A B OILING H ISTOTRIPSY S YSTEM FOR D EEP T ISSUE A BLATION Adam Maxwell 1,2 , Wayne Kreider 1 , Petr Yuldashev 3 , Tanya Khokhlova 1 , Oleg Sapozhnikov 1,3 , Michael Bailey 1 , and Vera Khokhlova 1,3 1 CIMU, Applied Physics Laboratory, University of Washington 2 Dept. of Urology, University of Washington School of Medicine 3 Dept. of Acoustics, Physics Faculty, Moscow State University ISTU 2013 May 15, 2013 Shanghai, China 1
B OILING H ISTOTRIPSY Boiling histotripsy is a noninvasive method to generate mechanical lesions in tissue using millisecond-long focused ultrasound pulses with shock fronts. 100 80 60 Pressre (MPa) 40 20 0 -20 5.5 6 6.5 7 7.5 Time ( µ s) Millisecond Boiling Tissue Emulsification Shock-induced Heating Study Objectives: 1. Develop a system capable of producing boiling histotripsy lesions at clinically relevant depths (> 5-6 cm) 2. Characterize output of this system 3. Evaluate use of a derating procedure to determine exposure for boiling histotripsy at different tissue depths 2
B OILING H ISTOTRIPSY S YSTEM Transducer Amplifier • 7-element lens-focused design • Class D RF Amplifier 1 • Center Frequency: 1 MHz • Peak Power: >30 kW for ≤10 ms pulse • Aperture: 14.7 cm • Frequency Range: 0.1 – 4.0 MHz • Focal Length: 14.0 cm Capacitor HV Supply Array FPGA Amplifier Matching Timing Board Network Board Elements LV Supply MATLAB Transducer 3 1 Hall TL and Cain CA. AIP Conf Proc. 2005;829: 445-449
T RANSDUCER O UTPUT C HARACTERIZATION Acoustic Holography + Nonlinear Simulation 1 : Compare simulated results with direct measurement of focal waveforms using fiber optic probe hydrophone (FOPH 2000) 1. Acquire low amplitude waveforms with hydrophone in prefocal plane 2. Create 2D pressure at z = -5.5 cm 80 field map from data 60 Pressre (MPa) 40 3. Apply linear 20 backward propagation 0 algorithm to obtain a -20 1 Kreider W et al., IEEE Trans. Ultrason. -5 0 5 10 source hologram in a Time ( µ s) Ferroelectr. Freq. Control 2013; in press. plane z = 0 4. Apply nonlinear forward propagation 4 Yuldashev P.V., Khokhlova V.A. at increased p 0 amplitude to obtain 4 Acoust. Phys. 2011, 57(3): 334-343. focal waveforms and beam profiles
N ONLINEAR PROPAGATION MODEL 3D Westervelt equation in the retarded temporal coordinates æ ö ¶ ¶ ¶ ¶ b ¶ d ¶ 2 2 2 2 2 2 3 c p p p p p p ç ÷ = + + + + 0 ç ÷ ¶ t ¶ ¶ ¶ ¶ r ¶ t ¶ t 2 2 2 3 2 3 3 z 2 x y z 2 c 2 c è ø 0 0 0 Boundary condition: Linear field distribution: back propagated hologram axial plane, dB scale 0 0.5 1 p / p max -80 -60 -40 -20 y , mm 0 20 40 60 80 -80 -60 -40-20 0 20 40 60 80 x , mm Yuldashev P.V., Khokhlova V.A. Simulation of three-dimensional nonlinear fields of ultrasound therapeutic 5 arrays. Acoust. Phys. 2011, 57(3), p.334-343.
