Optical Feedback for Safe Automatic Laser Lithotripsy: Tissue Sensor Implementation in a Clinical System

To develop an optical feedback system compatible with a commercial surgical laser for automatically distinguishing between urinary stones and soft tissues during laser lithotripsy, thereby enhancing procedural safety.

The system, based on diffuse reflectance spectroscopy (DRS), was implemented in an engineered clinical theranostic platform. In vivo experiments were conducted to collect and analyze DRS spectra of tissues during laser lithotripsy. Illumination was performed via the endoscope, and detection was performed via the treatment fiber. Classification of urinary stones and soft tissues was performed using machine learning methods, i.e., Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA).

The system demonstrated high diagnostic performance, with 93% sensitivity for soft tissue identification and 93% specificity for stone detection evaluated by the LDA method. This real-time differentiation effectively minimized unintended laser exposure to non-target tissues.

The developed optical guidance system provides real-time feedback during laser lithotripsy, improving safety and precision by reducing the risk of accidental tissue damage. The proposed technology is expected to enhance outcomes in minimally invasive urological laser procedures.