AutoScore-Ordinal: An Interpretable Machine Learning Framework for Generating Scoring Models for Ordinal Outcomes
AutoScore-Ordinal is a novel machine learning framework to automate the development of interpretable clinical scoring models for ordinal outcomes, which expands the original AutoScore framework for binary outcomes. AutoScore-Ordinal modifies the six modules of the AutoScore framework to handle ordinal outcomes: 1) variable ranking with machine learning, 2) variable transformation, 3) score derivation (now from the proportional odds model), 4) model selection, 5) domain knowledge-based score fine-tuning, and 6) performance evaluation (using the mean AUC (mAUC) across binary classifications). The AutoScore-Ordinal is elaborated in the manuscript “AutoScore-Ordinal: An Interpretable Machine Learning Framework for Generating Scoring Models for Ordinal Outcomes” and its flowchart is shown in the following figure, where blue shading indicate modifications from the original AutoScore framework. AutoScore-Ordinal could seamlessly generate risk scores using a parsimonious set of variables, which can be easily implemented and validated in clinical practice. Moreover, it enables users to build transparent and interpretable clinical scores quickly in a straightforward manner.
AutoScore-Ordinal has been merged with the AutoScore package. Please visit AutoScore bookdown page for a full tutorial.
-
Saffari SE, Ning Y, Xie F, Chakraborty B, Volovici V, Vaughan R, Ong MEH, Liu N, AutoScore-Ordinal: An interpretable machine learning framework for generating scoring models for ordinal outcomes, BMC Medical Research Methodology 2022; 22: 286.
-
Xie F, Chakraborty B, Ong MEH, Goldstein BA, Liu N. AutoScore: A machine learning-based automatic clinical score generator and its application to mortality prediction using electronic health records. JMIR Medical Informatics 2020; 8(10): e21798.
- Yilin Ning (Email: yilin.ning@duke-nus.edu.sg)
- Nan Liu (Email: liu.nan@duke-nus.edu.sg)
Install from GitHub or CRAN:
# From Github
install.packages("devtools")
library(devtools)
install_github(repo = "nliulab/AutoScore", build_vignettes = TRUE)
# From CRAN (recommended)
install.packages("AutoScore")Load AutoScore package:
library(AutoScore)