Overview
Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive cancer with a very low survival rate due to its covert onset and low early diagnosis rate. This study uses a pancreatic imaging AI model to improve early detection and high-risk monitoring of pancreatic cancer through non-contrast CT scans. The goal is to validate the AI model's diagnostic performance, particularly in identifying early-stage resectable PDAC.
Description
PDAC is one of the most aggressive malignancies, with a dismal 5-year survival rate of merely 13%. The poor prognosis is primarily due to its insidious onset and low early diagnosis rate. Clinical studies have shown that 51% of pancreatic cancer patients present with distant metastasis at the time of diagnosis, with a 5-year survival rate of less than 5%; in contrast, patients at stage IA demonstrate a clear therapeutic benefit, with a postoperative 5-year survival rate of up to 83.7%. Early detection of pancreatic cancer is crucial for improving survival rates. In this study, investigators will utilize our developed pancreatic imaging AI model to conduct prospective clinical trials, aiming to validate the model's diagnostic performance in real-world applications, particularly its efficacy in detecting early-stage PDAC.
This study is led by Jin Gang, Director of Department of General Surgery of Shanghai Changhai Hospital, in collaboration with Yinzhou Hospital Affiliated to Medical School of Ningbo University, The Second Affiliated Hospital of Jiaxing University, The Central Hospital of Lishui City, Jingning County People's Hospital, Meinian Onehealth Health Examination Center, and the Medical AI Imaging Laboratory Team at Alibaba DAMO Academy. This is a translational study focused on early detection and high-risk monitoring of pancreatic cancer based on artificial intelligence, and it will directly impact clinical care pathways. The study population consists of individuals who have undergone non-contrast abdominal or chest CT scans at medical institutions or health examination centers, defined as an opportunistic screening population. The pancreatic imaging AI model will be used to automatically detect pancreatic lesions, including PDAC and non-PDAC subtypes. Subjects identified with positive lesions by the AI model will be required to undergo questionnaire, blood tests (including tumor marker analysis) and further imaging examinations (such as contrast CT, MRI, and EUS) to confirm the final pancreatic lesion status and formulate a treatment plan. The ultimate aim is to assess the AI model's diagnostic performance, particularly its detection rate for early resectable pancreatic cancer.
Eligibility
Inclusion Criteria:
- 1.The subject has undergone an abdominal or chest non-contrast CT scan, categorized
as follows:
- Physical exam population
- Outpatient population
- Inpatient population
- Emergency department population
Exclusion Criteria:
- 1.The subject has been diagnosed with pancreatic cancer or other malignant tumors within the last 5 years.
- 2.The subject has any medical condition that contraindicates the use of contrast in MRI/CT scans.
- 3.The subject cannot be followed up or is participating in other clinical trials.