Near-Infrared Blood Vessel Image Segmentation Using Background Subtraction and Improved Mathematical Morphology

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Image Processing

Background:

• Accurate visualization of superficial blood vessels is critical for medical procedures like intravenous injections and disease diagnosis.
• Current infrared imaging methods for blood vessels suffer from noise, breaks, and uneven data, limiting their clinical utility.
• Existing techniques often fail to provide the necessary clarity for detailed vascular analysis.

Purpose of the Study:

• To develop and evaluate a novel image segmentation algorithm for improving the quality of superficial blood vessel images.
• To address limitations in current imaging techniques, specifically noise, breaks, and unevenness in vascular data.
• To provide a robust method for enhancing vascular image clarity for medical applications.

Main Methods:

• The proposed algorithm utilizes background subtraction and improved mathematical morphology for image segmentation.
• It models the image as a superposition of blood vessels onto a background.
• Noise reduction is achieved by analyzing connected domain sizes, ensuring uniform vessel width and smoothed edges.

Main Results:

• The algorithm effectively removes noise and artifacts from superficial blood vessel images.
• It produces images with uniform blood vessel width and smoothed edges, accurately reflecting the actual vascular state.
• Subjective and objective evaluations confirm the algorithm's capability to extract accurate and clear vascular information.

Conclusions:

• The developed image segmentation algorithm significantly enhances the quality of superficial blood vessel imaging.
• This method offers a reliable approach for obtaining clear and accurate vascular data, beneficial for clinical diagnosis and procedures.
• The findings provide a foundation for improved vascular image quality assessment and application.