Study on lung CT image segmentation algorithm based on threshold-gradient combination and improved convex hull...

Area of Science:

• Medical Imaging
• Computer Vision
• Image Processing

Background:

• Lung image segmentation is challenging due to noise, uneven grayscale, and complex pathologies.
• Accurate lung segmentation is crucial for diagnosing and monitoring respiratory diseases.

Purpose of the Study:

• To develop an advanced lung image segmentation algorithm addressing noise and contour loss.
• To improve the accuracy, consistency, and robustness of lung segmentation in medical imaging.

Main Methods:

• Proposed an initial lung mask extraction using a combination of threshold and gradient-based methods.
• Utilized a time series feature extraction method based on differential memory (TFDM) for gradient calculation.
• Implemented an improved convex hull algorithm for lung contour repair to handle lesions.

Main Results:

• The proposed algorithm achieved superior segmentation results on a COVID-19 CT dataset.
• Demonstrated significant improvements in the consistency and accuracy of lung segmentation.
• The method effectively captured more lung information, leading to robust segmentation.

Conclusions:

• The developed algorithm offers an effective solution for challenging lung image segmentation tasks.
• The combination of TFDM and improved convex hull enhances segmentation accuracy and robustness.
• This method holds promise for improved analysis of respiratory conditions from medical images.