Research of the Application of Pancreatic Cancer Screening Artificial Intelligence Model "PANDA PLUS"

Trial Title: Research of the Application of Pancreatic Cancer Screening Artificial Intelligence Model "PANDA PLUS"

NCT ID: NCT06643715

Condition: Pancreatic Cancer

Conditions: Official terms:
Pancreatic Neoplasms

Conditions: Keywords:
Artificial Intelligence
Pancreatic Cancer
PANDA PLUS
PANDA

Study type: Interventional

Study phase: N/A

Overall status: Not yet recruiting

Study design:

Allocation: Randomized

Intervention model: Parallel Assignment

Intervention model description: The PANDA (Pancreatic Cancer Detection with Artificial Intelligence) early screening model was developed by Alibaba DAMO Academy.It employs a fully automated pancreatic segmentation model using the VoxelMorph algorithm, which enhances registration speed and accuracy. By building a self-learning framework, the model efficiently obtains precise annotations of non-contrast CT images. The core network is based on Transformer technology, combined with the MaskFormer model to improve diagnostic accuracy. The upgraded PANDA PLUS model has improved upon the original PANDA model by enhancing its ability to differentiate between pancreatitis, pancreatic cystic lesions, and eliminating interference from adjacent organs such as the common bile duct and duodenum. These advancements have effectively increased the model's clinical utility, making it more reliable for real-world applications in diagnosing pancreatic conditions.

Primary purpose: Diagnostic

Masking: None (Open Label)

Intervention:

Intervention type: Device
Intervention name: PANDA PLUS
Description: Using the PANDA PLUS model to assist in image interpretation, radiologists are provided with the model's output classification-non-PDAC, PDAC, or normal-alongside the initial CT image assessment. The radiologists then integrate the information from both the PANDA PLUS output and the original image interpretation to formulate the final report.
Arm group label: PANDA PLUS

Summary: This study aims to utilize large-scale real-time CT（Computed Tomography） imaging data from multiple centers and scenarios, including the First Affiliated Hospital of Zhejiang University, Lishui Central Hospital, and the Second Affiliated Hospital of Nanchang University, within a prospective real-world cohort. Radiologists will be stratified by years of experience and randomly assigned into two groups. The experimental group will use PANDA（Pancreatic Cancer Detection with Artificial Intelligence） Plus results to assist in generating imaging reports. The study will record the radiologists' interpretations before and after reviewing the PANDA Plus reports and will track patients with imaging reports suggesting suspicion of PDAC（Pancreatic ductal adenocarcinoma） positivity from both groups. Follow-up will be conducted to verify whether patients with negative imaging reports in both groups develop pancreatic cancer. The study will assess the benefit of PANDA Plus in assisting clinical radiologists in diagnosing PDAC under non-contrast CT conditions. It will compare the sensitivity, specificity, and positive predictive value of PDAC detection with and without the assistance of the PANDA Plus model, as well as differences in the initial TNM staging, resectability rates, and long-term survival data.

Detailed description: This study aims to utilize large-scale real-time imaging data from multiple centers, including the First Affiliated Hospital of Zhejiang University, Lishui Central Hospital, and the Second Affiliated Hospital of Nanchang University, within a prospective real-world cohort. Radiologists will be stratified by years of experience and randomly assigned into two groups. One group will use the pancreatic cancer screening model, PANDA PLUS, while the other will use a pseudo-PANDA model to assist in diagnosis during routine clinical practice. Both groups will be tracked for cases with imaging reports suggesting possible PDAC positivity, and follow-up will verify the final outcomes of patients with negative imaging reports. Patients with positive reports will receive regular follow-up every three months to obtain clinical history, pathological gold standards, initial tumor markers, resectability classification, TNM staging, and long-term survival data. The study will evaluate the effectiveness of PANDA PLUS in assisting clinical radiologists in diagnosing PDAC under non-contrast CT conditions and compare the sensitivity, specificity, and positive predictive value of PDAC detection between the pseudo-PANDA and PANDA PLUS models. Additionally, it will compare differences in initial TNM staging, resectability rates, and long-term survival data. The study will also assess the detection efficiency of PANDA PLUS for triple-negative PDAC (negative clinical symptoms, negative tumor markers, and negative non-contrast CT imaging).

Criteria for eligibility:
Criteria:
Inclusion Criteria: - Subjects who have undergone chest and/or abdominal non-contrast CT scans at outpatient clinics, inpatient departments, or physical examination centers; - Age at the time of the scan between 18-90 years old, with no restriction on gender; Exclusion Criteria: - Chest CT scans that do not cover the pancreas; - Non-contrast CT scans performed in emergency settings; - Patients who have undergone thoracic/abdominal surgeries affecting or altering the anatomical display of the pancreas (e.g., post-esophageal, gastric, pancreatic, vascular surgeries, or post-ERCP); - Non-standard scans (e.g., hands placed on either side of the body or abdomen, severe respiratory motion artifacts, perfusion contamination, etc.); - CT scans ordered by hepatobiliary and pancreatic surgeons or oncologists; - Patients referred to a higher-level hospital due to a pancreatic mass found during local hospital examination; - Patients who, for personal reasons, did not follow up with pancreatic cancer diagnosis or treatment at the hospital, or were lost to follow-up midway; - Patients with concurrent malignancies in other locations or those undergoing comprehensive cancer treatment for malignant tumors; - Imaging reports made by radiologists without referring to AI during the image interpretation; - Patients who underwent enhanced CT, MRI, or PET-CT examinations concurrently.

Gender: All

Minimum age: 18 Years

Maximum age: 90 Years

Healthy volunteers: Accepts Healthy Volunteers

Locations:

Facility:
Name: the First Affiliated Hospital, School of Medicine, Zhejiang University

Address:
City: Hangzhou
Zip: 310000
Country: China

Start date: November 1, 2024

Completion date: October 31, 2026

Lead sponsor:
Agency: Zhejiang University
Agency class: Other

Collaborator:
Agency: Lishui Municipal Central Hospital
Agency class: Other

Collaborator:
Agency: Second Affiliated Hospital of Nanchang University
Agency class: Other

Source: Zhejiang University

Record processing date: ClinicalTrials.gov processed this data on November 12, 2024

Source: ClinicalTrials.gov page: https://clinicaltrials.gov/ct2/show/NCT06643715