-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathData_wrangling2.R
267 lines (225 loc) · 8.52 KB
/
Data_wrangling2.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
<<<<<<< HEAD
library(readr)
library(glmnet)
library(dplyr)
library(tidyverse)
#imputated data 2 is only the vessel data with improved feature extraction
source <- read.csv("raw_data2.csv", fileEncoding = "UTF-8")
#remove NAs present in the CT number data. This data should be discarded for analyses (Data cleaning)
valid_data <- source %>%
filter(!M1 %in% NA)
valid_data$M1 = as.numeric(valid_data$M1)
comparable_data <- valid_data
#selecting a specific organ for comparisons and performing the Si-component removal (Data preprocessing)
organ <- "aorta"
df <- comparable_data %>%
filter(X4 %in% organ) %>%
filter(X3 %in% "AGUIX") %>%
filter(!X5 %in% c(10674, 10308, 26328, 26361)) %>% # I need to remove non-paired samples s7-s10 since these may introduce bias
mutate(M1 = as.numeric(M1)*(1 - 0.0569))
#df <- df %>%
# filter(!X2 %in% 1)
### Machine learning
# Split the data into training and test sets
set.seed(123) # For reproducibility
regression <- lm(N1 ~ M1, data = df)
print(summary(regression))
mean_df <- df %>%
group_by(X2) %>%
summarize(M1 = mean(M1), N1 = mean(N1))
#means
regression <- lm(N1 ~ M1, data = mean_df)
print(summary(regression))
# Fit the linear model
#####
### Plot df$y versus df$X1 with custom axes and title
plot(df$M1, df$N1, xlab = "CT number", ylab = "[Gd] (mg/mL)", main = paste0("OLS-fit (", organ,")"))
# Add red dots representing the mean
points(mean_df$M1, mean_df$N1, col = "red", pch = 19)
# Add a regression line in blue
abline(regression, col = "blue")
# Get summary of the regression model
summary_regression <- summary(regression)
# Extract R-squared value
rsquared <- summary_regression$r.squared
rse <- sqrt(mean(residuals(regression)^2))
# Format R-squared and RSE for legend
rsq_label <- paste("R² =", round(rsquared, 3))
rse_label <- paste("RSE =", round(rse, 3))
# Add legend with abbreviated labels
legend("bottomright", legend = c(rsq_label, rse_label),
bg = "white", cex = 0.8)
#####
coefficients <- coef(regression) # Extract coefficients
intercept <- coefficients[1] # Intercept
slopes <- coefficients[-1]
regression_sd <- lm(y1 ~ X11, data = df)
summary(regression_sd)
coefficients_sd <- coef(regression_sd) # Extract coefficients
intercept_sd <- coefficients_sd[1] # Intercept
slopes_sd <- coefficients_sd[-1]
comparable_data <- comparable_data %>%
mutate(
y1 = case_when(
X3 == "GBCA" & X4 == organ & is.na(y1) ~ intercept_sd + X11 * slopes_sd[1],
TRUE ~ y1), # Keep original y1 values unchanged otherwise
y = case_when(
X3 == "GBCA" & X4 == organ & is.na(y) ~ intercept + X1 * slopes[1],
TRUE ~ y # Keep original N1 values unchanged otherwise
),
N1 = case_when(
X3 == "GBCA" & X4 == organ & is.na(N1) ~ intercept + M1 * slopes[1],
TRUE ~ N1 # Keep original N1 values unchanged otherwise
),
N2 = case_when(
X3 == "GBCA" & X4 == organ & is.na(N2) ~ intercept + M2 * slopes[1],
TRUE ~ N2 # Keep original N2 values unchanged otherwise
),
N3 = case_when(
X3 == "GBCA" & X4 == organ & is.na(N3) ~ intercept + M3 * slopes[1],
TRUE ~ N3 # Keep original N3 values unchanged otherwise
),
N4 = case_when(
X3 == "GBCA" & X4 == organ & is.na(N4) ~ intercept + M4 * slopes[1],
TRUE ~ N4 # Keep original N4 values unchanged otherwise
),
N5 = case_when(
X3 == "GBCA" & X4 == organ & is.na(N5) ~ intercept + M5 * slopes[1],
TRUE ~ N5 # Keep original N5 values unchanged otherwise
)
)
# Calculate mean and standard deviation for M1 and N1, N2, N3, N4, N5
mean_data <- comparable_data %>%
group_by(X2, X3, X4) %>%
summarize(
X1 = mean(X1, na.rm = TRUE),
X11 = mean(X11, na.rm = TRUE),
y = mean(y, na.rm = TRUE),
y1 = mean(y1, na.rm = TRUE),
M1 = mean(M1, na.rm = TRUE),
M2 = mean(M2, na.rm = TRUE),
M3 = mean(M3, na.rm = TRUE),
M4 = mean(M4, na.rm = TRUE),
M5 = mean(M5, na.rm = TRUE),
N1 = mean(N1, na.rm = TRUE),
N2 = mean(N2, na.rm = TRUE),
N3 = mean(N3, na.rm = TRUE),
N4 = mean(N4, na.rm = TRUE),
N5 = mean(N5, na.rm = TRUE)
)
