R Notebook
# Load libraries
library(aplore3)
library(caret) ## Loading required package: ggplot2## Loading required package: latticelibrary(pROC) ## Type 'citation("pROC")' for a citation.##
## Attaching package: 'pROC'## The following objects are masked from 'package:stats':
##
## cov, smooth, varlibrary(randomForest)## randomForest 4.7-1.1## Type rfNews() to see new features/changes/bug fixes.##
## Attaching package: 'randomForest'## The following object is masked from 'package:ggplot2':
##
## marginlibrary(class)
library(rpart)
library(xgboost)
library(ROSE) ## Loaded ROSE 0.0-4library(smotefamily)
library(memoise)
library(doParallel)## Loading required package: foreach## Loading required package: iterators## Loading required package: parallel# Load data
data("glow_bonemed")
glow_bonemed$fracture <- as.factor(glow_bonemed$fracture)
# Remove less predictive identifiers
trimmed_data <- glow_bonemed[, !(names(glow_bonemed) %in% c("sub_id", "site_id", "phy_id"))]
# Handling class imbalance with ROSE
set.seed(123)
rose_data <- ROSE(fracture ~ ., data = trimmed_data, seed = 1)$data
# Splitting data into training, validation, and test sets
set.seed(123)
# Split into temporary training and a test set
temp_train_indices <- createDataPartition(rose_data$fracture, p = 0.8, list = FALSE)
train_temp <- rose_data[temp_train_indices, ]
test_set <- rose_data[-temp_train_indices, ]
# Further split the temporary training set into actual training and validation sets
index <- createDataPartition(train_temp$fracture, p = 0.8, list = FALSE)
train_set <- train_temp[index, ]
validation_set <- train_temp[-index, ]
# Set up parallel processing
cl <- makeCluster(detectCores() - 1) # Use one less than the total number of cores
registerDoParallel(cl)
# Feature selection using recursive feature elimination
control <- rfeControl(functions = rfFuncs, method = "cv", number = 10, returnResamp = "all", saveDetails = TRUE)
results <- rfe(train_set[, -ncol(train_set)], train_set$fracture, sizes = c(1:5), rfeControl = control)
# Set control for training models
fit_control <- trainControl(method = "cv", number = 10, savePredictions = "final", classProbs = TRUE, verboseIter = TRUE, allowParallel = TRUE)
list(train_set = train_set, validation_set = validation_set, test_set=test_set, results = results, control=control)## $train_set
## priorfrac age weight height bmi premeno momfrac armassist
## 1 No 76.57131 63.88782 157.0700 23.17194 No No No
## 4 No 73.27055 51.91283 155.7938 24.26608 No No No
## 5 No 66.69603 52.28338 167.9820 24.54984 No Yes Yes
## 6 Yes 87.65635 52.99006 163.4894 16.05605 No No No
## 8 No 68.04976 86.31009 169.1141 24.32254 Yes No Yes
## 9 Yes 80.27934 74.98722 155.9962 25.25380 No No No
## 10 No 58.33089 76.35809 156.9728 28.18456 No No No
## 11 Yes 63.54905 72.69218 158.6275 33.15255 No No No
## 13 No 60.14699 58.58503 167.0455 22.99401 No No No
## 16 No 57.84117 50.66219 156.1355 19.99090 No Yes No
## 17 No 52.86429 70.37763 166.7190 27.88284 No No No
## 20 Yes 74.51750 70.22003 166.2621 28.92920 No No Yes
## 22 No 59.63402 92.06890 169.2035 29.97890 Yes Yes Yes
## 23 No 60.95411 88.46920 155.0449 31.42366 No No No
## 25 No 65.35568 43.73006 162.8557 22.37878 No No No
## 26 No 63.80493 72.70912 161.7124 31.06083 No No No
## 29 No 72.40462 87.42665 165.9957 20.91771 No No No
## 30 No 58.02408 54.57663 153.6212 22.34889 No No No
## 32 No 61.16975 49.78011 155.0126 20.56597 No No No
## 33 No 76.43237 89.18340 157.6523 30.02923 No Yes Yes
## 34 No 65.55628 72.76036 197.6156 15.12155 No No No
## 35 No 49.70777 70.30708 161.6883 37.27509 No No No
## 37 No 50.18331 80.75951 161.7041 25.72136 No No No
## 38 No 61.70856 73.35858 150.6925 28.17296 No No No
## 39 No 65.69610 48.32013 163.8642 20.84948 No No No
## 40 No 67.09018 37.43641 155.3514 25.44229 No No No
## 41 No 63.06337 61.76295 161.7022 25.54845 Yes No No
## 42 Yes 89.71297 54.92796 148.2578 27.42241 No No No
## 43 No 68.38781 72.27746 166.3598 25.10359 Yes No Yes
## 45 No 57.58984 73.86638 161.9960 28.95851 No No No
## 46 No 55.78411 90.47149 159.5099 38.03624 Yes Yes Yes
## 48 No 67.43021 97.81432 164.1404 32.72300 No No Yes
## 49 No 63.88991 79.74476 159.4074 32.01824 No No No
## 50 No 71.61804 72.29098 161.8455 24.89789 No No No
## 51 No 48.06555 73.23659 199.5345 16.06424 No No No
## 53 No 76.55119 61.64445 154.4373 25.95900 No No No
## 55 Yes 64.82297 69.88371 162.0614 22.12841 Yes No No
## 57 No 68.55877 79.70077 160.0377 29.33269 No No No
## 58 Yes 81.34873 65.39016 156.1578 26.41587 No No No
## 59 No 60.23245 81.13798 158.1197 31.39832 No No No
## 60 No 63.29111 66.72906 179.6172 24.65570 No No No
## 61 No 68.42247 54.39467 172.3637 19.21300 No No No
## 63 No 73.67577 60.43332 155.3129 23.84775 No No No
## 64 No 74.05605 83.64089 156.8395 33.71213 Yes No No
## 66 No 71.93684 62.82082 155.3228 30.45029 No No No
## 67 No 56.69458 93.55792 152.4289 43.40528 No No Yes
## 68 No 61.40555 68.65670 165.5752 24.00874 No No No
## 69 No 65.37224 59.39231 162.2945 28.90244 No No No
## 70 Yes 76.08935 42.28548 149.5545 23.41127 No No No
## 72 No 57.20965 69.53457 161.3670 19.66897 No No No
## 74 No 69.42618 70.05748 167.6061 20.65429 No No No
## 75 No 77.95200 61.45815 161.8999 24.47534 No No No
## 76 No 75.95498 70.16400 168.1501 24.33465 No No No
## 77 No 69.92030 65.30667 151.8914 25.16314 No No No
## 78 No 60.44902 59.85023 160.5904 27.57348 Yes No No
## 79 No 72.54137 71.62773 165.3125 28.55999 Yes No No
## 80 No 73.45231 85.34661 164.4443 36.13777 No No Yes
## 81 No 65.91401 86.96654 150.4487 34.04721 No No Yes
## 82 No 69.34865 81.49230 156.2082 31.98439 No Yes Yes
## 85 No 68.64869 93.73309 164.4083 43.25205 Yes No Yes
## 86 Yes 88.75806 54.72523 151.6941 29.81687 No No Yes
## 87 No 78.81139 38.27674 148.8836 23.04740 No No No
## 89 No 52.91022 71.04832 158.4762 26.98634 No No No
## 90 No 69.08347 65.71074 166.5208 29.84021 No No Yes
## 91 No 65.35362 77.38872 161.9371 24.70073 No No Yes
## 93 Yes 89.02528 50.75574 162.0195 19.74167 No No Yes
## 97 Yes 69.72050 78.37471 166.4963 28.38413 No No No
## 98 