discretizeKPS: Discretize KPS according to expert opinion

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

Description

KPS is categorized into UNCH, SigDECR, SigINCR, DECR, and INCR.

Usage

discretizeKPS(kps)

Arguments

kps	a dataframe of KPS scores, with columns: startdate date of KPS evaluation) agent_detail KPS

Details

KPS categories are

• initial: first KPS in sequence

• UNCH: unchanged, KPS is the same as the previous KPS

• SigDECR: significant decrease, KPS drops to <= 60 OR KPS was already <= 60 and drops further

• SigINCR: significant increase, KPS is > 60 AND previous KPS was <=60

• DECR: decrease, KPS drops but remains above 60 AND psrevious KPS was above 60

• INCR: increase, KPS improves AND is above 60 AND previous KPS was above 60

Value

A dataframe with original data kps plus discretized KPS categories in eventName column.

Examples

data("fake_data")

# get one person's set of KPS exams
kps_set <- fake_data$events[fake_data$events$agent=='KPS' & fake_data$events$iois=='2' , ]

# convert to categorical
kps_set_discretized <- discretizeKPS(kps_set)

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