A Factored Generalized Additive Model for Clinical Decision Support in the Operating Room

Logistic regression (LR) is widely used in clinical prediction because it is simple to deploy and easy to interpret. Nevertheless, being a linear model, LR has limited expressive capability and often has unsatisfactory performance. Generalized additive models (GAMs) extend the linear model with transformations of input features, though feature interaction is not allowed for all GAM variants. In this paper, we propose a factored generalized additive model (F-GAM) to preserve the model interpretability for targeted features while allowing a rich model for interaction with features fixed within the individual. We evaluate F-GAM on prediction of two targets, postoperative acute kidney injury and acute respiratory failure, from a single-center database. We find superior model performance of F-GAM in terms of AUPRC and AUROC compared to several other GAM implementations, random forests, support vector machine, and a deep neural network. We find that the model interpretability is good with results with high face validity.

[1]  S. Lage,et al.  Acute kidney injury prediction following elective cardiac surgery: AKICS Score. , 2007, Kidney international.

[2]  Wendong Ge,et al.  An Interpretable ICU Mortality Prediction Model Based on Logistic Regression and Recurrent Neural Networks with LSTM Units , 2018, AMIA.

[3]  Matt J. Kusner,et al.  Deep Manifold Traversal: Changing Labels with Convolutional Features , 2015, ArXiv.

[4]  Parisa Rashidi,et al.  Application of Machine Learning Techniques to High-Dimensional Clinical Data to Forecast Postoperative Complications , 2016, PloS one.

[5]  Yixin Chen,et al.  Optimal Action Extraction for Random Forests and Boosted Trees , 2015, KDD.

[6]  J. Hanley,et al.  The meaning and use of the area under a receiver operating characteristic (ROC) curve. , 1982, Radiology.

[7]  Sameer Antani,et al.  Visualization and Interpretation of Convolutional Neural Network Predictions in Detecting Pneumonia in Pediatric Chest Radiographs , 2018, Applied sciences.

[8]  Esslli Site,et al.  Probabilistic Models in the Study of Language , 2012 .

[9]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[10]  Jairo A. Gutiérrez,et al.  ISeeU: Visually interpretable deep learning for mortality prediction inside the ICU , 2019, J. Biomed. Informatics.

[11]  Muhan Zhang,et al.  Deep Embedding Logistic Regression , 2018, 2018 IEEE International Conference on Big Knowledge (ICBK).

[12]  Matthew M. Churpek,et al.  The Development of a Machine Learning Inpatient Acute Kidney Injury Prediction Model* , 2018, Critical care medicine.

[13]  Wojciech Samek,et al.  Methods for interpreting and understanding deep neural networks , 2017, Digit. Signal Process..

[14]  E. Cook,et al.  Preoperative renal risk stratification. , 1997, Circulation.

[15]  Franco Turini,et al.  A Survey of Methods for Explaining Black Box Models , 2018, ACM Comput. Surv..

[16]  May D. Wang,et al.  Interpretable Predictions of Clinical Outcomes with An Attention-based Recurrent Neural Network , 2017, BCB.

[17]  Yan Liu,et al.  Interpretable Deep Models for ICU Outcome Prediction , 2016, AMIA.

[18]  J. Kellum,et al.  Diagnosis, evaluation, and management of acute kidney injury: a KDIGO summary (Part 1) , 2013, Critical Care.

[19]  William J. E. Potts,et al.  Generalized additive neural networks , 1999, KDD '99.

[20]  W. Henderson,et al.  Multifactorial Risk Index for Predicting Postoperative Respiratory Failure in Men After Major Noncardiac Surgery , 2000, Annals of surgery.

[21]  K. Tremper,et al.  Development and Validation of an Acute Kidney Injury Risk Index for Patients Undergoing General Surgery: Results from a National Data Set , 2009, Anesthesiology.

[22]  Yixin Chen,et al.  Density-based logistic regression , 2013, KDD.

[23]  L. Neumayer,et al.  Multivariable predictors of postoperative respiratory failure after general and vascular surgery: results from the patient safety in surgery study. , 2007, Journal of the American College of Surgeons.

[24]  Xiang Fang,et al.  Development and validation of a risk calculator predicting postoperative respiratory failure. , 2011, Chest.

[25]  Rohit J. Kate,et al.  Prediction and detection models for acute kidney injury in hospitalized older adults , 2016, BMC Medical Informatics and Decision Making.