoscopy. AI-positive segments will be compared with corresponding endoscopist assessments and available tissue or cytologic findings.
Time frame
During index procedureSponsors and contacts
Click on the lead sponsor to view all of their trials.
Source ClinicalTrials.gov
This trial is not recruiting at the moment. You can still explore other options: