Automated image quality assessment for chest CT scans

Area of Science:

• Medical imaging
• Radiology
• Computer-aided diagnosis

Background:

• Maintaining high medical image quality is crucial for accurate clinical interpretation.
• Automated methods offer potential for consistent and objective image quality assessment.

Purpose of the Study:

• To present an automated method for assessing the quality of chest CT scans used in lung cancer screening.
• To characterize image noise and intensity calibration using automated algorithms.

Main Methods:

• The method utilizes measurements from three automatically segmented homogeneous regions: external air, trachea lumen air, and descending aorta blood.
• It computes profiles of CT scanner behavior, including noise and calibration characteristics.

Main Results:

• The automated method demonstrated repeatable noise and calibration measurements on both phantom and real low-dose chest CT scans.
• Distinct differences in noise and calibration profiles were observed across various scanners and imaging protocols.

Conclusions:

• Automated image quality assessment provides a valuable tool for quality control in lung cancer screening programs.
• This approach can enhance the performance of automated computer analysis methods used in medical imaging.