setContinDiscrete: Discretize tumor volumes into intervals, where each interval...

In novasmedley/gbmSpm: Find predictive sequential patterns for clinical glioblastoma patients

Description

setContinDiscrete: convert volumes, measured in original units, to intervals.

setPercentDiscrete: convert volumes, measured in percent change, to intervals.

setRateDiscrete: convert volumes, measured in rate change, to intervals.

setBaselineDiscrete: convert baseline volumes, measured in original units, to intervals.

Usage

setContinDiscrete(volumes, categories = 10)

setPercentDiscrete(volumes, categories = 10)

setRateDiscrete(volumes, categories = 10)

setBaselineDiscrete(volumes, categories = 10)

Arguments

volumes	a dataframe, rows are tumor volumes
categories	number of categories or a vector with boundaries (all values outside the boundaries will be set to NA).

Details

Volumes are discretized into 10 intervals, where each is an event. Event names are named after the intervals. Note:

• intervals are based on distribution of values in volumes

• each interval or bin will be used as different categories of tumor volume

• also see prepTumorVol

volumes contains tumor measurement events as rows and at least the following columns:

• agent_detail = event attribute, i.e., measured tumor volume in mm3 char

Value

volumes with eventName as an additional column

See Also

discretize

Examples

data("fake_data")
vols <- fake_data$events[fake_data$events$agent=='measurement', ]

# format data
vols$startdate <- as.Date(as.character(vols$startdate), format='%Y-%m-%d')
vols$enddate <- as.Date(as.character(vols$enddate), format='%Y-%m-%d')
vols <- resetEventIDsToDays(vols)
vols <- removeData(vols) # get rid of excess details and empty or no data

vols_dist <- setContinDiscrete(vols)

data("fake_data")
vols <- fake_data$events[fake_data$events$agent=='measurement', ]

# format data
vols$startdate <- as.Date(as.character(vols$startdate), format='%Y-%m-%d')
vols$enddate <- as.Date(as.character(vols$enddate), format='%Y-%m-%d')
vols <- resetEventIDsToDays(vols)
vols <- removeData(vols) # get rid of excess details and empty or no data

# convert vol to correct type for all patients
vols <- split(vols, vols$iois) # list of dataframes, where each is a patient's set of tumor volumes
vols <- lapply(vols, function(v) setPercentChange(v))
vols <- do.call('rbind', vols)

# discretize
vols_dist <- setPercentDiscrete(vols)

novasmedley/gbmSpm documentation built on May 17, 2019, 10:39 a.m.