Data-driven Temporal Prediction of Surgical Site Infection

Analysis of data from Electronic Health Records (EHR) presents unique challenges, in particular regarding nonuniform temporal resolution of longitudinal variables. A considerable amount of patient information is available in the EHR - including blood tests that are performed routinely during inpatient follow-up. These data are useful for the design of advanced machine learning-based methods and prediction models. Using a matched cohort of patients undergoing gastrointestinal surgery (101 cases and 904 controls), we built a prediction model for post-operative surgical site infections (SSIs) using Gaussian process (GP) regression, time warping and imputation methods to manage the sparsity of the data source, and support vector machines for classification. For most blood tests, wider confidence intervals after imputation were obtained in patients with SSI. Predictive performance with individual blood tests was maintained or improved by joint model prediction, and non-linear classifiers performed consistently better than linear models.

[1]  Mary R. Kwaan,et al.  An Apgar score for surgery. , 2007, Journal of the American College of Surgeons.

[2]  Bernhard Schölkopf,et al.  New Support Vector Algorithms , 2000, Neural Computation.

[3]  Yan Liu,et al.  Granger Causality Analysis in Irregular Time Series , 2012, SDM.

[4]  Sergio Huerta,et al.  Surgical site infections after colorectal surgery: do risk factors vary depending on the type of infection considered? , 2007, Surgery.

[5]  Janet L. Peacock,et al.  Oxford Handbook of Medical Statistics , 2010 .

[6]  Melinda Maggard-Gibbons,et al.  Surgical site infections following ambulatory surgery procedures. , 2014, JAMA.

[7]  Clifford Y. Ko,et al.  Reliability of Evaluating Hospital Quality by Colorectal Surgical Site Infection Type , 2013, Annals of surgery.

[8]  A. Hill,et al.  Systematic review and meta‐analysis of use of serum C‐reactive protein levels to predict anastomotic leak after colorectal surgery , 2014, The British journal of surgery.

[9]  Ron Kohavi,et al.  Wrappers for Feature Subset Selection , 1997, Artif. Intell..

[10]  José Luis Rojo-Álvarez,et al.  Support Vector Feature Selection for Early Detection of Anastomosis Leakage From Bag-of-Words in Electronic Health Records , 2016, IEEE Journal of Biomedical and Health Informatics.

[11]  Shahram Ebadollahi,et al.  Data-driven approach for assessing utility of medical tests using electronic medical records , 2015, J. Biomed. Informatics.

[12]  Walter Hellinger,et al.  Prevalence of Healthcare-Associated Infections in Acute Care Hospitals in Jacksonville, Florida , 2012, Infection Control & Hospital Epidemiology.

[13]  W. Cromwell,et al.  Predicting Surgical Site Infections in Real-Time , 2014 .

[14]  Paris P. Tekkis,et al.  Comparison of POSSUM Scoring Systems and the Surgical Risk Scale in Patients Undergoing Surgery for Complicated Diverticular Disease , 2006, Diseases of the colon and rectum.

[15]  Nicole A. Lazar,et al.  Statistical Analysis With Missing Data , 2003, Technometrics.

[16]  José Luis Rojo-Álvarez,et al.  On the differential benchmarking of promotional efficiency with machine learning modelling (II): Practical applications , 2012, Expert Syst. Appl..

[17]  D. Hosmer,et al.  Model‐Building Strategies and Methods for Logistic Regression , 2005 .

[18]  Masoud Nikravesh,et al.  Feature Extraction - Foundations and Applications , 2006, Feature Extraction.

[19]  Alexander J. Smola,et al.  Learning with kernels , 1998 .

[20]  Renee Demski,et al.  Implementation of a surgical comprehensive unit-based safety program to reduce surgical site infections. , 2012, Journal of the American College of Surgeons.

[21]  A. Senagore,et al.  Use of the American College of Surgeons NSQIP Surgical Risk Calculator for Laparoscopic Colectomy: how good is it and how can we improve it? , 2015, Journal of the American College of Surgeons.

[22]  T. Lasko,et al.  Computational Phenotype Discovery Using Unsupervised Feature Learning over Noisy, Sparse, and Irregular Clinical Data , 2013, PloS one.

[23]  Suchi Saria,et al.  Developing Predictive Models Using Electronic Medical Records: Challenges and Pitfalls , 2013, AMIA.

[24]  G. de Lissovoy,et al.  Surgical site infection: incidence and impact on hospital utilization and treatment costs. , 2009, American journal of infection control.

[25]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[26]  Aixia Guo,et al.  Gene Selection for Cancer Classification using Support Vector Machines , 2014 .

[27]  José Luis Rojo-Álvarez,et al.  On the differential benchmarking of promotional efficiency with machine learning modeling (I): Principles and statistical comparison , 2012, Expert Syst. Appl..

[28]  W J Martone,et al.  Surgical wound infection rates by wound class, operative procedure, and patient risk index. National Nosocomial Infections Surveillance System. , 1991, The American journal of medicine.

[29]  C. van Walraven,et al.  The Surgical Site Infection Risk Score (SSIRS): A Model to Predict the Risk of Surgical Site Infections , 2013, PloS one.

[30]  E. Lawson,et al.  Risk factors for superficial vs deep/organ-space surgical site infections: implications for quality improvement initiatives. , 2013, JAMA surgery.

[31]  Alain Rakotomamonjy,et al.  Variable Selection Using SVM-based Criteria , 2003, J. Mach. Learn. Res..