Computer-Generated R.E.N.A.L. Nephrometry Scores Yield Comparable Predictive Results to Those of Human-Expert Scores in Predicting Oncologic and Perioperative Outcomes
This article is commented on by the following:
is corrected by
Abstract
Purpose:
We sought to automate R.E.N.A.L. (for radius, exophytic/endophytic, nearness of tumor to collecting system, anterior/posterior, location relative to polar line) nephrometry scoring of preoperative computerized tomography scans and create an artificial intelligence-generated score (AI-score). Subsequently, we aimed to evaluate its ability to predict meaningful oncologic and perioperative outcomes as compared to expert human-generated nephrometry scores (H-scores).
Materials and Methods:
A total of 300 patients with preoperative computerized tomography were identified from a cohort of 544 consecutive patients undergoing surgical extirpation for suspected renal cancer at a single institution. A deep neural network approach was used to automatically segment kidneys and tumors, and geometric algorithms were developed to estimate components of R.E.N.A.L. nephrometry score. Tumors were independently scored by medical personnel blinded to AI-scores. AI- and H-score agreement was assessed using Lin’s concordance correlation and their predictive abilities for both oncologic and perioperative outcomes were assessed using areas under the curve.
Results:
Median age was 60 years (IQE 51–68), and 40% were female. Median tumor size was 4.2 cm and 91.3% had malignant tumors, including 27%, 37% and 24% with high stage, grade and necrosis, respectively. There was significant agreement between H-scores and AI-scores (Lin’s ⍴=0.59). Both AI- and H-scores similarly predicted meaningful oncologic outcomes (p <0.001) including presence of malignancy, necrosis, and high-grade and -stage disease (p <0.003). They also predicted surgical approach (p <0.004) and specific perioperative outcomes (p <0.05).
Conclusions:
Fully automated AI-generated R.E.N.A.L. scores are comparable to human-generated R.E.N.A.L. scores and predict a wide variety of meaningful patient-centered outcomes. This unambiguous artificial intelligence-based scoring is intended to facilitate wider adoption of the R.E.N.A.L. score.
References
- 1. : Renal tumor anatomic complexity clinical implications for urologists kidney cancer nephrometry score renal mass tumor complexity. Urol Clin North Am 2017; 44: 179. Google Scholar
- 2. : The R.E.N.A.L. nephrometry score: a comprehensive standardized system for quantitating renal tumor size, location and depth. J Urol 2009; 182: 844. Link, Google Scholar
- 3. : Preoperative aspects and dimensions used for an anatomical (PADUA) classification of renal tumours in patients who are candidates for nephron-sparing surgery. Eur Urol 2009; 56: 786. Google Scholar
- 4. : A multidisciplinary evaluation of inter-reviewer agreement of the nephrometry score and the prediction of long-term outcomes. J Urol 2011; 186: 1223. Link, Google Scholar
- 5. : Anatomic features of enhancing renal masses predict malignant and high-grade pathology: a preoperative nomogram using the RENAL nephrometry score. Eur Urol 2011; 60: 241. Google Scholar
- 6. : Interobserver variability of R.E.N.A.L., PADUA, and centrality index nephrometry score systems. World J Urol 2015; 33: 853. Google Scholar
- 7. : The RENAL nephrometry nomogram: statistically significant, but is it clinically relevant?Eur Urol 2011; 60: 249. Google Scholar
- 8. : Machine learning in medicine. N Engl J Med 2019; 380: 1347. Google Scholar
- 9. : Big data and machine learning in health care. JAMA 2018; 319: 1317. Google Scholar
- 10. : Deep learning. Nature 2015; 521: 436. Google Scholar
- 11. : ImageNet large scale visual recognition challenge. Int J Comput Vis 2015; 115:211. Google Scholar
- 12. : Skin cancer classification using convolutional neural networks: systematic review. J Med Internet Res 2018; 20: e11936. Google Scholar
- 13. : Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs. JAMA 2016; 316: 2402. Google Scholar
- 14. : Radiomics: extracting more information from medical images using advanced feature analysis. Eur J Cancer 2012; 48: 441. Google Scholar
- 15. : Radiomics: images are more than pictures, they are data. Radiology 2016; 278: 563. Google Scholar
- 16. : Machine learning methods for quantitative radiomic biomarkers. Sci Rep 2015; 5: 13087. Google Scholar
- 17. : An introduction to radiomics: an evolving cornerstone of precision medicine. In: Biomedical Texture Analysis. Edited by . London, UK: Elsevier Ltd 2017; pp 223. Google Scholar
- 18. : Public perceptions of artificial intelligence and robotics in medicine. J Endourol 2020; 34: 1041. Google Scholar
- 19. : Textural differences between renal cell carcinoma subtypes: machine learning-based quantitative computed tomography texture analysis with independent external validation. Eur J Radiol 2018; 107: 149. Google Scholar
- 20. : Machine learning-based quantitative texture analysis of CT images of small renal masses: differentiation of angiomyolipoma without visible fat from renal cell carcinoma. Eur Radiol 2017; 28: 1625. Google Scholar
- 21. : Kidney tumor location measurement using the C index method. J Urol 2010; 183: 1708. Link, Google Scholar
- 22. : A mathematical method to calculate tumor contact surface area: an effective parameter to predict renal function after partial nephrectomy. J Urol 2016; 196: 33. Link, Google Scholar
- 23. : Tumour contact surface area as a predictor of postoperative complications and renal function in patients undergoing partial nephrectomy for renal tumours. BJU Int 2019; 123: 639. Google Scholar
- 24. : Tumor contact surface area as a predictor of functional outcomes after standard partial nephrectomy: utility and limitations. Urology 2018; 116: 106. Google Scholar
- 25. : External validation of contact surface area as a predictor of postoperative renal function in patients undergoing partial nephrectomy. J Urol 2018; 199: 649. Link, Google Scholar
- 26. : Evaluation of surgery-related kidney volume loss to predict the outcomes of laparoscopic partial nephrectomy with segmental renal artery clamping. Int Urol Nephrol 2020; 52: 35. Google Scholar
- 27. : Comparison of 2 computed tomography-based methods to estimate preoperative and postoperative renal parenchymal volume and correlation with functional changes after partial nephrectomy. Urology 2015; 86: 80. Google Scholar
- 28. : The KiTS19 challenge data: 300 kidney tumor cases with clinical context, CT semantic segmentations, and surgical outcomes. arXiv 2019; e-prints arXiv:1904.00445. Google Scholar
- 29. : The state of the art in kidney and kidney tumor segmentation in contrast-enhanced CT imaging: results of the KiTS19 challenge. Med Image Anal 2020; 67: 101821. Google Scholar
Support: National Institutes of Health (NIH) and the National Science Foundation.