No 54.56123 65.13978 156.9619 28.13794 No No No
## 100 No 61.21514 62.44052 155.2634 25.55189 Yes No No
## 102 Yes 53.26732 83.23469 171.2607 30.93954 No No No
## 104 No 48.18090 82.69982 163.2234 32.72950 No No No
## 105 No 66.63236 52.49367 169.1644 21.60059 No No No
## 106 No 75.92141 83.79043 167.6849 27.91311 No No No
## 108 Yes 62.52021 131.35988 160.7475 44.57432 No Yes Yes
## 110 No 57.95767 63.78739 166.6214 25.29923 No No Yes
## 111 No 60.07390 57.82318 157.6502 18.08299 No No No
## 112 No 66.57753 60.98836 161.0837 31.57782 No No No
## 113 No 77.01021 48.51803 168.0461 17.66690 No No No
## 116 No 77.97838 50.39413 148.0934 17.93720 No No No
## 117 No 83.39859 68.71358 155.3717 19.76527 No No No
## 120 No 53.89866 66.68286 155.7395 30.25693 No No No
## 121 No 57.51412 66.76642 167.0738 24.99912 Yes No No
## 122 No 67.63735 89.32317 163.6974 25.95975 No No No
## 126 No 73.25314 57.30743 167.1577 23.59360 Yes No No
## 128 Yes 61.99341 76.69313 162.8776 26.86351 No No No
## 136 No 50.77891 86.45181 168.8790 30.58167 Yes Yes Yes
## 137 No 75.55462 89.99825 166.4626 29.65927 No No Yes
## 138 No 54.93355 49.85433 160.2234 19.96277 No Yes No
## 139 No 62.91543 60.96426 165.4807 19.69004 No No No
## 140 No 59.60464 99.24209 156.8374 42.76232 No No Yes
## 141 No 54.49651 63.30765 158.5724 30.05205 No No Yes
## 142 No 78.18879 60.36907 159.1585 23.96904 No No No
## 144 No 68.05407 68.22116 162.8419 29.26544 Yes No No
## 146 No 58.60195 54.40269 162.7842 24.65475 No No No
## 148 No 81.68119 65.63787 157.1828 24.67861 No No No
## 150 No 70.80581 79.16308 166.3999 18.92818 No No No
## 151 No 71.94866 53.20442 154.7816 27.73584 No No No
## 155 Yes 89.76802 67.20458 159.1340 23.60328 No Yes Yes
## 156 Yes 74.71325 72.00507 163.0891 28.30332 No No Yes
## 157 Yes 66.01006 108.00381 163.2054 33.98613 No No Yes
## 158 No 72.33118 81.53428 178.1060 24.67447 Yes No No
## 159 No 65.61970 67.01288 160.6549 23.93118 No No No
## 161 Yes 60.25266 66.35102 162.0286 32.87381 Yes No Yes
## 163 No 62.56367 112.88887 165.1064 41.80804 No No Yes
## 164 Yes 74.21691 99.40347 164.5443 37.79016 No No Yes
## 165 No 63.75080 73.66900 167.2403 19.81545 Yes No No
## 166 No 76.41093 49.52507 150.1695 26.09081 No No No
## 168 No 72.47964 68.20754 156.0775 29.92989 No Yes No
## 169 Yes 74.99600 70.84466 174.7185 32.69411 No No Yes
## 172 No 65.80582 60.05149 148.5044 25.05805 No No No
## 175 No 64.55336 97.04869 170.8184 30.03135 No Yes No
## 176 Yes 74.63085 50.89590 138.0992 28.15615 Yes Yes No
## 177 Yes 72.06779 72.27865 166.7321 27.35570 No No No
## 180 No 58.66876 85.22196 157.8631 33.26541 No No No
## 182 No 83.29858 65.45462 164.9832 23.56385 No No Yes
## 184 No 70.65003 56.49347 158.4865 28.28204 No No No
## 186 No 55.50082 64.63458 203.8041 20.79209 No No No
## 189 No 63.51063 51.02529 175.0532 19.24852 No No No
## 195 No 72.04903 57.47064 153.1823 30.72493 No Yes No
## 198 No 65.35718 54.94204 160.2074 22.85795 No No No
## 199 No 65.42057 66.22243 174.3681 31.92492 Yes Yes No
## 200 No 58.91906 90.31151 171.2119 38.42359 Yes No No
## 201 Yes 58.91353 84.45100 165.5047 32.31964 No No No
## 204 No 73.56971 56.26158 167.4063 25.24790 No No No
## 205 No 72.61347 59.08490 159.3913 25.40538 No No No
## 207 Yes 64.85298 64.44741 171.3122 24.47002 No No Yes
## 208 No 54.42656 67.48847 159.9195 23.11739 No No No
## 209 No 55.71590 115.67785 155.4795 45.59054 No No Yes
## 210 No 67.81973 81.00106 163.5243 31.16564 Yes No No
## 211 No 53.54565 118.99520 180.8187 38.07503 No No No
## 213 No 76.54151 93.89820 161.0139 44.34131 No No Yes
## 214 No 65.25090 105.90474 159.3252 41.25197 No No Yes
## 217 Yes 85.95233 44.66521 164.0144 18.71667 No No No
## 218 No 79.61854 62.56336 165.9123 18.76737 No Yes Yes
## 220 No 59.47491 77.53020 164.4164 29.50211 Yes Yes No
## 221 No 77.74262 72.92677 166.6596 22.41741 No Yes Yes
## 222 No 67.15308 73.08640 166.6697 23.63502 No No No
## 223 No 69.71343 103.81132 162.2581 42.75891 No No No
## 224 Yes 84.41827 79.66844 152.8082 31.20696 No No Yes
## 225 No 75.59390 60.15435 152.3581 18.95020 Yes No Yes
## 226 Yes 64.29895 104.24826 164.7552 40.33710 No No Yes
## 229 No 85.92501 51.95494 157.7248 18.71197 No No Yes
## 230 Yes 62.98702 63.67187 159.5823 25.22235 No No No
## 233 Yes 84.59392 83.62911 159.5016 25.23858 No No No
## 234 No 69.58875 59.69897 173.4377 22.00298 No No No
## 235 No 69.36775 62.44358 171.0205 20.10247 No No No
## 236 No 60.74788 67.03377 168.2304 24.06565 Yes No No
## 237 Yes 73.78915 64.74122 155.9576 24.29327 No No No
## 239 No 59.43999 47.09039 153.8264 23.76203 No No No
## 240 No 61.65398 83.00894 166.9995 25.24862 No No No
## 241 No 59.76665 77.31561 169.0312 20.72614 No No No
## 242 No 50.35766 84.15904 160.1023 29.76986 No No No
## 243 No 75.68233 104.15423 166.3423 40.28856 No No Yes
## 244 No 64.52847 53.90379 154.5395 20.40071 Yes No Yes
## 247 No 75.63243 49.13884 158.2997 27.57038 No No No
## 248 No 63.84105 106.38291 160.5449 35.90612 No Yes Yes
## 250 Yes 78.47852 62.20900 155.2304 28.79397 Yes No No
## 252 Yes 64.65015 89.39522 163.9750 41.95445 No No Yes
## 253 No 61.95383 88.51438 158.5410 35.44151 Yes No No
## 254 No 58.82361 35.00009 157.2596 22.80042 No No No
## 256 No 66.94031 89.82896 164.8864 33.70176 No Yes No
## 257 No 77.40710 72.75074 148.9472 29.61313 No No No
## 258 Yes 72.28360 70.93814 159.1046 23.61519 No No No
## 259 No 70.20572 66.09578 163.3483 26.16281 No Yes No
## 260 No 79.94988 61.97135 165.5186 18.60005 No No No
## 262 No 76.24805 59.31223 167.6599 28.79785 No No No
## 263 No 57.19995 54.06564 158.6906 24.25501 No No No
## 264 No 69.21154 43.10518 153.6337 21.50761 No No No
## 265 No 72.07484 77.11514 159.7339 28.10677 No No No
## 266 No 76.70386 50.31156 146.8502 27.22026 No No No
## 267 No 70.49334 68.87288 155.2312 24.74905 No No No
## 269 Yes 69.82384 83.89373 152.2026 33.32893 No No No
## 270 No 57.75567 59.92168 171.8263 18.55866 No No No
## 273 No 60.60236 55.08829 160.1128 26.12361 Yes No No
## 274 No 83.26318 77.34916 152.8539 30.79100 No No No
## 275 Yes 46.38760 41.24459 