Artificial Intelligence in Lung Cancer Screening

Completed

• Conditions: Lung Cancer Screening; Asbestos

• Registration Number: NCT06444373
• Lead Sponsor: Scientific Institute San Raffaele

Brief Summary

Single-center, non-profit, observational, retrospective study of collection of clinical and amnestic data and images to create, implement and develop a pilot model of an integrated virtual platform.

Detailed Description

The project we propose is a study whose objective was to develop an artificial intelligence program integrated into a web-based platform for the optimization of the performance of lung cancer screening for the diagnosis of lung nodules and risk stratification in subjects exposed to environmental carcinogens and/or cigarette smoke.

Inclusion criteria:

Age \> 50; smokers for at least 20 pack-years (20 cigarillos a day for 20 years) or former heavy smokers if they quit less than 15 years ago; and/or previous professional exposure to asbestos; absence of lung cancer symptoms; who performed lung cancer screening after the year 2000 upon approval of the study by the relevant EC.

Study Timeline

• Study Start: 2022-12-02
• Status Verified: 2024-05
• Primary Completion: 2024-05-22
• Study Completion: 2024-05-30
• Study First Submitted: 2024-05-30
• Study First Submitted QC: 2024-05-30
• Last Update Submitted: 2024-05-30
• Study First Posted: 2024-06-05
• Last Update Posted: 2024-06-05

Recruitment & Eligibility

• Status: COMPLETED
• Sex: All
• Target Recruitment: 728

Inclusion Criteria

• Age > 50 years;
• smokers for at least 20 pack-years (20 cigarettes a day for 20 years) or former heavy smokers if they quit less than 15 years ago;
• and/or previous professional exposure to asbestos;
• absence of lung cancer symptoms;
• who performed lung cancer screening after the year 2000 upon approval of the study by the relevant Etical Committee

Exclusion Criteria

• Age < 50 years
• never smokers
• lung cancer symptoms

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
AIM 1 Pilot deep learning model	from enrollment to the end of treatment at 2 years	Development and fine-tuning of a pilot deep learning model for automatic detection and diagnosis of screen-detected nodules for risk stratification in subjects with asbestos exposure as part of a lung cancer screening program in high-risk subjects for exposure to asbestos and smoking on retrospective data.

Secondary Outcome Measures

Name	Time	Method
AIM 2 Clinical database	from enrollment to the end of treatment at 2 years	Development of an integrated system between the clinical database and several existing imaging volumetric software and risk models for the creation of a pilot platform in order to optimize the organizational management of lung cancer screening.

Trial Locations

Locations (1)

IRCCS San Raffaele Scientific Institute; 🇮🇹 Milan, Italy