RecruitingNCT05375591

AI & Radiomics for Stratification of Lung Nodules After Radically Treated Cancer

Artificial Intelligence & Radiomics for Stratification Of Lung Nodules After Radically Treated Cancer (AI-SONAR)

Sponsor

Royal Marsden NHS Foundation Trust

Enrollment

1,000 participants

Start Date

Oct 13, 2021

Study Type

OBSERVATIONAL

Conditions

Lung Cancer Lung Metastases Indeterminate Pulmonary Nodules Second Primary Cancer

Summary

This study will assess the utility of radiomics and artificial intelligence approaches to new lung nodules in patients who have undergone radical treatment for a previous cancer.

Eligibility

Min Age: 18 Years

Plain Language Summary

Simplified for easier understanding

This study is using artificial intelligence and a technique called radiomics — which finds patterns in CT scan images — to better determine whether a new spot on the lung is cancer or something harmless, in people who have previously had cancer. **You may be eligible if...** - You previously had a solid organ cancer that was treated with the aim of a cure - A new spot (nodule) was found on your lung during a recent chest CT scan - Your previous cancer was confirmed by your care team and successfully treated **You may NOT be eligible if...** - Your new lung spot has already been confirmed as benign or does not need further assessment - You have never had a prior cancer diagnosis Talk to your doctor to see if this trial is right for you.

This summary was AI-generated to explain the trial in plain language. It is not medical advice. Always discuss eligibility with your doctor before enrolling in a clinical trial.

Interested in this trial?

Get notified about updates and connect with the research team.

Interventions

OTHERNon-Interventional Study

First nodule detection CT scans as per eligibility criteria will be used as input into in-house software to extract multiple radiomic features and used to develop a machine learning based classifier to differentiate nodule aetiology. Scans will also be used as input in to a deep learning/convolutional neural network models to perform automated imaging classification.