AI Assisted the Diagnosis of Pancreatic Solid Lesions

Enhanced Deep Learning Model for Diagnosis of Pancreatic Solid Lesions Through Multimodal Clinical Images

Solid lesions of the pancreas mainly include tumor and non tumor lesions. More than 90% of pancreatic tumor lesions are pancreatic cancer, which is characterized by high mortality and poor prognosis and requires surgical treatment; Non-tumor lesions of the pancreas are mainly inflammatory lesions, which usually do not require surgical treatment, but can be treated with drugs. The common ones are chronic pancreatitis and autoimmune pancreatitis, with a good prognosis. Clinically, the differential diagnosis between them is very difficult. Multi-disciplinary diagnosis and treatment of MDT makes our understanding of pancreatic diseases increasingly rich and in-depth. From disease diagnosis to preoperative evaluation and curative effect evaluation, non-invasive imaging involves almost every link under MDT mode. In view of this, improving the differential diagnosis of pancreatic solid space-occupying lesions on imaging will be more conducive to the diagnosis and treatment under MDT mode, so new technologies such as artificial intelligence should be considered. Our goal is to develop a clinically applicable artificial intelligence system, which uses multiple modes to simulate the routine clinical workflow and assist in the diagnosis of benign and malignant pancreatic solid space-occupying lesions.

Study Overview

• Status: Recruiting
• Conditions: AI Assist in the Diagnosis of Pancreatic Solid Lesions
• Intervention / Treatment: Diagnostic test: Clinicians will review the suggestions of a hypothetical AI

Detailed Description

The diagnosis of solid pancreatic lesions is challenging, MDT is a very effective method, but it has a certain misdiagnosis rate. This is a multi-center, prospective and observational clinical study. Our goal is to develop a clinically applicable artificial intelligence system. On the one hand, our artificial intelligence based on clinical data+CT imaging images can assist MDT doctors to diagnose the nature of pancreatic space-occupying lesions and reduce misdiagnosis; On the other hand, if a patient needs EUS-FNA puncture, the multimodal artificial intelligence system based on clinical data+CT+EUS developed by us can help MDT doctors understand the nature of pancreatic space-occupying lesions and reduce the probability of misdiagnosis or secondary puncture.

• Study Type: Observational
• Enrollment (Anticipated): 200

Contacts and Locations

Study Locations

• China
  • Shanghai
    • Shanghai, Shanghai, China, 200433
      • Recruiting
      • CT and EUS
      • Contact: Zhaoshen Li; Phone Number: 86-21-31162338; Email: changhaiec@126.com

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: 18 years to 75 years (ADULT, OLDER_ADULT)
• Accepts Healthy Volunteers: N/A
• Genders Eligible for Study: All

• Sampling Method: Non-Probability Sample

Study Population

Patients who had pancreatic solid mass will be enrolled in our study.

Description

Inclusion Criteria:

• pancreatic solid mass in CT and EUS

Exclusion Criteria:

• insufficient imaging quality of CT or EUS
• endoscopic ultrasound non accessible lesions

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
patients with solid lesions of pancreas	Diagnostic test: Clinicians will review the suggestions of a hypothetical AI; There is no intervention. Clinicians will review the suggestions of a hypothetical AI

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Researchers use artificial intelligence (AI) support system to assist in diagnosis of pancreatic solid space-occupying lesions	A multi-layer screening deep convolution network based on deep convolution network was developed to observe its accuracy, sensitivity and specificity in assisting MDT doctors to identify benign and malignant pancreatic space-occupying lesions.	2 months

Collaborators and Investigators

• Sponsor: Changhai Hospital
• Collaborators: Sun Yat-sen University; Ruijin Hospital; Wuhan Union Hospital, China; The Third People's Hospital of Chengdu

Study record dates

Study Major Dates

• Study Start (ANTICIPATED): January 21, 2023
• Primary Completion (ANTICIPATED): February 21, 2023
• Study Completion (ANTICIPATED): February 21, 2023

Study Registration Dates

• First Submitted: January 21, 2023
• First Submitted That Met QC Criteria: January 21, 2023
• First Posted (ACTUAL): January 31, 2023

Study Record Updates

• Last Update Posted (ACTUAL): January 31, 2023
• Last Update Submitted That Met QC Criteria: January 21, 2023
• Last Verified: January 1, 2023

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Pathologic Processes; Disease
• Other Study ID Numbers: 2022-AI and MDT

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No