140.3337 24.97969 No No No
## 276 Yes 70.70922 76.11559 166.3300 24.16084 Yes No Yes
## 278 No 78.43923 36.64810 156.1258 22.29567 No Yes No
## 279 Yes 82.76426 71.03725 170.2358 34.16549 No Yes Yes
## 280 No 67.70097 73.54600 160.7048 28.65408 No No Yes
## 281 No 65.44666 89.68101 164.3759 21.14638 Yes No No
## 282 No 63.78925 53.72173 155.2027 18.94224 No No Yes
## 283 No 58.88615 57.55866 156.7307 26.61142 Yes Yes No
## 285 Yes 73.88806 53.46309 165.7779 26.10380 No No Yes
## 286 Yes 80.43553 42.68423 165.8210 22.76357 No No No
## 289 Yes 72.72439 82.04630 174.5991 21.69065 No No Yes
## 290 No 57.66083 65.88925 161.3818 24.91021 Yes No No
## 291 No 81.32728 68.86735 162.9779 28.89571 No No Yes
## 292 Yes 69.41974 38.61288 156.9267 12.45061 No No Yes
## 293 No 56.35275 70.12563 158.3167 23.58519 No No No
## 294 No 66.43960 81.58341 157.4434 28.08125 No No Yes
## 295 Yes 84.15183 52.74020 169.9008 16.18418 No No Yes
## 296 No 73.18179 77.97488 157.3349 29.77726 Yes No Yes
## 297 Yes 74.32053 82.76845 155.6278 34.97146 Yes No Yes
## 300 No 66.03087 79.61865 159.8150 30.37141 No No No
## 301 No 74.44096 66.16692 156.4194 19.82553 Yes No Yes
## 303 Yes 67.03523 78.07136 169.6737 31.34369 No Yes Yes
## 305 No 70.18265 95.02379 157.0005 37.04174 No No No
## 306 Yes 74.85203 44.21353 157.4777 28.12140 No No No
## 309 No 85.74815 48.02739 152.4419 27.53492 No No No
## 310 No 67.43684 84.96143 168.0770 21.62557 No Yes No
## 311 No 45.93424 73.84658 158.7269 29.21502 No No No
## 313 Yes 63.73754 61.97923 146.5170 26.43730 No No No
## 314 No 83.78102 74.62683 148.8877 39.47253 Yes No Yes
## 317 No 83.90723 78.19699 156.5916 25.64706 No Yes Yes
## 318 No 57.77232 81.45207 165.7451 29.28221 No No Yes
## 319 Yes 53.61253 108.74047 165.3843 44.18439 No No Yes
## 320 No 72.11990 80.49063 160.5797 24.34752 Yes Yes No
## 321 Yes 82.05523 45.46947 157.1072 21.29910 No No No
## 323 Yes 69.66840 50.61663 172.8842 25.04158 No No Yes
## 325 No 83.27259 60.98643 164.6032 19.88682 No No No
## 326 Yes 61.28914 67.11543 161.1054 26.91154 Yes No Yes
## 327 No 54.79197 56.65707 161.3790 21.27140 No No No
## 331 No 69.08169 68.40223 160.1028 26.32341 No No Yes
## 333 No 83.79515 64.67362 179.4338 29.54568 No Yes Yes
## 335 Yes 69.43885 58.02443 163.5135 29.21528 Yes Yes No
## 336 No 83.79528 40.78875 153.9163 22.32644 No No No
## 337 Yes 60.34269 111.06965 158.4508 46.33546 Yes No Yes
## 340 No 78.48926 65.48700 161.4876 33.78708 Yes No Yes
## 343 Yes 55.60156 93.98092 158.3243 36.86835 Yes No Yes
## 344 No 93.76116 85.04453 150.5849 35.53928 No No No
## 347 Yes 75.90286 87.27356 164.8871 28.14711 No Yes No
## 348 Yes 75.79509 70.44906 152.3840 29.39584 No Yes Yes
## 350 Yes 67.70744 101.60508 178.0232 31.93218 No No Yes
## 351 No 82.85464 59.53088 158.9277 26.68091 No No Yes
## 352 No 68.50650 95.05596 174.6840 26.03142 Yes No Yes
## 353 Yes 71.10523 52.36601 142.0781 26.04843 No No No
## 354 No 52.23012 73.92121 165.7124 28.44697 No No No
## 355 Yes 84.80850 41.52257 151.2255 23.49013 No Yes Yes
## 356 No 67.93811 56.01038 168.9477 22.46949 No No No
## 357 No 67.88753 76.96048 169.8851 26.35606 Yes No Yes
## 358 No 74.01736 44.68971 152.8482 21.17373 No No No
## 360 Yes 80.93773 56.98814 172.9701 19.49347 No No Yes
## 361 Yes 51.48472 64.84106 146.5000 24.58107 No No No
## 363 Yes 60.81080 86.97586 156.8702 41.90067 Yes No Yes
## 364 Yes 92.04128 49.57913 150.9309 28.28054 No Yes Yes
## 366 No 72.27856 66.76499 144.2238 18.79390 No No No
## 368 Yes 69.24661 82.95092 168.1683 29.18698 Yes No Yes
## 374 Yes 76.98615 66.39509 152.1185 27.92886 No No No
## 376 No 69.43302 92.43833 149.6821 34.01435 No No Yes
## 378 No 71.34491 52.54473 163.0945 15.79146 No No No
## 379 No 65.13542 61.99858 169.0656 27.40711 No No No
## 380 No 63.60819 85.51391 161.1634 35.66653 Yes No Yes
## 381 No 74.71948 89.89549 180.3986 27.15927 No Yes Yes
## 382 No 59.49796 80.10863 162.7134 30.12211 No No Yes
## 383 Yes 76.26880 85.10175 163.3067 25.23924 No Yes No
## 384 Yes 70.41872 73.79582 164.7054 31.43898 No Yes Yes
## 385 Yes 52.33628 61.52211 155.3891 34.73393 No No No
## 387 No 69.42894 73.79435 141.7311 36.93245 Yes No Yes
## 389 Yes 81.09563 71.68336 151.1852 27.32187 No Yes Yes
## 390 No 73.41555 49.07399 161.8871 21.52896 No No No
## 393 Yes 52.86023 125.93325 165.4718 44.09721 No No Yes
## 395 No 69.72005 57.05186 163.2748 21.57143 No No No
## 397 Yes 56.28671 81.19006 162.1750 28.37678 No No No
## 398 No 66.79389 70.63431 140.0545 34.36417 Yes No Yes
## 399 No 58.40492 49.72487 163.7034 30.75719 No No No
## 403 Yes 78.85778 74.17253 156.8189 31.64947 No No Yes
## 406 Yes 65.86120 113.27631 160.5898 43.06834 Yes No Yes
## 407 No 57.25641 92.66655 162.2872 33.40318 Yes No Yes
## 411 Yes 78.13499 88.05451 174.5775 30.92752 No No Yes
## 412 Yes 83.25946 71.44112 159.3535 14.43559 No Yes No
## 413 No 98.08873 86.25317 152.9809 32.17877 No No No
## 415 Yes 74.67588 94.79964 152.1410 47.52916 No No Yes
## 416 Yes 87.23929 77.01296 149.9864 25.19950 No No Yes
## 418 Yes 62.59765 69.47001 158.4488 27.38662 Yes No Yes
## 420 Yes 73.61288 55.25366 155.8657 30.63091 No No No
## 421 Yes 56.93751 71.31165 149.7523 30.02160 No No No
## 423 Yes 62.68611 93.71993 154.9837 43.33611 No No Yes
## 424 Yes 80.98066 91.89845 169.5168 27.78036 No No Yes
## 426 Yes 84.52892 83.12032 147.5401 31.21707 No No Yes
## 428 Yes 60.01296 130.87512 177.5659 37.15756 No No Yes
## 429 No 67.93614 92.35300 160.2896 39.48748 No No Yes
## 430 No 80.42645 66.15513 162.0124 22.16638 No No No
## 431 No 59.36833 47.12627 152.7150 21.55162 No No No
## 432 Yes 84.71175 39.19898 171.5876 15.76485 No No No
## 433 Yes 54.53825 70.16895 160.5709 26.68173 No No No
## 434 No 57.88193 80.52171 168.1039 30.10928 Yes No No
## 435 No 90.06783 83.93441 153.6768 24.64934 No Yes Yes
## 437 No 83.54955 74.90091 134.1323 42.97548 No No Yes
## 438 Yes 54.31973 95.13594 162.4912 43.35716 No No Yes
## 439 Yes 82.06854 48.09827 153.9309 21.10450 No No Yes
## 440 No 86.88456 102.66682 168.4586 34.64790 No No Yes
## 443 Yes 58.07873 116.83126 156.4798 