Detection and Automatic Segmentation of Liver Nodules in Patients With Colorectal Adenocarcinoma

• Conditions: Detection of Liver Metastases

• Registration Number: NCT04834596
• Lead Sponsor: University Hospital, Angers

Brief Summary

Evaluation of the diagnostic performance of artificial intelligence-based software for the detection of liver nodules on scanner in patients at risk of liver metastases of colorectal cancer

Detailed Description

The aim of this study is to evaluate the performance of a software for liver nodule detection in patients with colorectal cancer compared to the consensus of two radiologists analysis.

300 hepatic scanners will be analysed.

Study Timeline

• Status Verified: 2021-04
• Study First Submitted: 2021-04-02
• Study First Submitted QC: 2021-04-06
• Last Update Submitted: 2021-04-06
• Study First Posted: 2021-04-08
• Last Update Posted: 2021-04-08
• Study Start: 2021-05-02
• Primary Completion: 2021-12-01
• Study Completion: 2022-05-02

Recruitment & Eligibility

• Status: UNKNOWN
• Sex: All
• Target Recruitment: 300

Inclusion Criteria

• Patients consecutively treated for colorectal cancer a
• Availability of imaging by injected abdominal scanner including at least one portal phase
• Age over 18

Exclusion Criteria

• Patients with treatment of the hepatic nodule by ablation or previous hepatic surgery
• Imaging follow-up <12 months
• Non-interpretable scanner (artefact, injection quality, etc.)
• Patients with more than 10 liver metastases

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
To evaluate the performance of the software for detection of hepatic nodules on CT scan of patients with colorectal cancer compared to a reference constituted by the evaluation established by two radiologists in consensus.	1 month	Number of nodules detected by the software compared to the reference method

Trial Locations

Locations (1)

UHAngers; 🇫🇷 Angers, France