+
+
+
+
+
+
+
+
+
+
+
+
+
+
data_dir <- '~/Box/CogNeuroLab/Wearables/data/'
+fit <- read_csv(paste0(data_dir, 'fitbit/fitbitSleepStatsGroup.csv'))
+act <- read_csv(paste0(data_dir, 'actiwatch/wa_act_combined.csv'))
+act_cl <- read_csv(paste0(data_dir, 'actiwatch/actiware_manual_clean/wa_act_combined.csv'))
+
+df_cx <- readr::read_csv(paste0(data_dir, 'df_main.csv'))
+
+head(act)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
df_cx %>%
+ mutate(age_group = recode(age_group, "Young Adults", "Older Adults")) %>%
+ mutate(handedness = recode(handedness, "Left", "Right", "Both")) %>%
+ mutate(actiwatch_arm = recode(actiwatch_arm, "Left", "Right")) %>%
+ mutate(dominant = ifelse(handedness == actiwatch_arm, 1, 0)) %>%
+ group_by(age_group, dominant) %>%
+ summarise(n = n())
+
+
+
+
+
+
+
+
+df_cx %>%
+ mutate(age_group = recode(age_group, "Young Adults", "Older Adults")) %>%
+ mutate(handedness = recode(handedness, "Left", "Right", "Both")) %>%
+ mutate(actiwatch_arm = recode(actiwatch_arm, "Left", "Right")) %>%
+ filter(handedness == "Right") %>%
+ mutate(dominant = ifelse(handedness == actiwatch_arm, 1, 0)) %>%
+ group_by(age_group, dominant) %>%
+ summarise(n = n())
+
+
+
+
+
+
+
+
+df_cx %>%
+ mutate(age_group = recode(age_group, "Young Adults", "Older Adults")) %>%
+ mutate(handedness = recode(handedness, "Left", "Right", "Both")) %>%
+ mutate(actiwatch_arm = recode(actiwatch_arm, "Left", "Right")) %>%
+ filter(handedness == "Left") %>%
+ mutate(dominant = ifelse(handedness == actiwatch_arm, 1, 0)) %>%
+ group_by(age_group, dominant) %>%
+ summarise(n = n())
+
+
+
+
+
+
+
+
+
+
+
+
exclude = c('10023', '10076', '20182', '20184', '20176', '10151', '10139')
+df <- df[!df$subject %in% exclude, ]
+
+df_means <- df %>%
+ select(subject_id, analysis_name_cl, where(is.numeric)) %>%
+ group_by(subject, analysis_name_cl) %>%
+ summarise(across(where(is.numeric), ~ mean(.x, na.rm = TRUE)))
+
+head(df_means)
+
+
+
+
+
+
+
+
NA
+
+
+
+
+
Correlations
+
+
+
+
library(corrr)
+library(purrr)
+library(dplyr)
+
+df %>%
+ group_by(subject) %>%
+ select(where(is.numeric)) %>%
+ group_by(subject) %>%
+ summarise_at(vars(-subject_id), funs(mean(., na.rm=TRUE))) %>%
+ corrr::correlate(method = 'spearman') %>%
+ corrr::focus(matches('sleep_duration'), mirror = TRUE)
+
+
+
+
+
+
+
+
NA
+
+
+
+
+
+
+
df %>%
+ group_by(subject) %>%
+ filter(handedness == 2) %>%
+ select(where(is.numeric)) %>%
+ group_by(subject) %>%
+ summarise_at(vars(-subject_id), funs(mean(., na.rm=TRUE))) %>%
+ corrr::correlate(method = 'spearman') %>%
+ corrr::focus(matches('sleep_duration'), mirror = TRUE)
+
+
+
+
+
+
+
+
+
+
+
+
df %>% #all
+ group_by(subject) %>%
+ select(where(is.numeric)) %>%
+ group_by(subject) %>%
+ summarise_at(vars(-subject_id), funs(mean(., na.rm=TRUE))) %>%
+ corrr::correlate(method = 'spearman') %>%
+ corrr::focus(matches('bed'), mirror = TRUE)
+
+
+
+
+
+
+
+
+
+
+
+
df %>% #right handed only
+ group_by(subject) %>%
+ filter(handedness == 2) %>%
+ select(where(is.numeric)) %>%
+ group_by(subject) %>%
+ summarise_at(vars(-subject_id), funs(mean(., na.rm=TRUE))) %>%
+ corrr::correlate(method = 'spearman') %>%
+ corrr::focus(matches('bed'), mirror = TRUE)
