PheVis: Automatic Phenotyping of Electronic Health Record at Visit Resolution

Description

Using Electronic Health Record (EHR) is difficult because most of the time the true characteristic of the patient is not available. Instead we can retrieve the International Classification of Disease code related to the disease of interest or we can count the occurrence of the Unified Medical Language System. None of them is the true phenotype which needs chart review to identify. However chart review is time consuming and costly. 'PheVis' is an algorithm which is phenotyping (i.e identify a characteristic) at the visit level in an unsupervised fashion. It can be used for chronic or acute diseases. An example of how to use 'PheVis' is available in the vignette. Basically there are two functions that are to be used: 'train_phevis()' which trains the algorithm and 'test_phevis()' which get the predicted probabilities. The detailed method is described in preprint by Ferté et al. (2020) doi:10.1101/2020.06.15.20131458.

Author(s)

Maintainer: Thomas Ferte thomas.ferte@u-bordeaux.fr

Authors:

• Boris P. Hejblum boris.hejblum@u-bordeaux.fr

boot_df

Description

Sample rows with replacement from a matrix

Usage

boot_df(x_matrix, y_sur, ID = NULL, size = 10^5, seed = 1, prob = NULL)

Arguments

Value

A list with the sampled explanatory matrix and the sampled qualitative surrogate (y_sur)

build_qantsur

Description

build quantile threshold based on icd variables and omega constant

Usage

build_qantsur(df, var.icd, omega)

Arguments

Value

A numeric vector with the thresholds for the extrema populations.

build_quali

Description

build_quali

Usage

build_quali(x, p, q)

Arguments

Value

The qualitative surrogate (x in three categories) defining the extrema populations

check_arg_test_phevis

Description

Function to check arguments passed to test_phevis()

Usage

check_arg_test_phevis(
  train_param,
  df_test,
  surparam,
  model,
  START_DATE,
  PATIENT_NUM,
  ENCOUNTER_NUM
)

Arguments

Value

No return value, stop the code execution if one condition is not met.

check_arg_train_phevis

Description

Function to check arguments passed to train_phevis()

Usage

check_arg_train_phevis(
  half_life,
  df,
  START_DATE,
  PATIENT_NUM,
  ENCOUNTER_NUM,
  var_vec,
  main_icd,
  main_cui,
  rf,
  p.noise,
  bool_SAFE,
  omega,
  GS
)

Arguments

Value

No return value, stop the code execution if one condition is not met.

cum_lag

Description

helpful function to cumulate information.

Usage

cum_lag(x, n_lag)

Arguments

Value

return numeric vector.

Control data for test

Description

Simulated dataset for PheVis phenotyping.

Usage

data(data_perf)

Format

An object of class numeric of length 2.

PheVis simulated dataset

Description

Simulated dataset for PheVis phenotyping.

Usage

data(data_phevis)

Format

An object of class data.frame with 19659 rows and 15 columns.

expcorrectC

Description

c++ function to compute exponential cumulation of information.

Usage

expcorrectC(mat, diffdate, lambda)

Arguments

Details

expcorrectC

Value

A matrix corresponding to the mat argument with cumulated exponential decay

fct_surrogate_quanti

Description

Compute the quantitative surrogate and then apply thresholds to get the qualitative surrogate.

Usage

fct_surrogate_quanti(
  main_icd,
  main_cui,
  df,
  half_life,
  date,
  patient_id,
  encounter_id,
  omega = 2,
  param = NULL
)

Arguments

Value

A list

• table - Main result: data.frame with the rolling variables and the surrogates

• param - the parameters for the standardisation of ICD and CUI

• roll_all - a subset of table with the rolling variables only

• quantile_vec - the quantile defining the extrema populations

ggindividual_plot

Description

Plot individual predictions.

Usage

ggindividual_plot(subject, time, gold_standard, prediction)

Arguments

Value

a ggplot graph

Examples

ggindividual_plot(subject = rep(1,10),
  time = 1:10,
  gold_standard = c(0,0,1,1,0,0,1,1,0,0),
  prediction = runif(n = 10, min = 0, max = 1))

matrix_exp_smooth

Description

Function to accumulate the information with exponential decay.

Usage

matrix_exp_smooth(half_life, df, date, patient_id, encounter_id)

Arguments

Value

A data.frame object with both the raw variables and the accumulated ones.

noising

Description

Noise a matrix

Usage

noising(X_boot, p = 0.3)

Arguments

Value

A noised matrix

norm_var

Description

Standardize a numeric variable

Usage

norm_var(x)

Arguments

Value

The standardized variable

phenorm_longit_fit

Description

Apply simplified 'PheNorm' algorithm on longitudinal data with bootstrap and noise.

Usage

phenorm_longit_fit(
  x_matrix,
  y_sur,
  ID,
  size = 10^5,
  seed = 1,
  p.noise = 0.3,
  do_sampling = TRUE,
  do_noise = TRUE,
  prob = NULL,
  calc.prob = TRUE,
  nAGQ = 0,
  glmer.control = glmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
)

Arguments

Value

A list with the fixed effects, the predicted responses and the model used (mixed effect or logistic regression)

phenorm_longit_simpl

Description

'PheNorm' like function adapted to longitudinal data.

