Peter A Noble et al., 2024: Stone decision engine accurately predicts stone removal and treatment complications for shock wave lithotripsy and laser ureterorenoscopy patients
Peter A Noble 1 , Blake D Hamilton 2 , Glenn Gerber 3
1Department of Microbiology, University of Alabama Birmingham, Birmingham, AL, United States of America.
2School of Medicine, University of Utah, Salt Lake City, UT, United States of America.
3University of Chicago Medical Center, Chicago, IL, United States of America.
Abstract
Kidney stones form when mineral salts crystallize in the urinary tract. While most stones exit the body in the urine stream, some can block the ureteropelvic junction or ureters, leading to severe lower back pain, blood in the urine, vomiting, and painful urination. Imaging technologies, such as X-rays or ureterorenoscopy (URS), are typically used to detect kidney stones. Subsequently, these stones are fragmented into smaller pieces using shock wave lithotripsy (SWL) or laser URS. Both treatments yield subtly different patient outcomes. To predict successful stone removal and complication outcomes, Artificial Neural Network models were trained on 15,126 SWL and 2,116 URS patient records. These records include patient metrics like Body Mass Index and age, as well as treatment outcomes obtained using various medical instruments and healthcare professionals. Due to the low number of outcome failures in the data (e.g., treatment complications), Nearest Neighbor and Synthetic Minority Oversampling Technique (SMOTE) models were implemented to improve prediction accuracies. To reduce noise in the predictions, ensemble modeling was employed. The average prediction accuracies based on Confusion Matrices for SWL stone removal and treatment complications were 84.8% and 95.0%, respectively, while those for URS were 89.0% and 92.2%, respectively. The average prediction accuracies for SWL based on Area-Under-the-Curve were 74.7% and 62.9%, respectively, while those for URS were 77.2% and 78.9%, respectively. Taken together, the approach yielded moderate to high accurate predictions, regardless of treatment or outcome. These models were incorporated into a Stone Decision Engine web application (http://peteranoble.com/webapps.html) that suggests the best interventions to healthcare providers based on individual patient metrics.
PLoS One. 2024 May 2;19(5):e0301812. doi: 10.1371/journal.pone.0301812. eCollection 2024.
PMID: 38696418
Comments 1
Selection of the most appropriate low-invasive method for stone removal has remained a matter of concern for clinicians over the years. In the current article the authors describe a STONE DECISIONS ENGINE for prediction of the outcome of SWL and URS in terms of treatment success and complications. The theoretical as well as computational details of the “engine” certainly is beyond the horizon for most urologists and so also for the reviewer.
The results presented in this extensive analysis is based on findings in 17242 patient records. Success was defined as stone-free and up to 4 mm fragments and complications as “0” or “1”.
There were more kidney stones and more SWL procedures in the analysis.
Like in previous analyses, calculations and nomograms, the important variables were: gender, age, SSD, BMI, number of stones, size, location, and stone characteristics. It is possible that future clinical decisions will be taken over by ANN or other AI-systems, but the ultimate outcome still must rely on personal expertise and on which attention that is paid to the details of SWL and URS treatments.
The authors should be congratulated to this ingenious decision engine, but my own opinion is that extensive personal experience plays an equal or even better basis for decisions.
Hans-Göran Tiselius