+
+
+
+
+
+
+
+
NA
+
+
+
+
+
+
+
library(RColorBrewer)
+
+df_means %>%
+ ggplot(aes(x = sleep_duration_cl, y = sleep_duration_fit, color = analysis_name_cl)) +
+ geom_point() +
+ stat_smooth(method = 'lm') +
+ scale_color_brewer(palette = 'Set1')
+
+
+
+
+
+
NA
+NA
+
+
+
+
+
+
+
df_means %>%
+ ggplot(aes(x = time_in_bed_cl, y = time_in_bed_fit, color = analysis_name_cl)) +
+ geom_point() +
+ stat_smooth(method = 'lm') +
+ scale_color_brewer(palette = 'Set1')
+
+
+
+
+
+
+
+
+
+
+
# excluded subject 10139
+df_means %>%
+ ggplot(aes(x = time_in_bed_cl)) +
+ geom_boxplot()
+
+
+
+
+
+
+df_means$subject[df_means$time_in_bed_cl > 700]
+
+
+
[1] 20108
+
+
+
+
+
Repeated measures correlations
+
Documentation
+
Repeated Measures Correlation, Bakdash1 and Marusich, 2017
+
+
Time in Bed
+
+
+
+
library(rmcorr)
+
+colourCount = length(unique(df$subject))
+getPalette = colorRampPalette(brewer.pal(9, "Spectral"))
+
+my.rmc <- rmcorr(participant = subject, measure1 = time_in_bed_act, measure2 = time_in_bed_fit, dataset = df)
+plot(my.rmc, overall = TRUE)
+
+
+
+
+
+
+
+
+
+
+
ggplot2::ggplot(df, ggplot2::aes(x = time_in_bed_act, y = time_in_bed_fit, group = factor(subject),
+ color = factor(subject))) +
+ ggplot2::geom_point(ggplot2::aes(colour = factor(subject))) +
+ ggplot2::geom_line(ggplot2::aes(y = my.rmc$model$fitted.values), linetype = 1) + xlab('Actiwatch 2.0') + ylab('Fitbit') + ggtitle('Time In Bed') + scale_color_manual(values = getPalette(colourCount)) + labs(color = 'Subject') + ggsave(paste0(data_dir, 'CARE2021/figures/timeinbedrmcorr.png'), height = 5, width = 8, dpi = 300)
+
+
+
+
+
+
+
+
+
+
Time in Bed: Cleaned Actiwatch data
+
+
+
+
my.rmc <- rmcorr(participant = subject, measure1 = time_in_bed_cl, measure2 = time_in_bed_fit, dataset = df[df$age_group == 1,])
+
+ggplot2::ggplot(df[df$age_group == 1,], ggplot2::aes(x = time_in_bed_cl, y = time_in_bed_fit, group = factor(subject),
+ color = factor(subject))) +
+ ggplot2::geom_point(ggplot2::aes(colour = factor(subject))) +
+ ggplot2::geom_line(ggplot2::aes(y = my.rmc$model$fitted.values), linetype = 1) + xlab('Actiwatch 2.0') + ylab('Fitbit') + ggtitle('Time In Bed') + scale_color_manual(values = getPalette(colourCount)) + labs(color = 'Subject')
+
+
+
+
+
+
+
+
+
+
Time in bed: correlations
+
+
+
+
#raw
+rmcorr(subject, time_in_bed_act, time_in_bed_fit, df)
+
+
+
+Repeated measures correlation
+
+r
+0.1353624
+
+degrees of freedom
+721
+
+p-value
+0.0002620058
+
+95% confidence interval
+0.06296976 0.2063381
+
+
+
rmcorr(subject, time_in_bed_act, time_in_bed_fit, filter(df, handedness == 2))
+
+
+
+Repeated measures correlation
+
+r
+0.1405546
+
+degrees of freedom
+671
+
+p-value
+0.0002544933
+
+95% confidence interval
+0.06556382 0.2139666
+
+
+
rmcorr(subject, time_in_bed_act, time_in_bed_fit, filter(df, (actiwatch_arm == 1) & (handedness == 2)))
+
+
+
+Repeated measures correlation
+
+r
+0.1816943
+
+degrees of freedom
+331
+
+p-value
+0.000865625
+
+95% confidence interval
+0.07537021 0.2839334
+
+
+
rmcorr(subject, time_in_bed_act, time_in_bed_fit, filter(df, (actiwatch_arm == 2) & (handedness == 2)))
+
+
+
+Repeated measures correlation
+
+r
+0.07252395
+
