Non-invasive Ultrasonic Temperature Measurement Imaging Algorithm Based On Sound Velocity Distribution Evaluation

Traditional methods towards cancer therapy,such as radiotherapy,chemotherapy and resection,not only attain insufficient efficacy,but also engender great harm to patients.Interventional therapy is an advanced method to head and kill the tumor by radiating energy into body,which can attain higher treatment efficiency,engender less harm and apply to more symptoms.However,noninvasive temperature imaging technique is significant while the operation.Thus,many ultrasonic non-invasive thermometry methods are proposed.Phase-contrast method based on the change of sound speed is effective,but it is difficult to put in clinical application because of practical issues caused by thermo-acoustic lensing,thermal expansion and patient motion.Therefore,a new method to overcome the above-mentioned difficulties is proposed in this paper.During the Interventional operation,as temperature rise,sound speed in therapeutic area will increase and it results in a local sound speed region,which blurs ultrasonic imaging.Thus,we present a new ultrasonic non-invasive thermometry algorithm applying to monitor interventional therapy,which is based on measuring spatial distribution of sound speed.In the previous stage,we set up a guide stars based experiment to verify that it is possible to measure the local sound speed of therapeutic area in one single frame of imaging data,so that the interference in thermometry,such as thermo-acoustic lensing,thermal expansion and patient motion,can be eliminated.In the follow-up study stage,to be closer to actual clinical,guide stars are replaced by contour structures in animal organ imaging.We also make phantoms of different sound speed(1400-1800m/s)to simulate the sound speed change in therapeutic area.In experiment,phantoms and pig liver are placed together in a water tank to simulate a clinical interventional operation procedure.The algorithm is established by following steps: Firstly,a baseline image is reconstructed by synthetic aperture method from ultrasound channel data.Then we set up an auto-searching method to find contour structures ROIs,such as vascular wall.Secondly,the distribution of local sound speed is initialized.According to current distribution,ROIs are reconstructed again and their focus quality is quantified.Iterative algorithm improves the focus quality by optimizing the parameters of sound speed distribution.Thus,the distribution adaptively approaches what it really is.Consequently,the truly spatial distribution of sound speed is measured and thermometry is attained by the correlation between sound speed and temperature.We use Prodigy,an ultrasonic research platform from S# corporation,Taiwan,to acquire 128 TR ultrasound channel data,with 6MHz probe center frequency and 25 MHz sample rate.The algorithm successfully measures the spatial distribution of sound speed.The measure error of position and hyperthermia degree(defined in this paper)is within 1.52 mm and 10% respectively.Generally,the presented method is potentially applied to thermometry monitor of interventional therapy.