45.36278 No No Yes
## 444 Yes 84.48926 37.64577 155.8863 16.23828 No No No
## 445 Yes 67.10743 124.83317 175.7802 43.61234 No No Yes
## 446 Yes 65.89133 56.39671 167.0893 21.18577 No No No
## 447 Yes 68.26818 97.12359 157.2665 45.24504 Yes No Yes
## 448 Yes 74.90943 60.76673 163.4109 21.10342 No No Yes
## 449 No 55.84162 54.28435 162.3112 17.80550 No No No
## 451 Yes 73.69203 63.52946 169.0309 25.54758 No No Yes
## 452 No 73.04945 84.56930 167.2016 23.76979 Yes No Yes
## 453 Yes 75.68019 80.95962 163.7927 25.31748 No Yes No
## 454 No 84.36394 50.20492 147.1212 22.96047 No No Yes
## 455 No 72.90050 88.17149 153.4256 33.31107 No No Yes
## 456 No 75.11071 77.44199 163.0177 21.75546 Yes No No
## 457 No 83.95010 62.04546 152.0883 14.62292 No No No
## 458 No 53.54874 89.15939 165.9391 27.92747 No No Yes
## 462 No 58.10834 77.06126 160.4586 26.01875 No No No
## 463 No 81.52549 52.69483 161.9888 24.76120 No No No
## 465 Yes 86.76560 49.16408 164.2629 25.17198 No No Yes
## 468 No 56.98651 54.92472 154.1229 25.24119 No No No
## 469 Yes 76.88964 59.71273 165.8057 28.15299 No Yes No
## 471 No 71.49142 66.99815 168.8577 22.82492 Yes No No
## 474 No 47.61039 101.22313 159.7559 38.50385 Yes Yes No
## 476 No 75.92687 54.37467 154.5943 22.05478 No No Yes
## 477 No 66.26418 71.70206 172.1890 19.12922 Yes No No
## 478 No 69.77042 65.94128 156.1697 26.20175 No No No
## 479 No 87.98983 98.27926 170.6731 34.34495 No No No
## 480 Yes 75.14289 129.03760 174.0125 39.81371 No No Yes
## 481 Yes 89.92041 57.68064 156.3366 23.69545 No No Yes
## 482 Yes 63.72045 68.55941 159.8879 30.57240 No No Yes
## 484 No 79.82220 87.31804 165.4748 34.98575 No No Yes
## 486 No 50.18171 54.62860 153.4871 17.87525 No No No
## 488 Yes 80.37850 83.17213 153.5205 31.07070 No No Yes
## 490 Yes 74.80035 90.11384 172.2727 26.42260 No No Yes
## 491 Yes 57.77853 60.26583 152.6209 28.33146 No No No
## 493 Yes 86.48388 78.10339 150.2436 33.98332 No No Yes
## 494 Yes 86.42408 59.42103 151.9400 23.80311 No No No
## 495 No 71.57434 66.60855 156.2115 18.39726 No No No
## 496 No 79.26159 79.16509 149.9979 36.64621 No No No
## 497 No 76.43735 61.25051 157.5315 24.29828 No No Yes
## smoke raterisk fracscore fracture bonemed bonemed_fu bonetreat
## 1 Yes Less 3.83759409 No No No No
## 4 Yes Less 3.61050088 No No No No
## 5 No Same 4.67662112 No No No No
## 6 No Same 7.97972892 No No No No
## 8 No Same 1.91930779 No No No No
## 9 No Same 9.37991205 No Yes No No
## 10 No Less -0.29720967 No No No No
## 11 No Same -0.57161644 No Yes Yes Yes
## 13 No Same 4.19968804 No No No No
## 16 No Greater 1.72022768 No No No No
## 17 No Same 1.10917965 No No No No
## 20 No Greater 7.86864237 No Yes Yes Yes
## 22 No Greater 2.14947103 No No No No
## 23 No Same 1.55837104 No No No No
## 25 No Less 5.58875758 No No No No
## 26 No Less 1.53032026 No No No No
## 29 No Same 1.74389847 No Yes Yes Yes
## 30 No Greater 0.72137763 No Yes Yes Yes
## 32 No Greater 1.67753630 No Yes Yes Yes
## 33 No Same 5.40393523 No No No No
## 34 No Greater -0.94346031 No No No No
## 35 No Greater -1.89181545 No No No No
## 37 No Greater 0.38755828 No No No No
## 38 No Same 0.75463181 No No No No
## 39 No Less 0.36011610 No No No No
## 40 No Same 2.29998503 No No No No
## 41 No Less -0.59676707 No No No No
## 42 No Same 5.28997421 No No No No
## 43 No Same 2.32195631 No No No No
## 45 No Greater 0.91809922 No No No No
## 46 Yes Same 4.82979454 No No No No
## 48 No Same 5.64646695 No No No No
## 49 No Same -1.33034671 No No No No
## 50 No Same 3.42778556 No No No No
## 51 No Greater -0.85115244 No No No No
## 53 No Less 2.96667684 No No No No
## 55 Yes Same 4.40149336 No No No No
## 57 No Less 2.90479878 No Yes Yes Yes
## 58 No Same 6.34254605 No Yes Yes Yes
## 59 No Greater 0.56522284 No No No No
## 60 No Less 2.11332631 No No No No
## 61 No Less 2.20643604 No No No No
## 63 No Same 2.54618443 No Yes Yes Yes
## 64 No Less 3.37526857 No No No No
## 66 Yes Less 4.11758493 No No No No
## 67 No Less 0.58494981 No No No No
## 68 No Greater -0.11277842 No Yes Yes Yes
## 69 No Less 1.35465941 No No No No
## 70 No Same 9.46372589 No Yes Yes Yes
## 72 No Greater 0.25139293 No No No No
## 74 No Less 2.79172364 No No No No
## 75 No Same 5.41916388 No No No No
## 76 Yes Less 4.01164162 No No No No
## 77 No Less 4.88239072 No No No No
## 78 No Same 1.61787629 No No No No
## 79 No Same -0.70121139 No No No No
## 80 No Same 6.54630627 No No No No
## 81 No Less 1.74312446 No No No No
## 82 No Same 5.41206052 No No No No
## 85 No Same 3.23666373 No No No No
## 86 No Greater 6.24604846 No Yes Yes Yes
## 87 No Less 7.62044725 No No No No
## 89 No Greater 0.96132537 No No No No
## 90 No Greater 5.98288799 No No Yes No
## 91 No Less 4.09530392 No Yes Yes Yes
## 93 No Greater 7.68602608 No Yes Yes Yes
## 97 No Less 4.41630194 No No No No
## 98 No Greater 0.86149313 No No No No
## 100 No Same 1.26402125 No No No No
## 102 No Same 0.06143779 No No No No
## 104 No Greater 0.80289120 No No No No
## 105 No Less 2.95242075 No No No No
## 106 Yes Less 6.39837376 No No No No
## 108 No Greater 5.46331638 No No No No
## 110 No Greater 4.11845619 No No No No
## 111 No Same 1.80489271 No Yes Yes Yes
## 112 No Same 3.41782867 No No No No
## 113 No Less 8.66209965 No No No No
## 116 No Greater 6.32224666 No Yes Yes Yes
## 117 No Less 6.72116350 No No No No
## 120 No Greater 0.91600782 No No No No
## 121 No Same 2.64741170 No No No No
## 122 No Same 1.96819217 No No No No
## 126 Yes Same 1.65817796 No No No No
## 128 No Same 1.01335040 No Yes Yes Yes
## 136 No Greater 4.15709998 No No No No
## 137 No Same 5.00593736 No No No No
## 138 No Greater 3.91846473 No No No No
## 139 No Greater 1.34607412 No Yes Yes Yes
## 140 No Less 2.24827519 No No No No
## 141 No Less 3.13449794 No No No No
## 142 No Same 4.72811380 No Yes Yes Yes
## 144 Yes Same 2.75933046 No No No No
## 146 No Same 1.35927350 No No No No
## 148 No Less 4.67670228 No No No No
## 150 No Less 4.01717896 No No No No
## 151 No Less 3.34076610 No Yes Yes Yes
## 155 No Greater 8.82376100 No Yes Yes Yes
## 156 No Less 6.21062877 