+degrees of freedom
+339
+
+p-value
+0.1815221
+
+95% confidence interval
+-0.03425988 0.1776706
+
+
+
# manually cleaned
+rmcorr(subject, time_in_bed_cl, time_in_bed_fit, df)
+
+
+
+Repeated measures correlation
+
+r
+0.1597513
+
+degrees of freedom
+721
+
+p-value
+1.590169e-05
+
+95% confidence interval
+0.08775993 0.2300826
+
+
+
rmcorr(subject, time_in_bed_cl, time_in_bed_fit, filter(df, handedness == 2))
+
+
+
+Repeated measures correlation
+
+r
+0.1674614
+
+degrees of freedom
+671
+
+p-value
+1.258939e-05
+
+95% confidence interval
+0.09295126 0.2401063
+
+
+
rmcorr(subject, time_in_bed_cl, time_in_bed_fit, filter(df, (actiwatch_arm == 1) & (handedness == 2)))
+
+
+
+Repeated measures correlation
+
+r
+0.1786162
+
+degrees of freedom
+331
+
+p-value
+0.001062066
+
+95% confidence interval
+0.07220616 0.2810059
+
+
+
rmcorr(subject, time_in_bed_cl, time_in_bed_fit, filter(df, (actiwatch_arm == 2) & (handedness == 2)))
+
+
+
+Repeated measures correlation
+
+r
+0.1483393
+
+degrees of freedom
+339
+
+p-value
+0.006062129
+
+95% confidence interval
+0.04249153 0.2508941
+
+
+
+
+
+
Sleep time
+
+
+
+
my.rmc <- rmcorr(participant = subject, measure1 = sleep_duration_act, measure2 = sleep_duration_fit, dataset = df)
+
+ggplot2::ggplot(df, ggplot2::aes(x = sleep_duration_act, y = sleep_duration_fit, group = factor(subject),
+ color = factor(subject))) +
+ ggplot2::geom_point(ggplot2::aes(colour = factor(subject))) +
+ ggplot2::geom_line(ggplot2::aes(y = my.rmc$model$fitted.values), linetype = 1) + xlab('Actiwatch 2.0') + ylab('Fitbit') + ggtitle('Time Asleep') + scale_color_manual(values = getPalette(colourCount)) + labs(color = 'Subject') + ggsave(paste0(data_dir, 'CARE2021/figures/sleeptimermcorr.png'), height = 5, width = 8, dpi = 300)
+
+
+
+
+
+
+
+
+
+
Sleep Time: Cleaned Actiwatch data
+
+
+
+
my.rmc <- rmcorr(participant = subject, measure1 = sleep_duration_cl, measure2 = sleep_duration_fit, dataset = df[df$age_group == 1,], nreps=500)
+
+ggplot2::ggplot(df[df$age_group == 1,], ggplot2::aes(x = sleep_duration_cl, y = sleep_duration_fit, group = factor(subject),
+ color = factor(subject))) +
+ ggplot2::geom_point(ggplot2::aes(colour = factor(subject))) +
+ ggplot2::geom_line(ggplot2::aes(y = my.rmc$model$fitted.values), linetype = 1) + xlab('Actiwatch 2.0') + ylab('Fitbit') + ggtitle('Time Asleep') + scale_color_manual(values = getPalette(colourCount)) + labs(color = 'Subject')
+
+
+
+
+
+
+
+
+
+
+
#raw
+rmcorr(subject, sleep_duration_act, sleep_duration_fit, df)
+
+
+
+Repeated measures correlation
+
+r
+0.2741189
+
+degrees of freedom
+721
+
+p-value
+6.280332e-14
+
+95% confidence interval
+0.2052111 0.3403224
+
+
+
rmcorr(subject, sleep_duration_act, sleep_duration_fit, filter(df, handedness == 2))
+
+
+
+Repeated measures correlation
+
+r
+0.2620646
+
+degrees of freedom
+671
+
+p-value
+4.950899e-12
+
+95% confidence interval
+0.190148 0.3311838
+
+
+
rmcorr(subject, sleep_duration_act, sleep_duration_fit, filter(df, (actiwatch_arm == 1) & (handedness == 2)))
+
+
+
+Repeated measures correlation
+
+r
+0.3734459
+
+degrees of freedom
+331
+
+p-value
+1.845944e-12
+
+95% confidence interval
+0.2767884 0.4626223
+
+
+
