Lung cancer identification: a review on detection and classification

Area of Science:

• Medical Imaging
• Radiology
• Computer-Aided Diagnosis

Background:

• Lung cancer is a leading cause of mortality worldwide.
• Early diagnosis via computed tomography (CT) screening and lung nodule identification can significantly reduce mortality.
• Assessing numerous CT images for accurate nodule detection presents challenges for radiologists due to data volume.

Purpose of the Study:

• To review and analyze various computer-aided diagnosis (CAD) approaches for lung nodule detection and classification.
• To provide a comprehensive overview of methods assisting radiologists in analyzing CT images for lung cancer screening.
• To highlight the evolution of handcraft and learned approaches in CAD systems for nodule analysis.

Main Methods:

• Review of existing literature on computer-aided diagnosis (CAD) systems for lung cancer.
• Analysis of different computational methods for lung nodule detection and classification from CT images.
• Categorization of approaches into handcraft and learned methods.

Main Results:

• Identified diverse CAD methods for lung nodule analysis in CT scans.
• Highlighted the challenges faced by radiologists in interpreting large volumes of CT data.
• Demonstrated the potential of CAD systems to assist in early lung cancer diagnosis.

Conclusions:

• Computer-aided diagnosis (CAD) systems offer significant assistance to radiologists in lung nodule detection and classification.
• The review provides a comprehensive analysis of current CAD approaches, aiding further research and development.
• Improved CAD tools can enhance the accuracy and efficiency of early lung cancer diagnosis through CT screening.