A modified Tseng algorithm approach to restoring thoracic diseases' computerized tomography images

Area of Science:

• Applied Mathematics
• Medical Imaging
• Computer Vision

Background:

• The Tseng algorithm and its modifications are established methods for approximating zeros of the sum of monotone operators.
• Medical imaging techniques like computed tomography (CT) are susceptible to degradation from blur and noise, impacting diagnostic accuracy.
• Restoring degraded medical images is crucial for accurate disease identification and treatment planning.

Purpose of the Study:

• To adapt and apply a modified Tseng algorithm for the restoration of thoracic CT images.
• To evaluate the effectiveness of the modified Tseng algorithm in deblurring and denoising medical images.
• To assess the potential of this approach for advancing monotone inclusion problem-solving in medical image recovery.

Main Methods:

• Utilized a modified Tseng algorithm to restore thoracic CT images degraded by known blur functions and additive noise.
• Applied the algorithm to a range of thoracic CT images depicting conditions such as aortic calcification, subcutaneous emphysema, tortuous aorta, pneumomediastinum, and pneumoperitoneum.
• Employed established image restoration tools for comparative analysis of image quality.

Main Results:

• Successfully restored degraded thoracic CT images, enhancing their quality.
• Demonstrated the algorithm's capability to handle various types of image degradations specific to thoracic pathologies.
• Quantitative and qualitative comparisons showed improvements in the restored images compared to the originals.

Conclusions:

• The modified Tseng algorithm is effective for restoring degraded thoracic CT images.
• This method holds significant potential for improving medical image recovery applications.
• The study highlights the broader applicability of monotone inclusion problem-solving in medical imaging.