rmcorr(subject, sleep_duration_act, sleep_duration_fit, filter(df, (actiwatch_arm == 2) & (handedness == 2)))
+
+
+
+Repeated measures correlation
+
+r
+0.09787818
+
+degrees of freedom
+339
+
+p-value
+0.07105332
+
+95% confidence interval
+-0.008732025 0.2022883
+
+
+
# manually cleaned
+rmcorr(subject, sleep_duration_cl, sleep_duration_fit, df)
+
+
+
+Repeated measures correlation
+
+r
+0.3072433
+
+degrees of freedom
+721
+
+p-value
+2.854961e-17
+
+95% confidence interval
+0.2396034 0.371915
+
+
+
rmcorr(subject, sleep_duration_cl, time_in_bed_fit, filter(df, handedness == 2))
+
+
+
+Repeated measures correlation
+
+r
+0.291924
+
+degrees of freedom
+671
+
+p-value
+1.092137e-14
+
+95% confidence interval
+0.2211215 0.3596654
+
+
+
rmcorr(subject, sleep_duration_cl, time_in_bed_fit, filter(df, (actiwatch_arm == 1) & (handedness == 2)))
+
+
+
+Repeated measures correlation
+
+r
+0.3609073
+
+degrees of freedom
+331
+
+p-value
+1.110282e-11
+
+95% confidence interval
+0.2633529 0.4511551
+
+
+
rmcorr(subject, sleep_duration_cl, time_in_bed_fit, filter(df, (actiwatch_arm == 2) & (handedness == 2)))
+
+
+
+Repeated measures correlation
+
+r
+0.17582
+
+degrees of freedom
+339
+
+p-value
+0.001113192
+
+95% confidence interval
+0.07062353 0.2771487
+
+
+
+
+
+
+
rmcorr(subject, sleep_duration_act, time_in_bed_fit, df)
+
+
+
+Repeated measures correlation
+
+r
+0.2704232
+
+degrees of freedom
+721
+
+p-value
+1.393078e-13
+
+95% confidence interval
+0.2013845 0.3367884
+
+
+
+
+
+
+
Efficiency
+
+
+
+
my.rmc <- rmcorr(participant = subject, measure1 = efficiency_act, measure2 = efficiency_fit, dataset = df[df$age_group == 1,], nreps=500)
+
+ggplot2::ggplot(df[df$age_group == 1,], ggplot2::aes(x = efficiency_act, y = efficiency_fit, group = factor(subject),
+ color = factor(subject))) +
+ ggplot2::geom_point(ggplot2::aes(colour = factor(subject))) +
+ ggplot2::geom_line(ggplot2::aes(y = my.rmc$model$fitted.values), linetype = 1) + xlab('Actiwatch 2.0') + ylab('Fitbit') + ggtitle('Sleep Efficiency') + scale_color_manual(values = getPalette(colourCount)) + labs(color = 'Subject')
+
+
+
+
+
+
+
+
+
+
+
rmcorr(subject, efficiency_act, efficiency_fit, df)
+
+
+
+Repeated measures correlation
+
+r
+0.06349924
+
+degrees of freedom
+721
+
+p-value
+0.08797509
+
+95% confidence interval
+-0.009560122 0.1358843
+
+
+
+
+
+
Next steps:
+
Need correlation value for each subject, then see how that’s related to measures of subjective sleep quality, handedness, arm assignment, percent watch off, beiwe, fitbit device
+
+
+
+
corrdf <- df %>%
+ group_by(subject) %>%
+ summarize(sleep_time = cor(sleep_duration_act, sleep_duration_fit, method = 'spearman'), time_in_bed = cor(time_in_bed_act, time_in_bed_fit, method = 'spearman'))
+
+head(corrdf)
+
+
+
+
+
+
+
+
NA
+
+
+
+
+
dfcor <- merge(df, corrdf, by = 'subject')
+
+head(dfcor)
+
+
+
+
+
+
+
+
NA
+
+
+
+
+
+
+
unique(dfcor$subject[dfcor$time_in_bed < 0.5])
+
+
+
[1] 10011 10012 10014 10033 10038 10040 10045 10062 10140 10146 10147 10148 10149 20103 20105 20108 20113 20115
+[19] 20120 20122 20123 20124 20125 20127 20131 20142 20154 20155 20157
+
+
+
length(unique(dfcor$subject[dfcor$time_in_bed < 0.5]))
