Review of automatic pulmonary lobe segmentation methods from CT

Area of Science:

• Medical imaging analysis
• Computational anatomy
• Pulmonary medicine

Background:

• Accurate pulmonary lobe segmentation from CT images is crucial for healthcare applications like surgical planning.
• Automated algorithms are being developed to address this challenging segmentation task.
• Existing methods vary significantly, lacking standardization across diverse clinical scenarios.

Purpose of the Study:

• To conduct a methodological review of automated pulmonary lobe detection algorithms from CT images.
• To compare the strengths and weaknesses of different algorithmic approaches.
• To propose a framework for developing more robust and versatile segmentation methods.

Main Methods:

• Systematic review of published automated pulmonary lobe segmentation algorithms.
• Analysis of common stages within these algorithms.
• Comparative assessment of methodologies employed by different authors.

Main Results:

• No single, universally accepted standard method for pulmonary lobe detection has emerged.
• Current algorithms have not been validated across a wide spectrum of pathologies and imaging protocols.
• Diverse approaches exist, each with specific advantages and limitations.

Conclusions:

• Combining different algorithmic strategies may lead to improved performance.
• A structured workflow is proposed for the development of next-generation pulmonary lobe segmentation algorithms.
• Further research is needed to establish standardized and validated methods for clinical application.