The use of an Artificial Neural Network (ANN) in the evaluation of the Extracorporeal Shockwave Lithotripsy (ESWL) as a treatment of choise for urinary lithiasis

Abstract

PurposeArtificial Neural Networks (ANNs) are simplified computational models simulating the central nervous system. They are widely applied in medicine, since they substantially increase the sensitivity and specificity of the diagnosis, classification and the prognosis of a medical condition. In this study we constructed an artificial neural network to evaluate several parameters of extracorporeal shockwave lithotripsy (ESWL), such as the outcome and safety of the procedure.Materials and methodsPatients with urinary lithiasis suitable for ESWL treatment were enrolled. An artificial neural network (ANN) was designed and a unique algorithm was executed with the use of the well-known numerical computing environment, MATLAB. Medical data were collected from all patients and 12 nodes were used as inputs (sex, age, B.M.I. (Body Mass Index), stone location, stone size, comorbidity, previous ESWL sessions, analgesia, number of shockwaves, shockwave intensity, presence of a ureteral stent and hydronephrosis). Conventional statistical analysis was also performed.Results716 patients were finally included in our study. Univariate analysis revealed that diabetes and hydronephrosis were positively correlated to the ESWL complications. Regarding efficacy, univariate analysis revealed that stone location, stone size, the number and density of shockwaves delivered and the presence of a stent in the ureter were independent factors of the ESWL outcome. This was further confirmed when adjusted for sex and age in a multivariate analysis.The performance of the ANN (predictive/real values) at the end of the training state reached 98,72%. The four basic ratios (sensitivity, specificity, PPV, NPV) were calculated for both training and evaluation data sets. The performance of the ANN at the end of the evaluation state was 81,43%.ConclusionsOur ANN achieved high score in predicting the outcome and the side effects of the extracorporeal shockwave lithotripsy treatment for urinary stones. In fact, the accuracy of the network may further improve by using larger sets of data, different architecture in designing the model or using different set of input variables, making ANNs thus, a quite promising instrument for effective, precise and swift medical diagnosis.