+
+
+
[1] 29
+
+
+
unique(dfcor$subject[dfcor$sleep_time < 0.5])
+
+
+
[1] 10011 10033 10038 10045 10140 10147 10148 10149 20103 20105 20113 20115 20120 20122 20123 20124 20125 20131
+[19] 20142 20154 20155
+
+
+
length(unique(dfcor$subject[dfcor$sleep_time < 0.5]))
+
+
+
[1] 21
+
+
+
+
+
+
+
+require(gridExtra)
+
+plot1 <- dfcor %>%
+ ggplot() +
+ geom_boxplot(aes(x = fitbit_device, y = sleep_time, fill = fitbit_device, group = fitbit_device)) +
+ ylab('Sleep Time Correlation (r)') +
+ scale_fill_brewer(palette = 'Set1')
+
+plot2 <- dfcor %>%
+ ggplot() +
+ geom_boxplot(aes(x = fitbit_device, y = time_in_bed, fill = fitbit_device, group = fitbit_device)) +
+ ylab('Time In Bed Correlation (r)') +
+ scale_fill_brewer(palette = 'Set1') + theme(axis.text.x = element_text(angle = 90))
+
+grid.arrange(plot1, plot2, ncol=2)
+
+
+
+
+
+
+
+
+
+
+
dfcor$handedness <- factor(dfcor$handedness, labels = c("Right", "Left", "Both"))
+
+plot1 <- dfcor %>%
+ filter(handedness != "Both") %>%
+ ggplot() +
+ geom_boxplot(aes(x = handedness, y = sleep_time, fill = handedness, group = handedness)) +
+ ylab('Sleep Time Correlation (r)') + xlab('Handedness') +
+ scale_fill_brewer(palette = 'Set1')
+
+plot2 <- dfcor %>%
+ filter(handedness != "Both") %>%
+ ggplot() +
+ geom_boxplot(aes(x = handedness, y = time_in_bed, fill = handedness, group = handedness)) +
+ ylab('Time In Bed Correlation (r)') + xlab('Handedness') +
+ scale_fill_brewer(palette = 'Set1')
+
+pnew <- arrangeGrob(plot1, plot2, ncol=2)
+ggsave("~/Desktop/handedness.png", pnew, height = 5, width = 10, units = "in")
+
+plot(pnew)
+
+
+
+
+
+
+
+
+
+
+
+
+
dfcor$actiwatch_arm <- factor(dfcor$actiwatch_arm, labels = c("Right", "Left"))
+
+plot1 <- dfcor %>%
+ filter(handedness != "Both") %>%
+ ggplot() +
+ geom_boxplot(aes(x = actiwatch_arm, y = sleep_time, fill = handedness, group = handedness)) +
+ ylab('Sleep Time Correlation (r)') +
+ scale_fill_brewer(palette = 'Set1') + xlab("Actiwatch Arm")
+
+plot2 <- dfcor %>%
+ filter(handedness != "Both") %>%
+ ggplot() +
+ geom_boxplot(aes(x = actiwatch_arm, y = time_in_bed, fill = handedness, group = handedness)) +
+ ylab('Time In Bed Correlation (r)') +
+ scale_fill_brewer(palette = 'Set1') + xlab("Actiwatch Arm")
+
+pnew <- arrangeGrob(plot1, plot2, ncol=2)
+ggsave("~/Desktop/actiwatchassignment.png", pnew, height = 5, width = 10, units = "in")
+
+plot(pnew)
+
+
+
+
+
+
+
+
+
+
+
+
+
dfcor$dominant_assignment <- factor(ifelse(as.character(dfcor$actiwatch_arm) == as.character(dfcor$handedness), 1, 0), labels = c("Dominant", "Non-Dominant"))
+
+df_means$dominant_assignment <- factor(ifelse(as.character(df_means$actiwatch_arm) == as.character(df_means$handedness), 1, 0), labels = c("Dominant", "Non-Dominant"))
+
+plot1 <- dfcor %>%
+ filter(handedness != "Both") %>%
+ ggplot() +
+ geom_boxplot(aes(x = dominant_assignment, y = sleep_time, fill = dominant_assignment, group = dominant_assignment)) +
+ ylab('Sleep Time Correlation (r)') +
+ scale_fill_brewer(palette = 'Set1') + xlab("Actiwatch Arm")
+
+plot2 <- dfcor %>%
+ filter(handedness != "Both") %>%
+ ggplot() +
+ geom_boxplot(aes(x = dominant_assignment, y = time_in_bed, fill = dominant_assignment, group = dominant_assignment)) +
