Research progress in computer-aided diagnosis systems for lung cancer

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Oncology

Background:

• Lung cancer is a leading cause of cancer mortality, necessitating improved diagnostic tools.
• Accurate and timely diagnosis is crucial for effective lung cancer treatment and patient survival.

Purpose of the Study:

• To review and synthesize advances in computer-aided diagnosis (CAD) for lung cancer.
• To highlight clinically relevant and bedside-proven AI strategies in lung cancer detection and prognosis.

Main Methods:

• Review of computer-aided diagnosis across classical imaging, machine learning, and deep learning.
• Emphasis on multimodal CT/PET-clinical fusion, small-data strategies, interpretable AI, and privacy-preserving multi-center learning.

Main Results:

• AI systems achieve high accuracy (AUC ≥ 0.95) with low false positives (<0.1/CT).
• AI demonstrates significant potential in boosting early lung cancer detection (~20-30%) and improving prognostic accuracy (C-index ~0.85-0.90).

Conclusions:

• Clinically validated AI advances offer significant potential for lung cancer diagnosis and prognosis.
• Implementation strategies are needed to translate AI accuracy into tangible patient benefits.