Using Passive Cavitation Images to Classify High-Intensity Focused Ultrasound Lesions

摘要: Abstract Passive cavitation imaging provides spatially resolved monitoring of emissions. However, the diffraction limit a linear array results in relatively poor range resolution. Poor resolution has limited prior analyses spatial specificity and sensitivity passive predicting thermal lesion formation. In this study, limitation is overcome by orienting orthogonal to high-intensity focused ultrasound propagation direction performing imaging. Fourteen lesions were formed ex vivo bovine liver samples as result 1.1-MHz continuous-wave exposure. The classified focal, “tadpole” or pre-focal based on their shape location. images beamformed from emissions at fundamental, harmonic, ultraharmonic inharmonic frequencies with an established algorithm. Using area under receiver operating characteristic curve (AUROC), harmonic found be significant predictors formation for all types. For both emissions, most successfully (AUROC values 0.87 0.88, respectively), followed tadpole 0.77 0.64, respectively) focal 0.65 0.60, respectively).