Laser lithotripsy (LL) is commonly used for kidney stone treatment, with cavitation playing a crucial role in stone fragmentation. However, the relationship between cavitation activities and stone damage remains unclear. This supplement study by Anthony DiSpirito will utilize deep learning to predict stone damage based on passive cavitation mapping (PCM) signals. A three-dimensional PCM system will be employed, along with B-mode Ultrasound (US) for data acquisition. Deep learning enables the optimization of LL procedures by automating feature selection and identifying key factors driving stone damage. Our approach will offer valuable insights into medical practice, advancing the efficacy of LL treatments. We will adopt deep learning module combined with cavitation activities information, and we can better predict the potential damaged induced by bubble collapsing. Our result will also demonstrate the strong correlation between bubble collapsing information with stone crater damage. This approach can also be further explored with more sophisticated scenarios and clinical applications.