+ ylab('Time In Bed Correlation (r)') +
+ scale_fill_brewer(palette = 'Set1') + xlab("Actiwatch Arm")
+
+pnew <- arrangeGrob(plot1, plot2, ncol=2)
+ggsave("~/Desktop/actiwatchassignment_dominant.png", pnew, height = 5, width = 10, units = "in")
+
+summary(lm(sleep_duration_act ~ sleep_duration_fit + dominant_assignment, data = df_means))
+
+
+
+Call:
+lm(formula = sleep_duration_act ~ sleep_duration_fit + dominant_assignment,
+ data = df_means)
+
+Residuals:
+ Min 1Q Median 3Q Max
+-118.901 -42.467 4.308 41.687 176.725
+
+Coefficients:
+ Estimate Std. Error t value Pr(>|t|)
+(Intercept) 469.1405 87.8572 5.340 6.77e-06 ***
+sleep_duration_fit -0.0311 0.2008 -0.155 0.878
+dominant_assignmentNon-Dominant 8.1282 22.4001 0.363 0.719
+---
+Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
+
+Residual standard error: 66.78 on 33 degrees of freedom
+Multiple R-squared: 0.005072, Adjusted R-squared: -0.05523
+F-statistic: 0.08412 on 2 and 33 DF, p-value: 0.9195
+
+
+
summary(lm(time_in_bed_act ~ time_in_bed_fit + dominant_assignment, data = df_means))
+
+
+
+Call:
+lm(formula = time_in_bed_act ~ time_in_bed_fit + dominant_assignment,
+ data = df_means)
+
+Residuals:
+ Min 1Q Median 3Q Max
+-113.165 -54.558 3.155 31.991 179.772
+
+Coefficients:
+ Estimate Std. Error t value Pr(>|t|)
+(Intercept) 571.5440 95.8090 5.965 1.07e-06 ***
+time_in_bed_fit -0.1331 0.2043 -0.651 0.519
+dominant_assignmentNon-Dominant 4.7718 23.9942 0.199 0.844
+---
+Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
+
+Residual standard error: 71.18 on 33 degrees of freedom
+Multiple R-squared: 0.0152, Adjusted R-squared: -0.04448
+F-statistic: 0.2547 on 2 and 33 DF, p-value: 0.7766
+
+
+
plot(pnew)
+
+
+
+
+
+
+
+
+
+
+
+
+
dfcor$age_group <- factor(dfcor$age_group, labels = c("Young Adults", "Older Adults"))
+
+plot1 <- dfcor %>%
+ ggplot() +
+ geom_boxplot(aes(x = age_group, y = sleep_time, fill = age_group, group = age_group)) +
+ ylab('Sleep Time Correlation (r)') +
+ scale_fill_brewer(palette = 'Set1') + xlab("Age Group")
+
+plot2 <- dfcor %>%
+ ggplot() +
+ geom_boxplot(aes(x = age_group, y = time_in_bed, fill = age_group, group = age_group)) +
+ ylab('Time In Bed Correlation (r)') +
+ scale_fill_brewer(palette = 'Set1') + xlab("Age Group")
+
+pnew <- arrangeGrob(plot1, plot2, ncol=2)
+ggsave("~/Desktop/agegroup.png", pnew, height = 5, width = 10, units = "in")
+
+
+
+
+
+
+
+
write.csv(dfcor, paste0(data_dir, 'sleep_summary_condition_correlation.csv'))
+
+
+
+
+
+
+
+
Bland Altman plots
+
About the package: blandr Interpretation: Statology
+
e.g.
+
The average of the values in the Difference column turns out to be 0.5. (middle line) The standard deviation of values in the Difference column turns out to be 1.235. The upper and lower limits of the confidence interval for the average difference can be calculated as:
+
Upper Limit: x + 1.96s = 0.5 + 1.961.235 = 2.92 Lower Limit: x – 1.96s = 0.5 – 1.961.235 = -1.92
+
On average, instrument A weighs frogs to be 0.5 grams heavier than instrument B. 95% of the differences in weight between the two instruments are expected to fall in the range of -1.92 grams and 2.92 grams.