# Order the data by X4
mean_data <- mean_data[order(mean_data$X4), ]
print(mean_data)
=======
library(readr)
library(glmnet)
library(dplyr)
library(tidyverse)
#imputated data 2 is only the vessel data with improved feature extraction
source <- read.csv("raw_data2.csv", fileEncoding = "UTF-8")
#remove NAs present in the CT number data. This data should be discarded for analyses (Data cleaning)
valid_data <- source %>%
filter(!M1 %in% NA)
valid_data$M1 = as.numeric(valid_data$M1)
comparable_data <- valid_data
#selecting a specific organ for comparisons and performing the Si-component removal (Data preprocessing)
organ <- "aorta"
df <- comparable_data %>%
filter(X4 %in% organ) %>%
filter(X3 %in% "AGUIX") %>%
filter(!X5 %in% c(10674, 10308, 26328, 26361)) %>% # I need to remove non-paired samples s7-s10 since these may introduce bias
mutate(M1 = as.numeric(M1)*(1 - 0.0569))
#df <- df %>%
# filter(!X2 %in% 1)
### Machine learning
# Split the data into training and test sets
set.seed(123) # For reproducibility
regression <- lm(N1 ~ M1, data = df)
print(summary(regression))
mean_df <- df %>%
group_by(X2) %>%
summarize(M1 = mean(M1), N1 = mean(N1))
#means
regression <- lm(N1 ~ M1, data = mean_df)
print(summary(regression))
# Fit the linear model
#####
### Plot df$y versus df$X1 with custom axes and title
plot(df$M1, df$N1, xlab = "CT number", ylab = "[Gd] (mg/mL)", main = paste0("OLS-fit (", organ,")"))
# Add red dots representing the mean
points(mean_df$M1, mean_df$N1, col = "red", pch = 19)
# Add a regression line in blue
abline(regression, col = "blue")
# Get summary of the regression model
summary_regression <- summary(regression)
# Extract R-squared value
rsquared <- summary_regression$r.squared
rse <- sqrt(mean(residuals(regression)^2))
# Format R-squared and RSE for legend
rsq_label <- paste("R² =", round(rsquared, 3))
rse_label <- paste("RSE =", round(rse, 3))
# Add legend with abbreviated labels
legend("bottomright", legend = c(rsq_label, rse_label),
bg = "white", cex = 0.8)
#####
coefficients <- coef(regression) # Extract coefficients
intercept <- coefficients[1] # Intercept
slopes <- coefficients[-1]
regression_sd <- lm(y1 ~ X11, data = df)
summary(regression_sd)
coefficients_sd <- coef(regression_sd) # Extract coefficients
intercept_sd <- coefficients_sd[1] # Intercept
slopes_sd <- coefficients_sd[-1]
comparable_data <- comparable_data %>%
mutate(
y1 = case_when(
X3 == "GBCA" & X4 == organ & is.na(y1) ~ intercept_sd + X11 * slopes_sd[1],
TRUE ~ y1), # Keep original y1 values unchanged otherwise
y = case_when(
X3 == "GBCA" & X4 == organ & is.na(y) ~ intercept + X1 * slopes[1],
TRUE ~ y # Keep original N1 values unchanged otherwise
),
N1 = case_when(
X3 == "GBCA" & X4 == organ & is.na(N1) ~ intercept + M1 * slopes[1],
TRUE ~ N1 # Keep original N1 values unchanged otherwise
),
N2 = case_when(
X3 == "GBCA" & X4 == organ & is.na(N2) ~ intercept + M2 * slopes[1],
TRUE ~ N2 # Keep original N2 values unchanged otherwise
),
N3 = case_when(
X3 == "GBCA" & X4 == organ & is.na(N3) ~ intercept + M3 * slopes[1],
TRUE ~ N3 # Keep original N3 values unchanged otherwise
),
N4 = case_when(
X3 == "GBCA" & X4 == organ & is.na(N4) ~ intercept + M4 * slopes[1],
TRUE ~ N4 # Keep original N4 values unchanged otherwise
),
N5 = case_when(
X3 == "GBCA" & X4 == organ & is.na(N5) ~ intercept + M5 * slopes[1],
TRUE ~ N5 # Keep original N5 values unchanged otherwise
)
)
# Calculate mean and standard deviation for M1 and N1, N2, N3, N4, N5
mean_data <- comparable_data %>%
group_by(X2, X3, X4) %>%
summarize(
X1 = mean(X1, na.rm = TRUE),
X11 = mean(X11, na.rm = TRUE),
y = mean(y, na.rm = TRUE),
y1 = mean(y1, na.rm = TRUE),
M1 = mean(M1, na.rm = TRUE),
M2 = mean(M2, na.rm = TRUE),
M3 = mean(M3, na.rm = TRUE),
M4 = mean(M4, na.rm = TRUE),
M5 = mean(M5, na.rm = TRUE),
N1 = mean(N1, na.rm = TRUE),
N2 = mean(N2, na.rm = TRUE),
N3 = mean(N3, na.rm = TRUE),
N4 = mean(N4, na.rm = TRUE),
N5 = mean(N5, na.rm = TRUE)
)
# Order the data by X4
mean_data <- mean_data[order(mean_data$X4), ]
print(mean_data)
>>>>>>> 77a3982bf0902d3f28a8584daad61788e60cc341