---
title: "Example 2: Real EHR Data"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Example 2: Real EHR Data}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r}
library(PheCAP)
```

Load Data.

```{r}
data(ehr_data)
data <- PhecapData(ehr_data, "healthcare_utilization", "label", 0.4, patient_id = "patient_id")
data
```

Specify the surrogate used for surrogate-assisted feature extraction (SAFE). The typical way is to specify a main ICD code, a main NLP CUI, as well as their combination. In some cases one may want to define surrogate through lab test. The default lower_cutoff is 1, and the default upper_cutoff is 10. Feel free to change the cutoffs based on domain knowledge.

```{r}
surrogates <- list(
  PhecapSurrogate(
    variable_names = "main_ICD",
    lower_cutoff = 1, upper_cutoff = 10),
  PhecapSurrogate(
    variable_names = "main_NLP",
    lower_cutoff = 1, upper_cutoff = 10),
  PhecapSurrogate(
    variable_names = c("main_ICD", "main_NLP"),
    lower_cutoff = 1, upper_cutoff = 10))
```

Run surrogate-assisted feature extraction (SAFE) and show result.

```{r}
system.time(feature_selected <- phecap_run_feature_extraction(data, surrogates))
```

```{r}
feature_selected
```

Train phenotyping model and show the fitted model, with the AUC on the training set as well as random splits

```{r}
suppressWarnings(model <- phecap_train_phenotyping_model(data, surrogates, feature_selected))
model
```

Validate phenotyping model using validation label, and show the AUC and ROC

```{r}
validation <- phecap_validate_phenotyping_model(data, model)
validation
round(validation$valid_roc[validation$valid_roc[, "FPR"] <= 0.2, ], 3)
```

```{r}
phecap_plot_roc_curves(validation)
```

Apply the model to all the patients to obtain predicted phenotype.

```{r}
phenotype <- phecap_predict_phenotype(data, model)
idx <- which.min(abs(validation$valid_roc[, "FPR"] - 0.05))
cut.fpr95 <- validation$valid_roc[idx, "cutoff"]
case_status <- ifelse(phenotype$prediction >= cut.fpr95, 1, 0)
predict.table <- cbind(phenotype, case_status)
predict.table[1:10, ]
```