Scientific Methods for Health Sciences: HS 853: Homework Projects

Winter 2018: HS 853

Scientific Methods for Health Sciences: Special Topics

Homework 2

Problem 2.1: Use the University of Queensland Vital Signs Dataset data to do some QC charts and vital sign monitoring for several patients (e.g., cases 1, 3, 4). Some cases are available here. Filenames “UQ_VitalSignsData_Case0#.csv” (where “#” is a number indexing the case).

Generate some (clinically relevant) plots of the data.
Use Control Charts to identify possible critical time-points (events).

Note:


				VitalData <- read.csv('https://umich.instructure.com/files/366336/download?download_frd=1',as.is=T, header=T)

				VitalData$Time <- gsub("_", ".", VitalData$Time) # replace the “_” with “.” In the Time to indicate milliseconds
				head(strptime(VitalData$Time, "%H:%M:%OS"))

				library("reshape2"); library("ggplot2")

				VitalData_long <- melt(VitalData, id.vars="Time",

            			# The source columns

            			measure.vars=c("SpO2", "CO2", "ECG", "AWF", "inO2", "Pulse", "Pleth", "NBP_.Mean.", "awRR", "MAC", "Minute_Volume", "inN2O", "imCO2"),
 
            			# Name of the destination column that will identify the original

            			# column that the measurement came from

            			variable.name="Feature", 

            			value.name="Measurement")  # convert to long format

				ggplot(data=VitalData_long, aes(x=strptime(VitalData_long$Time, "%H:%M:%OS"), y=Measurement, colour=Feature)) +

    			geom_line()

Problem 2.2: Fit in a mixed effects model to the lung cancer data. Determine several covariates of clinical interest and interpret your findings in terms of the specific measures/results.

SOCR Resource Visitor number