compute_auc_val_ord

Compute mean AUC based on validation set for plotting parsimony

A novel interpretable machine learning-based framework to automate the development of a clinical scoring model for predefined outcomes. Our novel framework consists of six modules: variable ranking with machine learning, variable transformation, score derivation, model selection, domain knowledge-based score fine-tuning, and performance evaluation.The details are described in our research paper<doi:10.2196/21798>. Users or clinicians could seamlessly generate parsimonious sparse-score risk models (i.e., risk scores), which can be easily implemented and validated in clinical practice. We hope to see its application in various medical case studies.

Feng Xie

AutoScore

An Interpretable Machine Learning-Based Automatic Clinical Score
Generator

Yilin Ning

Han Yuan

Mingxuan Liu

Siqi Li

Ehsan Saffari

Bibhas Chakraborty

Nan Liu

compute_auc_val_ord function

<dl><dt>train_set_1</dt>
<dd>Processed training set</dd>
<dt>validation_set_1</dt>
<dd>Processed validation set</dd>
<dt>variable_list</dt>
<dd>List of included variables</dd>
<dt>link</dt>
<dd>The link function used to model ordinal outcomes. Default is
<code>"logit"</code> for proportional odds model. Other options are
<code>"cloglog"</code> (proportional hazards model) and <code>"probit"</code>.</dd>
<dt>categorize</dt>
<dd>Methods for categorize continuous variables. Options include "quantile" or "kmeans"</dd>
<dt>quantiles</dt>
<dd>Predefined quantiles to convert continuous variables to categorical ones. Available if <code>categorize = "quantile"</code>.</dd>
<dt>max_cluster</dt>
<dd>The max number of cluster (Default: 5). Available if <code>categorize = "kmeans"</code>.</dd>
<dt>max_score</dt>
<dd>Maximum total score</dd></dl>

Arguments

Internal function: Compute mean AUC for ordinal outcomes based on
  validation set for plotting parsimony — compute_auc_val_ord

<dl>

<dt>train_set_1</dt>
<dd>Processed training set</dd>


<dt>validation_set_1</dt>
<dd>Processed validation set</dd>


<dt>variable_list</dt>
<dd>List of included variables</dd>


<dt>link</dt>
<dd>The link function used to model ordinal outcomes. Default is
<code>"logit"</code> for proportional odds model. Other options are
<code>"cloglog"</code> (proportional hazards model) and <code>"probit"</code>.</dd>


<dt>categorize</dt>
<dd>Methods for categorize continuous variables. Options include "quantile" or "kmeans"</dd>


<dt>quantiles</dt>
<dd>Predefined quantiles to convert continuous variables to categorical ones. Available if <code>categorize = "quantile"</code>.</dd>


<dt>max_cluster</dt>
<dd>The max number of cluster (Default: 5). Available if <code>categorize = "kmeans"</code>.</dd>


<dt>max_score</dt>
<dd>Maximum total score</dd>

</dl>

compute_auc_val_ord: Internal function: Compute mean AUC for ordinal outcomes based on validation set for plotting parsimony

Description

Usage

Value

Arguments