AI-powered ultrasonic thermometry for HIFU therapy in deep organ

• DOI: 10.1016/j.ultsonch.2024.107154
• Publication Date: 2024-11-12T17:30:04Z
• Authors: Shunyao Luan; Yongshuo Ji; Yumei Liu; Linling Zhu; Hong Zhao; Haoyu Zhou; Ke Li; Weizhen Zhu; Benpeng Zhu
• Keywords: Ultrasonic cavitation; Noninvasive acoustic energy distribution monitoring; Respiratory compensation; Acoustics. Sound; QC221-246; VSI: Power Ultrasound & Cavitation; Deep learning; High Intensity focused ultrasound; Chemistry; 03 medical and health sciences; 0302 clinical medicine; QD1-999

High-intensity focused ultrasound (HIFU) is considered as an important non-invasive way for tumor ablation in deep organs. However, accurate real-time monitoring of the temperature field within HIFU focal area remains a challenge. Although technology, compared with other approaches, good choice noninvasive and on distribution, traditional ultrasonic thermometry mainly relies backscattered signal, which difficult high (>50 °C) measurement. Given that artificial intelligence (AI) shows significant potential biomedical applications, we propose AI-powered using end-to-end neural network termed Breath-guided Multimodal Teacher-Student (BMTS), possesses capability to elucidate interaction between complex heterogeneous biological media. It has been demonstrated experimentally two-dimension distribution organ can be accurately reconstructed average error frame speed 0.8 °C 0.37 s, respectively. Most importantly, maximum measurable technology successfully expanded record value 67 °C. This breakthrough indicates development beneficial precise therapy planning future.