Comparison of Cox Regression With Other Methods for Determining Prediction Models and Nomogramsстатья из журнала
Аннотация: No AccessJournal of UrologyOverview Consensus Statement1 Dec 2003Comparison of Cox Regression With Other Methods for Determining Prediction Models and Nomograms MICHAEL W. KATTAN MICHAEL W. KATTANMICHAEL W. KATTAN View All Author Informationhttps://doi.org/10.1097/01.ju.0000094764.56269.2dAboutFull TextPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookLinked InTwitterEmail Abstract Purpose: There is controversy as to whether artificial neural networks and other machine learning methods provide predictions that are more accurate than those provided by traditional statistical models when applied to censored data. Materials and Methods: Several machine learning prediction methods are compared with Cox proportional hazards regression using 3 large urological datasets. As a measure of predictive ability, discrimination that is similar to an area under the receiver operating characteristic curve is computed for each. Results: In all 3 datasets Cox regression provided comparable or superior predictions compared with neural networks and other machine learning techniques. In general, this finding is consistent with the literature. Conclusions: Although theoretically attractive, artificial neural networks and other machine learning techniques do not often provide an improvement in predictive accuracy over Cox regression. References 1 : A catalog of prostate cancer nomograms. J Urol2001; 165: 1562. Link, Google Scholar 2 : A simulation of factors affecting machine learning techniques: an examination of partitioning and class proportions. Omega Int J Mgmt Sci2000; 28: 501. Google Scholar 3 : The predictive accuracy of computer-based classification decision techniques. A review and research directions. Omega Int J Mgmt Sci1998; 26: 467. Google Scholar 4 : Comparison of artificial neural networks with other statistical approaches: results from medical data sets. Cancer2001; 91: 1636. Google Scholar 5 : Pretreatment nomogram for predicting freedom from recurrence after permanent prostate brachytherapy in prostate cancer. Urology2001; 58: 393. Google Scholar 6 : A postoperative prognostic nomogram for renal cell carcinoma. J Urol2001; 166: 63. Link, Google Scholar 7 : Pretreatment nomogram for predicting the outcome of three-dimensional conformal radiotherapy in prostate cancer. J Clin Oncol2000; 18: 3352. Google Scholar 8 : Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med1996; 15: 361. Google Scholar 9 : Regression Modeling Strategies With Applications to Linear Models, Logistic Regression, and Survival Analysis. New York: Springer-Verlag2001: 568. Google Scholar 10 : rpart: Recursive partitioning and regression trees. http://lib.stat.cmu.edu/S/rpart. Google Scholar 11 : Regression trees for censored data. Biometrics1988; 44: 35. Google Scholar 12 : Classification and Regression Trees. Monterey: Wadsworth1984. Google Scholar 13 : Polychotomous regression. J Am Stat Assoc1997; 92: 117. Google Scholar 14 : Martingale-based residuals for survival models. Biometrika1990; 77: 147. Google Scholar 15 : Neural networks as statistical methods in survival analysis. In: Artificial Neural Networks: Prospects for Medicine. Edited by . Austin, Texas: Landes Biosciences1997: 1. Google Scholar 16 : Modern Applied Statistics with S. New York: Springer-Verlag2002: 495. Google Scholar 17 : Experiments to determine whether recursive partitioning (CART) or an artificial neural network overcomes theoretical limitations of Cox proportional hazards regression. Comput Biomed Res1998; 31: 363. Google Scholar 18 : Comparing tumor staging and grading systems: a case study and a review of the issues, using thymoma as a model. Stat Med2000; 19: 1997. Google Scholar 19 : Evaluating the yield of medical tests. JAMA1982; 247: 2543. Google Scholar 20 : Artificial neural networks for diagnosis and prognosis in prostate cancer. Semin Urol Oncol2002; 20: 89. Google Scholar From the Health Outcomes Research Group, Department of Epidemiology and Biostatistics, and Department of Urology, Memorial Sloan-Kettering Cancer Center, New York, New York© 2003 by American Urological Association, Inc.FiguresReferencesRelatedDetailsCited ByKattan M (2018) Editorial CommentJournal of Urology, VOL. 200, NO. 6, (1377-1377), Online publication date: 1-Dec-2018.Labadie K, Okhunov Z, Akhavein A, Moreira D, Moreno-Palacios J, del Junco M, Okeke Z, Bird V, Smith A and Landman J (2018) Evaluation and Comparison of Urolithiasis Scoring Systems Used in Percutaneous Kidney Stone SurgeryJournal of Urology, VOL. 193, NO. 1, (154-159), Online publication date: 1-Jan-2015.Abbod M, Catto J, Linkens D and Hamdy F (2018) Application of Artificial Intelligence to the Management of Urological CancerJournal of Urology, VOL. 178, NO. 4, (1150-1156), Online publication date: 1-Oct-2007. Volume 170Issue 6SDecember 2003Page: S6-S10 Advertisement Copyright & Permissions© 2003 by American Urological Association, Inc.Keywordsneural network modelspredictive value of testsstatistical modelmachine learningMetricsAuthor Information MICHAEL W. KATTAN More articles by this author Expand All Advertisement PDF DownloadLoading ...
Год издания: 2003
Авторы: Michael W. Kattan
Издательство: Lippincott Williams & Wilkins
Источник: The Journal of Urology
Ключевые слова: Insurance, Mortality, Demography, Risk Management
Другие ссылки: The Journal of Urology (HTML)
PubMed (HTML)
PubMed (HTML)
Открытый доступ: closed
Том: 170
Выпуск: 6S