+
First look at which device may be under or over estimating - Fitbit underestimating sleep time and time in bed
+
+
+
+
+# fitbit underestimating sleep time
+df %>%
+ ggplot(aes(x = sleep_duration_act, y = sleep_duration_fit)) +
+ geom_point() +
+ stat_smooth(method = "lm")
+
+
+
+
+
+
+df %>%
+ ggplot(aes(x = sleep_duration_cl, y = sleep_duration_fit)) +
+ geom_point() +
+ stat_smooth(method = "lm")
+
+
+
+
+
+
+#fitbit underestimating time in bed
+df %>%
+ ggplot(aes(x = time_in_bed_act, y = time_in_bed_fit)) +
+ geom_point() +
+ stat_smooth(method = "lm")
+
+
+
+
+
+
+df %>%
+ ggplot(aes(x = time_in_bed_cl, y = time_in_bed_fit)) +
+ geom_point() +
+ stat_smooth(method = "lm")
+
+
+
+
+
+
+
+
+
+
+
library(blandr)
+
+df.stats <- blandr.statistics(df$sleep_duration_act , df$sleep_duration_fit)
+p1 <- blandr.draw(df$sleep_duration_act , df$sleep_duration_fit ) + ggtitle(paste("Raw Actiwatch, \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+df.stats <- blandr.statistics(df$sleep_duration_cl , df$sleep_duration_fit)
+p2 <- blandr.draw(df$sleep_duration_cl , df$sleep_duration_fit) + ggtitle(paste("Cleaned Actiwatch, \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+df.stats <- blandr.statistics(df$sleep_duration_act[(df$actiwatch_arm == 1) & (df$handedness == 2)] , df$sleep_duration_fit[(df$actiwatch_arm == 1) & (df$handedness == 2)])
+p3 <- blandr.draw(df$sleep_duration_act[(df$actiwatch_arm == 1) & (df$handedness == 2)] , df$sleep_duration_fit[(df$actiwatch_arm == 1) & (df$handedness == 2)]) + ggtitle(paste("Raw, RH, Non-Dominant, \n \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+df.stats <- blandr.statistics(df$sleep_duration_act[(df$actiwatch_arm == 2) & (df$handedness == 2)] , df$sleep_duration_fit[(df$actiwatch_arm == 2) & (df$handedness == 2)])
+p4 <- blandr.draw(df$sleep_duration_act[(df$actiwatch_arm == 2) & (df$handedness == 2)] , df$sleep_duration_fit[(df$actiwatch_arm == 2) & (df$handedness == 2)]) + ggtitle(paste("Raw, RH, Dominant, \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+df.stats <- blandr.statistics(df$sleep_duration_cl[(df$actiwatch_arm == 1) & (df$handedness == 2)] , df$sleep_duration_fit[(df$actiwatch_arm == 1) & (df$handedness == 2)])
+p5 <- blandr.draw(df$sleep_duration_cl[(df$actiwatch_arm == 1) & (df$handedness == 2)] , df$sleep_duration_fit[(df$actiwatch_arm == 1) & (df$handedness == 2)]) + ggtitle(paste("Cleaned, RH, Non-Dominant, \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+df.stats <- blandr.statistics(df$sleep_duration_cl[(df$actiwatch_arm == 2) & (df$handedness == 2)] , df$sleep_duration_fit[(df$actiwatch_arm == 2) & (df$handedness == 2)])
+p6 <- blandr.draw(df$sleep_duration_cl[(df$actiwatch_arm == 2) & (df$handedness == 2)] , df$sleep_duration_fit[(df$actiwatch_arm == 2) & (df$handedness == 2)]) + ggtitle(paste("Cleaned, RH, Dominant, \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+gridExtra::grid.arrange(p1, p2, p3, p4, p5, p6, nrow=3, ncol=2)
+
+
+
+
+
+
+pall <- gridExtra::arrangeGrob(p1, p2, p3, p4, p5, p6, nrow=3, ncol=2)
+
+ggsave("~/Box/CogNeuroLab/Wearables/results/sleep_duration_bland_altman.png", pall, scale = 1.5, dpi = 300)
+
+
+
+
+
+
+
library(blandr)
+
+df.stats <- blandr.statistics(df$time_in_bed_act , df$time_in_bed_fit)
+p1 <- blandr.draw(df$time_in_bed_act , df$time_in_bed_fit ) + ggtitle(paste("Raw Actiwatch, \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+df.stats <- blandr.statistics(df$time_in_bed_cl , df$time_in_bed_fit)
+p2 <- blandr.draw(df$time_in_bed_cl , df$time_in_bed_fit) + ggtitle(paste("Cleaned Actiwatch, \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+df.stats <- blandr.statistics(df$time_in_bed_act[(df$actiwatch_arm == 1) & (df$handedness == 2)] , df$time_in_bed_fit[(df$actiwatch_arm == 1) & (df$handedness == 2)])
