In Vivo Multithread Ultrasound Thermal Strain Imaging

Area of Science:

• Biomedical Engineering
• Medical Imaging
• Therapeutic Monitoring

Background:

• Real-time temperature monitoring is crucial for optimizing thermal therapy planning, control, and evaluation.
• Thermal strain imaging (TSI) shows promise for in vitro temperature estimation, but in vivo applications are hindered by motion artifacts and errors.
• Previous respiration-separated TSI (RS-TSI) methods have limitations in addressing complex physiological motion.

Purpose of the Study:

• To develop and evaluate a novel multithread thermal strain imaging (MT-TSI) approach for enhanced in vivo thermometry.
• To improve the accuracy, reduce noise, and increase the temporal density of temperature estimations during thermal therapy.
• To overcome the challenges of physiological motion artifacts in ultrasound-based thermometry.

Main Methods:

• A flag image frame is identified using ultrasound image correlation analysis.
• Respiration phase profiles are determined and divided into multiple subranges for parallel processing.
• Independent TSI calculations, including motion compensation and thermal strain estimation, are performed in parallel threads.
• Temporal extrapolation, spatial alignment, and interthread noise suppression are applied before averaging results.

Main Results:

• The MT-TSI approach demonstrated thermometry accuracy comparable to RS-TSI in porcine perirenal fat microwave heating experiments.
• MT-TSI exhibited significantly lower noise levels compared to RS-TSI.
• The proposed method achieved higher temporal density in temperature estimations.

Conclusions:

• MT-TSI offers a robust solution for overcoming motion artifacts in ultrasound-based in vivo thermometry.
• This advanced TSI technique enhances the reliability and precision of real-time temperature monitoring during thermal therapies.
• MT-TSI represents a significant advancement for improving the clinical application and outcomes of thermal therapies.