No No No No
## 157 No Less 3.55709921 No No No No
## 158 No Same -0.78232166 No Yes No No
## 159 No Same 1.39095269 No No No No
## 161 No Same 3.17921267 No No No No
## 163 No Same 3.22457894 No No No No
## 164 No Greater 7.36120004 No No No No
## 165 No Same 0.16551703 No No No No
## 166 No Greater 2.56570551 No Yes Yes Yes
## 168 Yes Less 2.14492035 No No No No
## 169 No Less 4.05002328 No No No No
## 172 No Same 6.24836531 No Yes Yes Yes
## 175 No Same 3.24949051 No No No No
## 176 No Same 5.12047459 No No No No
## 177 No Same 4.82913003 No Yes Yes Yes
## 180 No Less 1.74976774 No No No No
## 182 No Less 9.35442251 No No No No
## 184 No Less 3.52374206 No No No No
## 186 No Greater 0.41132174 No No No No
## 189 No Less 1.89773269 No No No No
## 195 Yes Less 3.93790350 No No No No
## 198 No Greater 5.15289018 No No Yes No
## 199 No Less 2.94497644 No No No No
## 200 No Same 1.15646949 No No No No
## 201 No Same 4.01190225 No Yes Yes Yes
## 204 No Same 5.80295445 No No Yes No
## 205 No Greater 1.92560829 No Yes Yes Yes
## 207 No Greater 4.96325237 No Yes Yes Yes
## 208 No Less -0.75996179 No No No No
## 209 No Same 1.33522410 No No No No
## 210 No Greater 3.64666439 No Yes Yes Yes
## 211 No Same -0.68956506 No No No No
## 213 No Same 6.76817482 No No No No
## 214 No Less 2.53174105 No Yes No No
## 217 No Same 11.38953826 No No No No
## 218 No Less 5.74726402 No No No No
## 220 No Less 3.00970224 No No No No
## 221 No Less 8.08118412 No No No No
## 222 No Less 2.18976499 No No No No
## 223 No Less 2.71172989 No No No No
## 224 No Greater 7.70617402 No Yes Yes Yes
## 225 No Same 5.96325680 No Yes Yes Yes
## 226 No Less 5.52486190 No No No No
## 229 No Greater 7.85038748 No No No No
## 230 No Same 1.97593277 No Yes Yes Yes
## 233 No Less 4.86599201 No No No No
## 234 No Less 1.31834184 No No No No
## 235 No Greater 2.34047610 No No Yes No
## 236 No Same 0.74733630 No No No No
## 237 No Less 4.72446095 No No No No
## 239 No Less 3.38120993 No No No No
## 240 No Less -1.44870730 No No No No
## 241 No Less -0.08508828 No No No No
## 242 No Same 1.00250349 No No No No
## 243 No Less 6.24785082 No No No No
## 244 Yes Same 4.76098298 No No No No
## 247 No Same 3.49947135 No Yes Yes Yes
## 248 No Less 2.50698263 No No No No
## 250 No Less 4.77173214 No No No No
## 252 No Less 4.35302363 No No No No
## 253 No Same -0.29269025 No No No No
## 254 No Greater 0.10665709 No No No No
## 256 No Less 4.00016084 No No No No
## 257 No Less 4.35152115 No No No No
## 258 No Less 5.26076836 No No No No
## 259 No Same 6.22948927 No Yes Yes Yes
## 260 No Less 2.72741566 No No No No
## 262 No Same 4.77155449 No No Yes No
## 263 Yes Less 0.44426206 No No No No
## 264 No Less 3.37439077 No Yes Yes Yes
## 265 No Less 2.93519922 No No No No
## 266 No Less 6.06190764 No No No No
## 267 No Greater 0.64458540 No Yes Yes Yes
## 269 Yes Same 3.62753645 No No No No
## 270 No Less 0.97131656 No No No No
## 273 No Greater -0.74247039 Yes Yes Yes Yes
## 274 No Same 1.04932962 Yes Yes Yes Yes
## 275 No Less 2.00076566 Yes No Yes No
## 276 No Greater 7.09897052 Yes No No No
## 278 No Same 4.57201173 Yes Yes Yes Yes
## 279 No Same 9.07508688 Yes No No No
## 280 No Less 4.25636994 Yes No No No
## 281 No Less 2.79437514 Yes No No No
## 282 No Same 7.70009891 Yes Yes Yes Yes
## 283 No Same 2.76857740 Yes No No No
## 285 No Greater 8.43658594 Yes No No No
## 286 No Less 4.88255349 Yes Yes Yes Yes
## 289 Yes Greater 8.41256590 Yes Yes Yes Yes
## 290 No Greater 4.70468281 Yes No No No
## 291 No Less 7.65297226 Yes No No No
## 292 No Greater 8.74110735 Yes Yes Yes Yes
## 293 No Same 0.68373865 Yes No No No
## 294 No Less 1.20483122 Yes No No No
## 295 No Greater 7.44023714 Yes Yes Yes Yes
## 296 No Greater 4.75523137 Yes No Yes No
## 297 No Greater 4.50419323 Yes No No No
## 300 No Greater -0.81279302 Yes Yes Yes Yes
## 301 No Same 9.28621670 Yes No Yes No
## 303 No Same 8.61168071 Yes No No No
## 305 No Less 3.06827022 Yes No No No
## 306 No Less 7.89000782 Yes Yes Yes Yes
## 309 No Less 7.35009023 Yes No Yes No
## 310 No Greater 5.41858461 Yes No No No
## 311 No Same 0.07860770 Yes No No No
## 313 No Less -1.60743367 Yes No Yes No
## 314 No Less 4.38902205 Yes No No No
## 317 No Same 8.67012308 Yes No No No
## 318 No Greater 2.39796235 Yes No No No
## 319 No Same 5.27285794 Yes No Yes No
## 320 Yes Greater 7.30820682 Yes Yes Yes Yes
## 321 No Less 8.24887754 Yes No No No
## 323 Yes Greater 8.57396577 Yes Yes Yes Yes
## 325 No Less 2.83372252 Yes Yes No No
## 326 No Greater 6.64971075 Yes No No No
## 327 No Greater 1.61319386 Yes No No No
## 331 No Less 6.66910425 Yes No No No
## 333 No Less 6.77158807 Yes No No No
## 335 No Greater 3.54798426 Yes No No No
## 336 No Greater 4.14695512 Yes Yes Yes Yes
## 337 No Greater 6.55935310 Yes Yes Yes Yes
## 340 No Less 4.12696607 Yes Yes Yes Yes
## 343 No Greater 4.11011950 Yes No No No
## 344 No Less 3.50104796 Yes No No No
## 347 No Same 4.37117044 Yes No No No
## 348 No Greater 8.92591560 Yes No Yes No
## 350 No Greater 6.54746904 Yes No No No
## 351 No Same 8.68206001 Yes No Yes No
## 352 No Less 4.90885976 Yes No No No
## 353 No Less 2.59822439 Yes No Yes No
## 354 No Same -2.06900789 Yes No No No
## 355 No Greater 9.14902357 Yes Yes Yes Yes
## 356 No Less 2.25699386 Yes No No No
## 357 No Less 6.06177509 Yes No No No
## 358 No Less 5.93677359 Yes No Yes No
## 360 No Greater 7.18480313 Yes No No No
## 361 No Less 2.74603901 Yes No Yes No
## 363 No Greater 4.27769782 Yes Yes Yes Yes
## 364 No Greater 9.55470266 Yes Yes Yes Yes
## 366 No Greater 3.25784588 Yes Yes Yes Yes
## 368 No Greater 2.55554187 Yes No No No
## 374 No Less 7.97002598 Yes Yes Yes Yes
## 376 No Same 6.71271146 Yes Yes No No
## 378 No Less 3.71843159 Yes Yes No No
## 379 No Same 1.03408930 Yes No No No
## 380 Yes Same 3.66020718 Yes No No No
## 381 No Less 8.46664486 Yes No No No
## 382 No Greater 2.87390524 Yes No No No
## 383 No Same 5.69059976 Yes No No No
## 384 No Same 8.23919202 Yes No No No
## 385 No Greater 2.50111483 Yes No Yes No
## 387 No Less 4.21229887 Yes No No