J Neurol Surg B Skull Base 2018; 79(02): 123-130
DOI: 10.1055/s-0037-1604393
Original Article
Georg Thieme Verlag KG Stuttgart · New York

Using a Guided Machine Learning Ensemble Model to Predict Discharge Disposition following Meningioma Resection

Whitney E. Muhlestein
1   Department of Neurosurgery, Vanderbilt University Medical Center, Vanderbilt University, Nashville, Tennessee, United States
,
Dallin S. Akagi
2   DataRobot, Inc., Boston, Massachusetts, United States
,
Justiss A. Kallos
1   Department of Neurosurgery, Vanderbilt University Medical Center, Vanderbilt University, Nashville, Tennessee, United States
,
Peter J. Morone
1   Department of Neurosurgery, Vanderbilt University Medical Center, Vanderbilt University, Nashville, Tennessee, United States
,
Kyle D. Weaver
1   Department of Neurosurgery, Vanderbilt University Medical Center, Vanderbilt University, Nashville, Tennessee, United States
,
Reid C. Thompson
1   Department of Neurosurgery, Vanderbilt University Medical Center, Vanderbilt University, Nashville, Tennessee, United States
,
Lola B. Chambless
1   Department of Neurosurgery, Vanderbilt University Medical Center, Vanderbilt University, Nashville, Tennessee, United States
› Author Affiliations
Further Information

Publication History

17 January 2017

14 June 2017

Publication Date:
08 August 2017 (online)

Preview

Abstract

Objective Machine learning (ML) algorithms are powerful tools for predicting patient outcomes. This study pilots a novel approach to algorithm selection and model creation using prediction of discharge disposition following meningioma resection as a proof of concept.

Materials and Methods A diversity of ML algorithms were trained on a single-institution database of meningioma patients to predict discharge disposition. Algorithms were ranked by predictive power and top performers were combined to create an ensemble model. The final ensemble was internally validated on never-before-seen data to demonstrate generalizability. The predictive power of the ensemble was compared with a logistic regression. Further analyses were performed to identify how important variables impact the ensemble.

Results Our ensemble model predicted disposition significantly better than a logistic regression (area under the curve of 0.78 and 0.71, respectively, p = 0.01). Tumor size, presentation at the emergency department, body mass index, convexity location, and preoperative motor deficit most strongly influence the model, though the independent impact of individual variables is nuanced.

Conclusion Using a novel ML technique, we built a guided ML ensemble model that predicts discharge destination following meningioma resection with greater predictive power than a logistic regression, and that provides greater clinical insight than a univariate analysis. These techniques can be extended to predict many other patient outcomes of interest.

Supplementary Material