+p3 <- blandr.draw(df$time_in_bed_act[(df$actiwatch_arm == 1) & (df$handedness == 2)] , df$time_in_bed_fit[(df$actiwatch_arm == 1) & (df$handedness == 2)]) + ggtitle(paste("Raw, RH, Non-Dominant, \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+df.stats <- blandr.statistics(df$time_in_bed_act[(df$actiwatch_arm == 2) & (df$handedness == 2)] , df$time_in_bed_fit[(df$actiwatch_arm == 2) & (df$handedness == 2)])
+p4 <- blandr.draw(df$time_in_bed_act[(df$actiwatch_arm == 2) & (df$handedness == 2)] , df$time_in_bed_fit[(df$actiwatch_arm == 2) & (df$handedness == 2)]) + ggtitle(paste("Raw, RH, Dominant, \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+df.stats <- blandr.statistics(df$time_in_bed_cl[(df$actiwatch_arm == 1) & (df$handedness == 2)] , df$time_in_bed_fit[(df$actiwatch_arm == 1) & (df$handedness == 2)])
+p5 <- blandr.draw(df$time_in_bed_cl[(df$actiwatch_arm == 1) & (df$handedness == 2)] , df$time_in_bed_fit[(df$actiwatch_arm == 1) & (df$handedness == 2)]) + ggtitle(paste("Cleaned, RH, Non-Dominant, \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+df.stats <- blandr.statistics(df$time_in_bed_cl[(df$actiwatch_arm == 2) & (df$handedness == 2)] , df$time_in_bed_fit[(df$actiwatch_arm == 2) & (df$handedness == 2)])
+p6 <- blandr.draw(df$time_in_bed_cl[(df$actiwatch_arm == 2) & (df$handedness == 2)] , df$time_in_bed_fit[(df$actiwatch_arm == 2) & (df$handedness == 2)]) + ggtitle(paste("Cleaned, RH, Dominant, \n bias=", signif(df.stats$bias, 4), "lower=", signif(df.stats$lowerLOA, 4), "upper=", signif(df.stats$upperLOA, 4)))
+
+gridExtra::grid.arrange(p1, p2, p3, p4, p5, p6, nrow=3, ncol=2)
+
+
+
+
+
+
+pall <- gridExtra::arrangeGrob(p1, p2, p3, p4, p5, p6, nrow=3, ncol=2)
+
+ggsave("~/Box/CogNeuroLab/Wearables/results/time_in_bed_bland_altman.png", pall, scale = 1.5, dpi = 300)
+
+
+
+
+
+
Correspondance with Sleep Diary Data
+
+
+
+
+
+
for (participant in unique(dfn1$subject)[1]) {
+ print(participant)
+ n_pages <- dfn1 %>%
+ filter(subject == participant) %>%
+ drop_na(date) %>%
+ summarise(ceiling(length(date) / 4))
+ print(n_pages)
+
+ rawact <-
+ read_csv(act_files[grep(participant, act_files)], col_names = c("date", "activity"))
+
+ rawact <- rawact %>%
+ mutate(clock_time = as.numeric(
+ lubridate::hour(lubridate::ymd_hms(date)) + lubridate::minute(lubridate::ymd_hms(date)) /
+ 60
+ )) %>%
+ mutate(date = format(ymd_hms(date), "%Y-%m-%d"))
+
+
+ p <- dfn1 %>%
+ filter(subject == participant) %>%
+ drop_na(date) %>%
+ ggplot() +
+ geom_vline(aes(
+ xintercept = bed_time_clock_act,
+ color = 'Actiwatch',
+ linetype = 'dashed'
+ )) +
+ geom_vline(aes(
+ xintercept = bed_time_clock_fit,
+ color = 'Fitbit',
+ linetype = 'dotted'
+ )) +
+ geom_vline(aes(
+ xintercept = bed_time_clock,
+ color = 'Self-Report',
+ linetype = 'longdash'
+ )) + #, y = 1, color = 'Self-Report', shape = 'Bed Time')) +
+ geom_vline(aes(
+ xintercept = wake_time_clock_act,
+ color = 'Actiwatch',
+ linetype = 'dashed'
+ )) +
+ geom_vline(aes(
+ xintercept = wake_time_clock_fit,
+ color = 'Fitbit',
+ linetype = 'dotted'
+ )) +
+ geom_vline(aes(
+ xintercept = wake_time_clock,
+ color = 'Self-Report',
+ linetype = 'longdash'
+ )) +
+ geom_line(data = rawact,
+ aes(x = clock_time, y = activity, size = 0.2)) +
+ theme_minimal() + theme_readable(base_size = 12) +
+ labs(color = "Method") + ggtitle(participant) +
+ theme(axis.text.y = element_blank(), axis.title.y = element_blank()) +
+ scale_x_continuous(limits = c(0, 24), breaks = seq(0, 24, 3)) + xlab("Hour") +
+ facet_wrap_paginate(. ~ date, nrow = 2, ncol = 2, page = as.numeric(n_pages))
+
+ pdf(
+ paste0(
+ "~/Box/CogNeuroLab/Wearables/results/diary_comparison/",
+ as.character(participant),
+ ".pdf"
+ ),
+ onefile = T
+ )
+ print(p)
+ dev.off()
+
+}
+
+
+
+
[1] "10011"
+
+
+
+
+
+
+
+
+
+
+
+
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
+
+
+
+