Lung nodule conspicuity using unsharp mask filters with storage-phosphor-based computed radiography

Area of Science:

• Radiology
• Medical Imaging
• Image Processing

Background:

• Storage-phosphor computed radiography (SR) is a key imaging modality for lung nodule detection.
• Early detection of lung cancer, often presenting as faint nodules, is crucial for patient outcomes.
• Unsharp mask filtering is a common post-processing technique used to enhance image details.

Purpose of the Study:

• To evaluate and compare the effectiveness of different unsharp mask filter types in SR.
• To optimize lung nodule detection, particularly for faint nodules indicative of early lung cancer.
• To identify filter parameters that improve the visibility of small pulmonary nodules.

Main Methods:

• Utilized an anthropomorphous chest phantom with simulated nodules (5-mm sphere, 5-mm and 10-mm hemispheres).
• Acquired 120 SR radiographs with nodules placed at various sites.
• Applied five distinct unsharp mask filter parameter combinations, including mid-frequency suppressing and enhancing types.
• Conducted a reader study with eight observers and analyzed results using Receiver Operating Characteristic (ROC) analysis.

Main Results:

• All five unsharp mask filter types demonstrated similar visibility for 10-mm nodules.
• Filters with mid-frequency suppression and very-low-frequency enhancement significantly improved the detection of 5-mm nodules compared to conventional filters.
• The enhanced filters were particularly effective in revealing subtle nodular opacities.

Conclusions:

• Mid-frequency suppressing unsharp mask filters enhance the visualization of faint nodules, crucial for detecting early bronchogenic carcinoma.
• These advanced filters show potential to replace conventional filters in clinical practice for improved lung nodule detection.
• The findings support the use of specific unsharp mask filter settings to improve diagnostic accuracy in SR lung imaging.