Usage

phenorm_longit_simpl(
  df,
  var_surrogate,
  surrogates_quali,
  id_rnd,
  rf = FALSE,
  ntree = 100,
  bool_weight = FALSE,
  p.noise = 0.3,
  bool_SAFE = TRUE,
  size = 10^5
)

Arguments

Value

A list with the logistic model, the random forest model, the variables selected for prediction and the predictions

pred_lme4model

Description

function to predict probability from 'lme4' or 'glm' objects

Usage

pred_lme4model(model = NULL, fe.model = NULL, df)

Arguments

Value

A vector of the predictions

pretty_cv.glmnet

Description

Train a 'glmnet' with cross validation (cv) model and return convenient results (model and results with non zero coefficients)

Usage

pretty_cv.glmnet(
  x_glmnet,
  y,
  alpha = 1,
  family = "binomial",
  s = "lambda.1se",
  weights = rep(1, nrow(x_glmnet)),
  ...
)

Arguments

Value

A list with the model, the coefficient associated with variables and the selected variables.

roll_time_sum

Description

Compute the cumulated information of what happened in past month and past year.

Usage

roll_time_sum(
  id,
  id_encounter,
  var,
  start_date,
  win_size1 = 30,
  win_size2 = 365,
  name1 = "cum_month",
  name2 = "cum_year"
)

Arguments

Value

A dataframe containing the rolling variables.

rolling_var

Description

Compute rolling variables (last visit, last 5 visits, last month and last year)

Usage

rolling_var(id, var, start_date, id_encounter)

Arguments

Value

A dataframe containing the rolling variables.

safe_selection

Description

Select the variables from dataframe by removing the rare variables and apply 'SAFE' on it.

Usage

safe_selection(
  df,
  var_surrogate,
  surrogate_quali,
  threshold = 0.05,
  alpha = 0.5,
  remove_var_surrogate = TRUE,
  bool_weight = FALSE,
  ...
)

Arguments

Value

A list

• glmnet_model - A list of three elements: the cv.glmnet fitted model, the coefficients of non zero variables and the vector of non zero coefficient variables.

• important_var - A vector with the variables used for the surrogate and the non zero variables.

• surrogate_quali - The surrogate_quali argument.

sur_exp_smooth

Description

Function to cumulate surrogate with exponential decay

Usage

sur_exp_smooth(half_life, sur, date, patient_id, encounter_id)

Arguments

Value

A dataframe with the cumulated surrogate.

test_phevis

Description

test_phevis

Usage

test_phevis(
  train_param,
  df_test,
  surparam,
  model,
  START_DATE,
  PATIENT_NUM,
  ENCOUNTER_NUM
)

Arguments

Value

A dataframe with the predictions.

Examples

library(dplyr)
library(PRROC)
PheVis::data_phevis
PheVis::data_perf

var_vec <- c(paste0("var",1:10), "mainCUI", "mainICD")
main_icd <- "mainICD"
main_cui <- "mainCUI"
GS <- "PR_state"
half_life <- Inf

df <- data_phevis %>%
        mutate(ENCOUNTER_NUM = row_number(),
               time = round(as.numeric(time)))

trainsize <- 0.8*length(unique(df$subject))
trainid <- sample(x = unique(df$subject), size = trainsize)
testid <- unique(df$subject)[!unique(df$subject) %in% trainid]

df_train <- as.data.frame(df[df$subject %in% trainid,])
df_test <- as.data.frame(df[df$subject %in% testid,])

##### train and test model #####
train_model <- PheVis::train_phevis(half_life = half_life,
                                    df = df_train,
                                    START_DATE = "time",
                                    PATIENT_NUM = "subject",
                                    ENCOUNTER_NUM = "ENCOUNTER_NUM",
                                    var_vec = var_vec,
                                    main_icd = main_icd,
                                    main_cui = main_cui)

test_perf <- PheVis::test_phevis(train_param = train_model$train_param,
                                 df_test = df_test,
                                 START_DATE = "time",
                                 PATIENT_NUM = "subject",
                                 ENCOUNTER_NUM = "ENCOUNTER_NUM",
                                 surparam = train_model$surparam,
                                 model = train_model$model)

pr_curve <-PRROC::pr.curve(scores.class0 = test_perf$df_result$PREDICTION,
                           weights.class0 = df_test$PR_state)

roc_curve <- PRROC::roc.curve(scores.class0 = test_perf$df_result$PREDICTION,
                              weights.class0 = df_test$PR_state)

train_phevis

Description

Global function to train phevis model.

Usage

train_phevis(
  half_life,
  df,
  START_DATE,
  PATIENT_NUM,
  ENCOUNTER_NUM,
  var_vec,
  main_icd,
  main_cui,
  rf = TRUE,
  p.noise = 0.3,
  bool_SAFE = TRUE,
  omega = 2,
  GS = NULL
)

Arguments

Value

A list

• surparam - the parameters used to compute the surrogate

• model - the random intercept logistic regression

• df_train_result - the data.frame containing the output predictions

• train_param - parameters for the model training (variables used, main ICD and CUIS, half_life, gold standard)

Examples

library(dplyr)
PheVis::data_phevis
df <- data_phevis %>%
         mutate(ENCOUNTER_NUM = row_number(),
         time = round(as.numeric(time)))
model <- PheVis::train_phevis(half_life = Inf,
         df = df,
         START_DATE = "time",
         PATIENT_NUM = "subject",
         ENCOUNTER_NUM = "ENCOUNTER_NUM",
         var_vec = c(paste0("var",1:10), "mainCUI", "mainICD"),
         main_icd = "mainICD",
         main_cui = "mainCUI")