No
## 389 No Greater 10.51828391 Yes No Yes No
## 390 No Same 6.09667514 Yes Yes Yes Yes
## 393 No Same 3.62304272 Yes No Yes No
## 395 No Less 3.65280768 Yes No No No
## 397 No Greater 1.67440945 Yes No No No
## 398 No Less 5.42553906 Yes No No No
## 399 No Same 4.29952991 Yes Yes Yes Yes
## 403 No Same 8.66020579 Yes No No No
## 406 No Greater 6.57287655 Yes Yes Yes Yes
## 407 Yes Same 4.03329352 Yes No No No
## 411 No Less 5.81824491 Yes No No No
## 412 No Greater 6.61803071 Yes No Yes No
## 413 No Less 6.49168115 Yes No No No
## 415 No Less 6.45346885 Yes No No No
## 416 No Greater 10.56253259 Yes Yes Yes Yes
## 418 No Greater 3.77006630 Yes No No No
## 420 No Same 5.49290971 Yes Yes Yes Yes
## 421 No Greater 0.16956908 Yes No Yes No
## 423 No Greater 3.86185642 Yes No No No
## 424 Yes Same 6.19195701 Yes Yes Yes Yes
## 426 No Same 8.10231111 Yes No No No
## 428 No Greater 5.87738639 Yes No No No
## 429 No Same 4.73173900 Yes Yes No No
## 430 No Same 6.73148245 Yes Yes Yes Yes
## 431 No Greater 1.58219216 Yes Yes Yes Yes
## 432 No Less 8.70965457 Yes Yes No No
## 433 No Same 4.58052431 Yes No No No
## 434 No Less 1.87377794 Yes No No No
## 435 No Same 6.41442423 Yes No No No
## 437 No Same 8.53356949 Yes Yes Yes Yes
## 438 No Same 2.71069991 Yes No Yes No
## 439 No Same 10.54521592 Yes Yes No No
## 440 Yes Greater 9.60187570 Yes No No No
## 443 No Greater 4.97391836 Yes No No No
## 444 No Same 6.17806939 Yes Yes Yes Yes
## 445 No Greater 4.49621430 Yes No No No
## 446 Yes Greater 3.27451231 Yes No No No
## 447 No Greater 6.46355061 Yes Yes Yes Yes
## 448 Yes Greater 6.37369505 Yes Yes Yes Yes
## 449 No Greater -3.65928443 Yes No No No
## 451 Yes Greater 9.04530582 Yes Yes Yes Yes
## 452 No Same 3.74611483 Yes No No No
## 453 No Same 5.68820022 Yes No No No
## 454 No Same 10.21613213 Yes No No No
## 455 No Same 2.05726891 Yes Yes No No
## 456 No Same 0.53483305 Yes No No No
## 457 No Less 8.00477902 Yes No No No
## 458 No Less 4.52556961 Yes No No No
## 462 No Less 2.26869445 Yes No No No
## 463 No Same 7.45526175 Yes Yes Yes Yes
## 465 No Greater 11.56377469 Yes No No No
## 468 No Less 0.58444062 Yes No No No
## 469 No Same 5.27943661 Yes No No No
## 471 No Greater 3.45128958 Yes No No No
## 474 No Same -2.77595806 Yes No No No
## 476 No Greater 5.59218704 Yes No Yes No
## 477 No Greater 3.07443269 Yes No No No
## 478 No Same 5.09854000 Yes Yes Yes Yes
## 479 No Less 5.31110242 Yes No No No
## 480 No Greater 2.50910256 Yes No No No
## 481 No Less 9.18075236 Yes Yes No No
## 482 No Same 3.59515664 Yes No No No
## 484 No Greater 5.42323021 Yes No No No
## 486 No Greater -1.36206112 Yes Yes Yes Yes
## 488 No Greater 3.58586507 Yes Yes Yes Yes
## 490 Yes Same 7.05704271 Yes Yes Yes Yes
## 491 No Less 2.81953428 Yes No No No
## 493 No Same 7.50239320 Yes No No No
## 494 No Less 6.85702281 Yes No No No
## 495 No Same 7.25699505 Yes Yes Yes Yes
## 496 No Same 6.11402764 Yes Yes Yes Yes
## 497 No Same 8.60250851 Yes No No No
##
## $validation_set
## priorfrac age weight height bmi premeno momfrac armassist
## 7 No 63.09999 98.38587 165.2956 36.07534 No No No
## 14 No 85.43227 82.58867 168.1624 19.20851 No No Yes
## 21 No 71.43618 65.16896 153.5457 30.21873 Yes No No
## 24 No 76.32431 72.75578 156.7735 27.23678 No No No
## 36 Yes 64.07745 107.17234 158.1017 38.56302 No No Yes
## 44 No 66.98171 79.35094 152.6520 36.35555 No No No
## 52 No 69.24648 90.26460 165.4306 27.38510 Yes No No
## 54 No 85.15716 49.69735 158.7345 20.50697 No No Yes
## 83 No 69.01927 59.53211 167.2678 31.03244 No Yes No
## 84 No 56.95850 71.06885 166.7907 21.45063 No No No
## 88 Yes 77.83976 79.17878 158.5095 20.78633 No No No
## 92 No 61.02796 90.50811 168.7041 32.54605 Yes No Yes
## 94 No 57.96584 59.13227 165.5098 13.34164 No No No
## 99 No 79.01303 46.16594 157.0925 23.43303 No No Yes
## 103 Yes 82.58248 66.37657 161.2251 27.31940 No No Yes
## 107 No 55.25743 92.18956 156.3635 36.01853 Yes No Yes
## 109 No 75.73181 67.73474 161.9628 25.31233 No No No
## 118 No 84.12562 66.43677 163.7166 20.10168 No No No
## 124 No 83.90791 50.19188 159.3852 23.40062 No No Yes
## 127 No 61.10793 74.46170 160.2720 25.74997 No No No
## 135 Yes 67.89122 52.68928 167.3072 18.46376 No Yes No
## 143 Yes 85.22721 58.63806 154.9021 19.46072 No Yes Yes
## 145 Yes 86.57534 73.81038 167.7477 26.67067 No No No
## 153 No 47.67964 72.64352 157.3533 27.27762 No No No
## 160 No 64.94077 75.86530 155.0726 33.28583 Yes No No
## 162 No 70.32017 78.39185 168.9146 29.13827 No No Yes
## 170 No 63.82716 55.08459 157.9080 17.33543 No No No
## 179 No 78.36768 62.66630 172.2398 10.39859 No No No
## 185 No 70.11945 46.05325 167.4525 20.22451 No No No
## 187 Yes 68.93607 76.32112 170.9543 25.49484 No No Yes
## 188 No 66.95645 63.43399 168.0714 19.27808 No Yes No
## 190 No 72.84271 72.01211 160.0936 22.30162 No No No
## 197 No 55.51276 53.70563 151.3146 22.87657 No No No
## 203 No 67.36791 50.49489 154.5410 28.37043 No No No
## 219 No 77.58593 61.58507 162.1470 26.02533 No No Yes
## 227 No 56.56464 90.02253 160.4327 28.90213 No No No
## 228 No 57.96528 81.66041 165.9887 33.68623 No No No
## 231 No 57.73496 96.29734 162.9539 32.35524 No No No
## 232 No 60.85505 122.86002 167.1154 44.36774 No No Yes
## 238 Yes 81.85510 85.71521 174.1840 25.85344 No No Yes
## 246 No 76.66766 55.79588 155.4614 23.68517 No No No
## 249 No 76.71198 77.19890 159.2938 24.00484 No No No
## 261 No 59.46618 127.36557 156.1342 50.14583 Yes No Yes
## 272 Yes 64.91002 100.22479 159.7180 40.02782 Yes No Yes
## 277 Yes 79.13201 57.09578 143.4361 27.34636 No Yes Yes
## 284 No 68.76284 70.13670 168.3837 30.20503 Yes No Yes
## 288 No 69.22187 38.46149 157.9816 23.52145 No No No
## 298 Yes 67.09091 90.44864 157.0495 28.78035 No No Yes
## 299 Yes 65.68944 115.59478 160.0064 47.20188 No No Yes
## 302 Yes 66.43379 65.43104 168.8795 26.12470 No No No
## 312 No 74.81840 60.81040 161.2411 33.29323 Yes No Yes
## 315 No 62.45618 84.70200 156.3099 26.64476 No No No
## 329 No 63.74804 54.35501 168.1770 23.42379 No Yes No
## 330 Yes 68.88205 79.52573 155.1615 33.07985 