Toward high-intensity focused ultrasound lesion quantification using compressive sensing theory

Area of Science:

• Medical Imaging
• Signal Processing
• Biomedical Engineering

Background:

• Compressive sensing (CS) is increasingly applied in pattern recognition, enabling signal recovery below the Nyquist limit.
• High-intensity focused ultrasound (HIFU) therapy requires effective monitoring and control for precise lesion creation.
• Existing imaging methods may have limitations in clearly visualizing HIFU-induced thermal lesions.

Purpose of the Study:

• To investigate a novel application of compressive sensing for controlling and monitoring HIFU therapy.
• To develop and evaluate a new method for detecting HIFU thermal lesions using modified CS, threshold algorithm, and wavelet transforms.

Main Methods:

• A modified compressive sensing approach was combined with a threshold algorithm and wavelet transforms.
• The proposed method was analyzed using both simulated and experimental ultrasound radio frequency data.
• Performance was compared against ultrasound B-mode and standard CS imaging.

Main Results:

• The proposed algorithm significantly enhanced the contrast of HIFU lesions compared to conventional methods.
• The modified CS method demonstrated effective detection of thermal lesions in vitro.
• Estimated lesion size (8.3 mm × 8.4 mm) closely matched the actual size (10.1 mm × 9 mm), with a small difference of 0.8 mm × 0.5 mm.

Conclusions:

• The developed modified compressive sensing method offers improved contrast and accurate detection of HIFU thermal lesions.
• This technique shows promise for enhanced monitoring and control in HIFU therapy.
• The findings support the potential of this algorithm for precise in vitro thermal lesion characterization.