Lung cancer diagnosis from CT scans using artificial intelligence techniques: A global perspective.

Area of Science:

• Medical Imaging Analysis
• Computational Pathology
• Oncology Diagnostics

Background:

• Lung cancer diagnosis relies heavily on imaging and pathological analysis.
• The integration of artificial intelligence (AI) offers a promising avenue for enhancing diagnostic accuracy and efficiency.
• Systematic reviews are crucial for understanding the current landscape of AI applications in this field.

Purpose of the Study:

• To systematically review artificial intelligence techniques employed in lung cancer diagnosis.
• To evaluate the performance and applicability of various AI models in detecting lung cancer.
• To identify trends and common AI methodologies used in the literature.

Main Methods:

• A comprehensive systematic search of major scientific databases (Web of Science, PubMed, Scopus, etc.) was conducted.
• Literature published up to June 2025 was included in the review.
• The Prediction Model Risk Of Bias Assessment Tool (PROBAST) was utilized to assess the quality of included studies.

Main Results:

• 204 studies were included, utilizing diverse AI techniques such as Convolutional Neural Networks (CNN), Support Vector Machines (SVM), and Deep Learning (DL) models.
• Convolutional Neural Networks (CNN) emerged as the most frequently applied model.
• Deep learning models demonstrated improved performance metrics (AUC, sensitivity, accuracy) when applied to preprocessed medical images, with accuracy ranging from 68.4% to 100%.

Conclusions:

• AI techniques demonstrate considerable potential for lung cancer diagnosis and detection, exhibiting varied diagnostic accuracy.
• Optimization of AI algorithms and further investigation into their clinical relevance and real-world applicability are warranted.
• The findings underscore the need for continued research to translate AI advancements into effective clinical tools for lung cancer management.