Yes No Yes
## 334 Yes 64.53708 90.15085 156.6594 34.45341 No No Yes
## 339 No 71.82003 54.13822 153.6135 22.02583 No Yes Yes
## 345 No 82.47358 53.25000 150.4497 22.82297 No No No
## 346 No 70.16199 61.29762 161.1071 15.65384 No No No
## 362 Yes 65.44637 40.97198 169.0493 25.82940 No No No
## 372 No 60.95954 55.40641 157.9703 20.92862 Yes No No
## 375 No 79.04405 62.59274 150.7158 25.87095 Yes No Yes
## 377 Yes 76.54905 67.90121 169.5580 25.66993 No No Yes
## 388 No 50.68470 79.83331 159.0045 33.90678 No Yes No
## 391 Yes 66.14656 68.86562 160.6296 22.79970 Yes Yes No
## 392 No 88.48579 76.80634 156.9343 27.48863 No No Yes
## 396 Yes 86.15843 45.78469 160.0708 17.22567 No No Yes
## 400 No 70.88265 89.01685 168.6406 39.38448 No No Yes
## 404 Yes 53.44571 54.91413 166.9835 25.92112 No No No
## 436 No 66.21311 68.80946 164.3404 23.36812 Yes No No
## 441 No 52.18638 81.71606 158.2459 34.26279 Yes Yes No
## 450 Yes 70.76631 82.59721 163.6327 30.34119 Yes No Yes
## 459 No 68.05979 73.35071 166.0946 25.23314 No Yes No
## 460 Yes 70.80824 81.75229 169.4380 27.34732 No Yes No
## 461 Yes 91.49829 64.19036 161.8275 22.92849 No No Yes
## 473 No 79.75396 58.04695 164.9298 23.97461 No Yes No
## 483 No 91.25935 77.50177 153.1730 32.79211 No No Yes
## 487 Yes 62.06655 94.65392 162.2634 25.67487 Yes No Yes
## 489 Yes 86.45346 75.11374 152.6609 35.58282 No No Yes
## 500 Yes 90.76761 58.12256 166.2035 19.84490 No No Yes
## smoke raterisk fracscore fracture bonemed bonemed_fu bonetreat
## 7 No Less 0.72021155 No No No No
## 14 No Same 6.77255566 No No No No
## 21 No Greater 4.15990092 No No No No
## 24 No Less 2.85114189 No No No No
## 36 No Less 5.66056064 No No No No
## 44 No Same 1.58050533 No No No No
## 52 No Same 2.18488555 No No No No
## 54 No Same 9.37571592 No No No No
## 83 No Same 3.11137221 No Yes Yes Yes
## 84 No Greater -0.13952477 No No No No
## 88 No Less 5.17744602 No No No No
## 92 No Same 2.17921424 No No No No
## 94 No Less 2.20596076 No No No No
## 99 No Greater 7.87051620 No No No No
## 103 No Same 8.73086412 No No No No
## 107 Yes Same 4.50771537 No No No No
## 109 No Less 0.57674499 No No No No
## 118 No Same 6.13306089 No Yes Yes Yes
## 124 No Same 8.41218933 No No No No
## 127 No Same 2.29392019 No No No No
## 135 No Less 2.16322153 No No No No
## 143 No Less 11.98538608 No No No No
## 145 No Less 4.15317732 No No No No
## 153 No Less -0.18253027 No No No No
## 160 No Greater 1.97848772 No No No No
## 162 No Less 6.54322606 No No No No
## 170 No Greater 4.58148318 No Yes Yes Yes
## 179 No Greater 2.82963779 No No Yes No
## 185 No Less 1.39157098 No No No No
## 187 No Greater 5.91677822 No Yes Yes Yes
## 188 No Same 2.84540516 No Yes Yes Yes
## 190 No Less 5.06565155 No No No No
## 197 No Less 2.24140765 No Yes Yes Yes
## 203 No Less 2.18081500 No No No No
## 219 No Less 8.36835504 No No No No
## 227 No Same -0.43508124 No No No No
## 228 No Same 5.29776008 No No No No
## 231 No Less -2.57725457 No No No No
## 232 Yes Same 3.45867194 No No No No
## 238 No Greater 6.65331884 No Yes Yes Yes
## 246 No Same 4.78841861 No Yes Yes Yes
## 249 No Same 2.42806991 No Yes Yes Yes
## 261 No Same 3.16549020 No No No No
## 272 No Greater 3.25594271 Yes Yes Yes Yes
## 277 No Greater 9.52826858 Yes Yes Yes Yes
## 284 No Same 5.41544706 Yes No No No
## 288 No Greater 6.68304988 Yes Yes Yes Yes
## 298 No Same 2.86716316 Yes No No No
## 299 No Greater 4.93428053 Yes Yes Yes Yes
## 302 No Same 3.14896781 Yes No No No
## 312 No Less 3.28269885 Yes Yes Yes Yes
## 315 No Same -2.06281465 Yes No No No
## 329 No Greater 3.31234430 Yes No No No
## 330 No Greater 6.78610478 Yes No No No
## 334 No Same 4.51333122 Yes No No No
## 339 No Greater 9.94657581 Yes No Yes No
## 345 No Less 7.67639213 Yes No No No
## 346 No Same 0.01049171 Yes Yes Yes Yes
## 362 Yes Greater 1.40994442 Yes No No No
## 372 No Greater 0.08408128 Yes Yes Yes Yes
## 375 No Same 4.98354938 Yes Yes Yes Yes
## 377 No Greater 7.35158083 Yes No No No
## 388 No Same 1.40421421 Yes No No No
## 391 No Greater 3.66365933 Yes No No No
## 392 No Greater 8.25951877 Yes No Yes No
## 396 No Greater 6.52296202 Yes Yes Yes Yes
## 400 No Less 3.22187611 Yes No No No
## 404 Yes Greater 2.04794203 Yes No No No
## 436 No Greater 1.32943644 Yes Yes Yes Yes
## 441 No Same 1.09503063 Yes No No No
## 450 No Same 6.74294963 Yes No No No
## 459 No Same 4.76236827 Yes No No No
## 460 No Same 4.90761429 Yes No No No
## 461 No Less 6.82133386 Yes Yes No No
## 473 No Less 4.00136712 Yes No No No
## 483 No Greater 4.65050510 Yes No No No
## 487 No Greater 6.83305180 Yes No No No
## 489 No Same 8.50350049 Yes No No No
## 500 No Greater 7.06562397 Yes No No No
##
## $test_set
## priorfrac age weight height bmi premeno momfrac armassist
## 2 No 62.72182 97.68371 158.1482 30.87807 No No Yes
## 3 No 65.04079 96.27363 170.7513 28.76940 No No Yes
## 12 No 64.85816 63.45075 151.7849 22.95598 Yes No No
## 15 No 62.88163 63.32804 157.7212 28.77763 Yes No No
## 18 No 54.90468 118.92517 158.6329 38.91459 No No No
## 19 Yes 71.96320 74.01754 169.4288 26.70969 No No Yes
## 27 No 50.79888 101.99564 168.2195 28.36861 No No Yes
## 28 No 63.58864 89.27746 158.0984 40.02523 No No No
## 31 Yes 78.58513 99.70824 165.8298 37.46544 Yes No Yes
## 47 No 56.76064 81.26793 170.5846 31.67434 No No No
## 56 No 52.77546 64.83244 162.9659 22.21471 Yes No No
## 62 No 69.19656 65.91146 160.5488 24.20225 Yes No No
## 65 Yes 70.45857 59.30130 161.4198 26.21285 Yes Yes No
## 71 No 94.30676 76.96663 164.2928 28.86742 No No Yes
## 73 Yes 75.73821 67.05063 167.3472 23.77511 No No Yes
## 95 Yes 77.88463 63.01132 151.0837 24.80266 No No No
## 96 No 67.36306 50.47521 155.5862 23.08630 No No No
## 101 No 63.84618 64.31109 157.2926 21.13323 No No No
## 114 No 77.15367 74.66060 156.6852 33.30785 Yes No No
## 115 No 52.41635 80.58769 167.1310 23.89609 No No Yes
## 119 No 62.43634 42.10377 165.4805 9.16625 Yes No Yes
## 123 No 49.34269 104.77434 161.4174 44.82807 No No No
## 125 No 59.30121 51.41380 165.1552 17.72833 No No No
## 129 No 73.77152 66.03944 150.5826 26.39790 