A warning system for urolithiasis via retrograde intrarenal surgery using machine learning: an experimental study

• DOI: 10.1186/s12894-022-01032-5
• Publication Date: 2022-06-06T09:47:09Z
• Authors: Jinho Jeong; Kidon Chang; Jisuk Lee; Jongeun Choi
• Keywords: Kidney stones; Research; Retrograde intrarenal surgery; Diseases of the genitourinary system. Urology; Machine Learning; Kidney Calculi; 03 medical and health sciences; Treatment Outcome; 0302 clinical medicine; Urolithiasis; Machine learning; Discrete wavelet transform; Humans; RC870-923; Ureter

Abstract Background To develop a warning system that can prevent or minimize laser exposure resulting in kidney and ureter damage during retrograde intrarenal surgery (RIRS) for urolithiasis. Our study builds on the hypothesis shock waves of different degrees are delivered to hand surgeon depending whether hits stone tissue. Methods A surgical environment was simulated RIRS by filling body raw whole chicken with water stones from human body. We developed an acceleration measurement recorded power signal data number hours, yielding distinguishable characteristics among three states (idle state, stones, tissue–laser interface) conducting fast Fourier transform (FFT) analysis. discrete wavelet (DWT) used feature extraction, random forest classification algorithm applied classify current state laser-tissue interface. Results The result FFT showed magnitude spectrum is within frequency range < 2500 Hz, indicating distinguishable. Each cut only 0.5-s increments transformed using DWT. were entered into classifier train model. test measured dataset isolated training dataset. maximum average accuracy > 95%. procedure repeated dummy data, 33.33% proving proposed method caused no bias. Conclusions monitoring receives shockwave signals generated urolithiasis treatment generates irradiance status rapidly recognizing (in 0.5 s) high (95%). postulate this significantly error RIRS.