Ultrasound thermal change detection based on steerable filters

Area of Science:

• Medical Imaging
• Biomedical Engineering
• Surgical Technology

Background:

• Laser and Radiofrequency (RF) knives require precise thermal control during surgical procedures.
• Ultrasound velocity is sensitive to the thermal properties of surrounding tissues, making it a potential tool for temperature monitoring.
• Accurate real-time temperature feedback is crucial for optimizing surgical interventions and patient outcomes.

Purpose of the Study:

• To develop and evaluate a novel ultrasound-based method for accurate tissue temperature detection.
• To assess the efficacy of multiresolution steerable filters for motion estimation in thermal imaging.
• To validate the technique on both simulated and in-vivo surgical scenarios.

Main Methods:

• Utilized multiresolution steerable filters for motion estimation to track speckle shifts in ultrasound images.
• Applied the technique to simulated thermal phantoms and real in-vivo cases involving surgical occlusion and reopening of renal arteries.
• Focused on analyzing the correlation between speckle shifts and ground truth temperature.

Main Results:

• The proposed method demonstrated promising results in observing internal organ temperature changes during surgical procedures.
• Steerable filters, due to their orientation in space and time, achieved more accurate thermal detection.
• In simulated images, the average thermal error was 0.68°C with a standard deviation of 0.79°C, indicating high precision.

Conclusions:

• Ultrasound thermal detection using steerable filter-based motion estimation is a viable, non-invasive technique for monitoring tissue temperature.
• This method offers a promising approach for enhancing diagnosis and therapy using standard digital ultrasound systems.
• The technique provides accurate temperature measurements, crucial for improving surgical safety and efficacy.