WCLC 2025 - Posters & ePosters-EP.04.43 Chest X-Ray Analysis With Artificial Intelligence Software Aids in the Early Diagnosis of Lung Cancer- a Case Series

EP.04.43 Chest X-Ray Analysis With Artificial Intelligence Software Aids in the Early Diagnosis of Lung Cancer- a Case Series

Back to course

Pdf Summary

This case series from Hacettepe University, Turkey, demonstrates the utility of artificial intelligence (AI)-enabled software applied to routine chest X-rays (CXRs) in the early detection of lung cancer. Lung cancer is the leading cause of cancer mortality globally and highly prevalent in Turkey, especially among males. Early diagnosis significantly improves prognosis, but widespread use of low-dose computed tomography (LDCT) screening remains limited. As chest X-rays are commonly performed and pulmonary nodules are often missed, the use of AI software capable of identifying high-risk nodules offers potential benefits.<br /><br />From February to November 2024, the AI tool qXR analyzed all CXRs prospectively, generating a Lung Nodule Malignancy Score (LNMS) to categorize nodules by risk. Radiologists and pulmonologists reviewed flagged nodules, with further evaluation by LDCT using Lung-RADS criteria. The report highlights three cases where AI-enabled CXR detected incidental nodules in patients undergoing imaging for unrelated medical issues: pre-anti-TNF therapy screening, fever evaluation in the emergency room, and smoking cessation counseling. In each case, the AI flagged suspicious nodules that were confirmed as early-stage lung cancers by biopsy (stages IA2 to IIB). Early detection facilitated timely surgical treatment with favorable outcomes.<br /><br />The study concludes that AI-enabled CXR interpretation can improve incidental detection of potentially malignant pulmonary nodules, particularly where radiologist availability is limited or workloads are high. This approach may be especially valuable in countries lacking formal lung cancer screening programs, helping to identify high-risk nodules earlier and enhance patient outcomes. The findings suggest AI-integrated CXR could be a cost-effective tool aiding early lung cancer diagnosis in routine clinical practice.

Asset Subtitle

Deniz Koksal

Meta Tag

Speaker Deniz Koksal

Topic Screening and Early Detection

Keywords

Artificial Intelligence

Chest X-Ray

Lung Cancer

Early Detection

Pulmonary Nodules

Low-Dose Computed Tomography

Lung Nodule Malignancy Score

Lung-RADS

Hacettepe University

Incidental Findings