Parallel integral projection transform for straight electrode localization in 3-D ultrasound images

Area of Science:

• Medical imaging
• Surgical technology
• Image processing

Background:

• Accurate localization of small metallic surgical instruments in human tissue is a significant challenge.
• Medical ultrasound offers real-time visualization of anatomical structures and surgical instruments.

Purpose of the Study:

• To develop an image-processing algorithm for automatic localization of line-segment-shaped metallic objects in biological tissue.
• To assess the accuracy and robustness of the proposed localization method.

Main Methods:

• An algorithm was developed to estimate the position of a line-segment-shaped object using the parallel integral projection transform (a form of the Radon transform).
• A hierarchical mesh-grid algorithm was implemented to accelerate the axis estimation step.
• Electrode tip localization was performed after determining the axis position.

Main Results:

• The algorithm was tested on simulated, phantom, and real ultrasound images (breast biopsy).
• Localization accuracy was found to be in the hundreds of micrometers, comparable to ultrasound axial resolution.
• The method demonstrated robustness against variations in electrode position and speckle noise.

Conclusions:

• The proposed image-processing algorithm enables accurate and robust localization of metallic surgical instruments using ultrasound.
• This technique has potential applications in enhancing surgical navigation and improving patient safety.