R ESULTS FROM C HARACTERIZATION Focal Peak Pressures Focal Pressure Waveforms: versus Transducer Voltage FOPH measurements versus simulations 100 80 p + 20 p + , MPa 60 Pressure (MPa) a) 10 40 35 V 20 0 0 0 40 experiment Pressure (MPa) b) -5 hologram 20 p - , MPa 45 V -10 p - 0 -15 10 15 20 -20 time, radians 0 20 40 60 80 100 120 Source voltage, V Simulations for higher input voltages are in progress 6
R ESULTS FROM C HARACTERIZATION Focal Pressure Waveforms Pressure (MPa) 100 simulations 100 FOPH measurements 60 V Pressure (MPa) 50 60 V 50 0 0 0 5 10 time, radians 100 Nonlinear peak pressure distributions: axial plane, dB scale Pa) p + p - 7
D ERATING FOR L ESION G ENERATION Hypothesis: a derating procedure enables us to determine the source output to provide a desired nonlinear focal waveforms when focusing at a depth L in tissue. Axial Beam Pressure Profile ( ) p 0 ,Liver = p 0 ,Water exp α L Bessonova et al. 60 40 α = 0.5 dB/cm Pressre (MPa) 20 0 -20 98 99 100 101 102 Time ( µ s) Canney et al. 60 p + 40 Pressre (MPa) 20 0 -20 98 99 100 101 102 Time ( µ s) p - 8 1 Bessonova et al. Acoustical Physics 2010;56(3):376–385. 2 Canney et al. Ultrasound Med Biol 2010;36:250-267
T HRESHOLD FOR B OILING Analytical Boiling Time 4 ρ 2 C s Δ t = Δ T 6 c 0 3 Pulse duration: 10 ms β f 0 A s Total on time: 500 ms PRF: 1 Hz Tissue Depth: 12 mm Present Study – Lesion Probability vs. Pressure Present Study Khokhlova et al 1 1 Khokhlova et al 1 Proportion with Lesion 0.8 5 mm p + = 75 MPa 5 mm 0.6 0.4 0.2 n ≥ 9 5 mm 0 p + = 58 MPa 20 40 60 80 Peak Positive Pressure (MPa) 9 1 TD Khokhlova et al. J Acoust Soc Am 2011; 130:3498-3510
B OILING VS . T ISSUE D EPTH • Lesions generated at tissue depths up to 6 cm • Linear derating underestimated pressure at greater depths by ≤14% • Lesions at 6 cm depth required 26% maximum p 0 of US transducer Transducer Voltage for Lesions vs. Depth Lesion Photographs 18 mm 35 mm 50 mm 60 mm 120 Transducer Driving Voltage (V) 100 80 70 V DC 60 90 V DC Maximum Pressure - No Lesion 40 Minimum Pressure - Lesion 20 Derated 0.5 dB/cm 100 V DC Derated 0.61 dB/cm (best fit) 105 V DC 0 0 1 2 3 4 5 6 7 Tissue Depth (cm) 10
C ONCLUSIONS • Boiling histotripsy can be applied to create lesions at clinically relevant (6 - 16 cm) tissue depths • Acoustic holography + nonlinear modeling provides characterization of transducer output in water and in tissue • Derating of focal waveforms obtained in water estimates in situ pressures and output needed from the transducer to induce boiling histotripsy 11
A CKNOWLEDGMENTS • Tim Hall and Yohan Kim at University of Michigan • Julianna Simon at APL - University of Washington • Work supported by: – NIH 2R01 EB007643-05 – Multidisciplinary Training Program in Benign Urology at UW (NIH 2T32 DK00779-11A1) 12
13
T RANSDUCER O UTPUT C HARACTERIZATION Acoustic Holography + Nonlinear Simulation 1 : 1. Acquire pressure waveforms at low amplitude in prefocal plane 2. Apply backward propagation algorithm to obtain transducer surface hologram 3. Apply nonlinear forward propagation simulation with Westervelt equation to obtain focal waveforms, beam profiles 80 60 Pressre (MPa) 40 20 0 -20 -5 0 5 10 Time ( µ s) (3) (1) (2) Results compared with direct measurement of focal waveforms using fiber optic probe hydrophone (FOPH 2000) 14 1 Kreider W et al., IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2013; in press.
P OWER O UTPUT M EASUREMENTS • Electrical Power – Voltage and current output from amplifier measured for 10 ms pulse duration – Average power calculated from V, I waveforms – Power measured to 250 V DC , extrapolated to 400 V DC • Acoustic Power – Power calculated from hologram at 5 V DC – Scaled for other power outputs – Surface pressure output was measured to be linear with input voltage to 150 V DC , extrapolated to 400 V DC 15
P OWER O UTPUT • Amplifier (Electric Power) – Maximum tested peak power: 36.1 kW @ 250 V DC – Maximum tested mean power over 10 ms: 10.1 kW @ 250 V DC – Maximum est. peak power: 92.4 kW @ 400 V DC – Maximum est. mean power over 10 ms: 25.9 kW @ 400 V DC • Transducer (Acoustic Power) – Peak power @ 250 V DC : 8.9 kW – Mean power @ 250 V DC : 4.5 kW – Maximum est. peak power: 23.0 kW @ 400 V DC – Maximum est. mean power over 10 ms: 11.5 kW @ 400 V DC • Transducer Power Efficiency (from Avg. Power) – 44-49% 16
P OWER O UTPUT M EASUREMENTS Maximum output we have tested Measurement Maximum rated error output 17
Recommend
More recommend
