Automated Lung Cancer Detection Using Artificial Intelligence (AI) Deep Convolutional Neural Networks: A Narrative...

Area of Science:

• Medical Imaging
• Artificial Intelligence in Oncology
• Radiology

Background:

• Lung cancer is a leading cause of mortality globally.
• Physician workload and burnout are significant challenges in radiology.
• Accurate lung nodule detection is crucial for early cancer diagnosis.

Purpose of the Study:

• To compare the performance of four artificial intelligence (AI) models for lung nodule cancer detection.
• To evaluate AI model performance against radiologist accuracy.
• To explore AI's potential in alleviating radiologist workload.

Main Methods:

• A narrative literature review of 648 selected articles (2008-2019).
• Inclusion criteria focused on CT chest scans, lung nodules, lung cancer, deep learning, ensemble, and classic AI methods in adults aged 18-65.
• Performance metrics included sensitivity, specificity, accuracy, ROC curve, and AUC.

Main Results:

• A hybrid deep-learning model demonstrated state-of-the-art performance.
• The model achieved high accuracy with a low false-positive rate.
• AI models show comparable or superior performance to human readers in specific tasks.

Conclusions:

• AI models, particularly hybrid deep-learning approaches, offer promising results for lung nodule detection.
• Automated physician-assist systems can support radiologists, potentially improving diagnostic accuracy and workflow efficiency.
• Further in-depth comparative studies of AI model accuracy are essential for clinical implementation.