AI-Based Prediction of Stage and Survival in Non-Small Cell Lung Cancer: A Retrospective Study

NCT07068139 | Active Not Recruiting

• 1 other identifier: B.30.2.ODM.0.20.08/194-318
• Study Type: observational
• Target: 150
• Locations: 0 countries
• Sites: N/A

Brief Summary

This study aims to evaluate the role of artificial intelligence (AI) in predicting disease stage and survival in patients diagnosed with non-small cell lung cancer (NSCLC). Using a retrospective design, the research will analyze radiologic imaging data (PET-CT and chest CT) and corresponding histopathological results of patients who underwent lung cancer surgery at Ondokuz Mayis University Hospital. The goal is to develop and validate a deep learning-based AI model that can automatically assess preoperative radiologic features and estimate postoperative tumor stage and survival outcomes. By integrating radiologic data with confirmed pathological diagnoses, the AI system is expected to provide clinical decision support that can improve diagnostic speed, reduce human error, and help clinicians predict prognosis more accurately. This study does not involve any experimental treatment or prospective follow-up of patients. All data will be collected from existing medical records. The findings may contribute to the digital transformation of healthcare and promote the use of AI tools in thoracic oncology.

Conditions

Non-Small Cell Lung Cancer; Artificial Intelligence (AI) in Diagnosis

Keywords

Non-Small Cell Lung Cancer; Artificial Intelligence; Deep Learning; Radiology; Thoracic Surgery

Outcome Measures

Primary Outcomes (1)

• Development of AI Model for Predicting Tumor Stage and Survival

  The primary outcome of this study is to develop and validate a deep learning-based artificial intelligence model that can predict postoperative tumor stage and survival in patients with non-small cell lung cancer using preoperative PET-CT and chest CT imaging data. The primary outcome will be considered achieved when at least 80% of the planned patient dataset (150 patients) has been successfully included and used for model development.

  From data extraction to completion of model training and validation (estimated by September 2025)

Study Arms (1)

NSCLC Surgery Cohort

This cohort includes patients who were diagnosed with non-small cell lung cancer (NSCLC) and underwent surgical treatment at Ondokuz Mayis University Hospital. Preoperative PET-CT and chest CT images and corresponding postoperative histopathological data were retrospectively collected and analyzed to develop an artificial intelligence model for predicting tumor stage and survival.

Other: AI-Based Predictive Modeling

Interventions

AI-Based Predictive Modeling OTHER

This is not a therapeutic or diagnostic intervention. The study uses a retrospective dataset of radiologic and pathological records to train and validate a deep learning model designed to predict tumor stage and survival in patients with non-small cell lung cancer (NSCLC). No experimental procedure is applied to participants.

Also known as: Deep Learning Algorithm, Retrospective Imaging Analysis

NSCLC Surgery Cohort

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

This study includes patients who underwent surgical treatment for non-small cell lung cancer (NSCLC) at Ondokuz Mayis University Hospital in Samsun, Türkiye. Eligible patients are selected from the hospital's electronic medical records and radiologic imaging archive between January 2010 and March 2025. The population reflects a clinical sample from a tertiary referral center serving a diverse adult patient population in the Black Sea region of Türkiye.

You may qualify if:

• Age ≥ 18 years
• Diagnosed with non-small cell lung cancer (NSCLC)
• Underwent surgical treatment for NSCLC at Ondokuz Mayis University Hospital
• Available preoperative PET-CT and chest CT imaging
• Available postoperative histopathological diagnosis and staging
• Signed informed consent form for data use in research

You may not qualify if:

• Age \< 18 years
• No available PET-CT or chest CT imaging in hospital records
• No available histopathological diagnosis in hospital records
• Diagnosed with a type of lung cancer other than NSCLC
• Patients who did not undergo surgery
• Patients who did not provide informed consent for retrospective data use

Sponsors & Collaborators

• Hilkat Fatih Elverdi: lead

MeSH Terms

Conditions

Carcinoma, Non-Small-Cell Lung

Condition Hierarchy (Ancestors)

Carcinoma, Bronchogenic; Bronchial Neoplasms; Lung Neoplasms; Respiratory Tract Neoplasms; Thoracic Neoplasms; Neoplasms by Site; Neoplasms; Lung Diseases; Respiratory Tract Diseases

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: SPONSOR INVESTIGATOR
• PI Title: Thoracic Surgery Resident

Study Record Dates

• First Submitted: July 7, 2025
• First Posted: July 16, 2025
• Study Start: January 1, 2010
• Primary Completion: September 1, 2025
• Study Completion: September 1, 2025
• Last Updated: August 8, 2025

Record last verified: 2025-07

Data Sharing

• IPD Sharing: Will share