No No No
## 130 No 70.84479 67.71070 160.9913 25.31320 No Yes Yes
## 131 No 58.42116 93.61188 161.7607 42.75914 No No Yes
## 132 No 59.94304 89.60734 160.7935 34.82814 No No No
## 133 No 55.54115 97.15453 161.7987 36.80392 No No No
## 134 No 64.90212 108.09562 166.1345 32.58611 No No Yes
## 147 No 85.93228 55.64044 159.1193 26.74693 No No No
## 149 No 49.68731 54.37133 167.1934 29.76133 No No No
## 152 No 75.61317 58.56937 169.5201 23.84951 Yes No No
## 154 No 78.45223 41.20816 151.4335 30.11445 No No No
## 167 No 86.62737 64.96704 165.1940 16.90350 No No No
## 171 No 75.61478 69.57171 159.4138 22.74078 No Yes Yes
## 173 No 55.67402 81.87472 166.8057 32.77548 No No No
## 174 Yes 68.01407 75.37379 167.9702 26.12219 Yes No Yes
## 178 No 75.70859 63.06373 159.2462 29.29897 No No No
## 181 Yes 85.87765 62.68355 159.7670 27.64493 No Yes Yes
## 183 Yes 60.58292 60.84491 162.8829 25.57512 No No No
## 191 No 63.79273 73.16381 164.0901 22.82040 No No No
## 192 No 56.09291 72.69166 165.6456 32.77971 No No No
## 193 No 85.90909 70.08774 165.9126 24.11410 No No Yes
## 194 No 57.16070 45.26298 158.8558 23.19351 No No No
## 196 No 70.09400 60.30959 162.6706 21.03660 No No No
## 202 Yes 69.31314 101.98188 169.6793 37.20184 Yes No Yes
## 206 Yes 90.48592 95.94250 164.2126 28.53617 No No Yes
## 212 No 68.40078 53.32512 159.8395 19.18465 No No Yes
## 215 Yes 80.45288 57.37017 161.7937 28.66887 No No No
## 216 No 54.19421 84.98605 163.8884 29.83594 No Yes No
## 245 Yes 79.85546 87.40246 165.6858 17.44596 No Yes No
## 251 No 67.06658 46.64128 168.7791 19.58438 No No No
## 255 No 49.02973 79.58278 157.9924 35.48278 No No No
## 268 No 66.07126 54.15574 156.3523 23.75320 No No No
## 271 No 66.75984 85.12307 155.6291 23.75203 No No No
## 287 Yes 78.52709 94.06132 154.8897 43.51361 No No Yes
## 304 Yes 83.15440 65.19466 155.5642 27.58775 No No No
## 307 No 62.03468 75.75596 166.0373 30.43237 No No No
## 308 No 74.89180 76.26679 178.4050 28.85425 No Yes Yes
## 316 No 83.32635 63.96169 161.4763 20.96325 No No Yes
## 322 Yes 65.07251 83.04850 162.5360 27.35425 No No Yes
## 324 No 79.22991 81.94381 153.9505 29.79879 No No No
## 328 Yes 68.80115 93.83109 153.2871 40.79668 Yes No Yes
## 332 Yes 80.92232 67.64243 171.8766 23.30824 No No Yes
## 338 Yes 71.82578 94.80002 157.7750 35.30464 Yes No Yes
## 341 No 66.38798 102.46474 165.4121 36.27937 No No No
## 342 No 78.72398 87.09566 150.1880 30.51290 Yes No Yes
## 349 No 64.50271 49.18944 156.7449 20.50738 No No No
## 359 No 62.52292 100.54627 168.0699 33.58132 No No No
## 365 Yes 68.18063 85.95007 165.6682 37.97065 Yes No Yes
## 367 Yes 72.52505 97.96208 152.4930 37.26589 Yes No Yes
## 369 No 67.34689 65.91308 163.2653 25.11595 Yes No No
## 370 Yes 63.98056 73.45444 159.9504 27.70143 No No No
## 371 Yes 71.63868 74.34485 162.4793 26.20660 No No No
## 373 No 70.19248 72.27677 165.9720 30.46132 Yes No Yes
## 386 No 79.95486 60.95334 156.5589 14.74609 No No No
## 394 No 60.60432 64.59239 162.0505 24.72927 No Yes No
## 401 Yes 99.97096 47.86834 171.0649 23.08673 No No Yes
## 402 Yes 80.66986 68.95356 158.2671 31.79764 No No No
## 405 Yes 59.76877 122.84484 155.6821 39.88175 No No Yes
## 408 No 62.55655 59.38388 164.8489 27.68278 Yes Yes No
## 409 No 73.22676 92.32035 174.4349 30.46722 No Yes Yes
## 410 No 77.68708 70.42943 160.2517 23.56759 No No Yes
## 414 Yes 51.68563 69.47707 161.5326 29.98921 Yes No Yes
## 417 No 75.47389 27.31324 151.2068 19.29066 No No No
## 419 No 65.74147 81.52171 161.2190 38.76250 No No Yes
## 422 No 74.52953 66.90172 172.3640 20.99491 No Yes No
## 425 No 73.35627 94.81187 158.9571 33.82040 No No Yes
## 427 No 69.15467 61.52779 159.7849 21.95372 Yes No No
## 442 No 75.25126 104.18491 155.5045 33.70470 No No No
## 464 No 58.21749 104.17692 167.9707 34.91109 Yes No Yes
## 466 Yes 57.71115 96.67726 158.2258 30.39015 No No Yes
## 467 Yes 65.05855 75.58737 158.4736 36.12302 Yes No Yes
## 470 No 61.56829 41.91596 160.0358 28.06184 No No No
## 472 No 71.79234 45.30784 154.5072 23.55765 No No No
## 475 Yes 81.15663 83.93356 148.8370 37.00882 No No Yes
## 485 No 69.52256 55.82314 150.2678 31.46179 Yes No Yes
## 492 Yes 70.63612 84.79887 153.6015 27.97732 No No Yes
## 498 No 63.21837 78.27270 160.4743 34.01245 Yes No No
## 499 Yes 98.02976 45.41424 171.3021 19.35205 No No No
## smoke raterisk fracscore fracture bonemed bonemed_fu bonetreat
## 2 No Less 1.1545226 No No No No
## 3 No Same 5.4329275 No No No No
## 12 No Same 2.5240105 No No No No
## 15 No Greater 2.2639318 No Yes Yes Yes
## 18 No Less 0.3632298 No No No No
## 19 No Greater 5.1172785 No Yes Yes Yes
## 27 No Less 1.9584144 No No No No
## 28 No Same 0.4305433 No No No No
## 31 No Greater 7.0736960 No No Yes No
## 47 No Same -0.6562039 No No No No
## 56 No Same 1.1351559 No No No No
## 62 No Greater 3.3393347 No No No No
## 65 No Same 5.1773926 No No No No
## 71 No Less 6.7914767 No No No No
## 73 No Greater 4.5702854 No Yes Yes Yes
## 95 No Greater 5.1219625 No No No No
## 96 No Less 1.5611527 No Yes Yes Yes
## 101 No Greater 1.5349583 No No No No
## 114 No Less 1.0402130 No No No No
## 115 Yes Greater 2.9905508 No No No No
## 119 No Greater 5.1722506 No Yes Yes Yes
## 123 No Less -1.0183783 No No No No
## 125 No Same 0.5921612 No Yes Yes Yes
## 129 No Same 1.7059026 No Yes Yes Yes
## 130 No Less 6.5049889 No No No No
## 131 No Less 2.7109491 No No No No
## 132 No Less 3.4291074 No No No No
## 133 Yes Less 0.5244556 No No No No
## 134 No Same 3.2446350 No No No No
## 147 No Less 2.1135076 No No No No
## 149 No Same -0.6414349 No No No No
## 152 No Less 3.3065610 No No No No
## 154 No Same 4.2138586 No Yes Yes Yes
## 167 No Same 2.7417315 No No No No
## 171 No Less 6.5812280 No No No No
## 173 No Less -1.6904375 No No No No
## 174 No Greater 5.7206517 No Yes Yes Yes
## 178 No Same 4.6750500 No No No No
## 181 No Same 9.6754326 No Yes Yes Yes
## 183 Yes Same 2.6116172 No No No No
## 191 No Same 1.0365044 No No No No
## 192 No Same 1.5778644 No No No No
## 193 No Same 7.1317797 No No No No
## 194 No Less 0.2023886 No No No No
## 196 No Less 2.5068103 No No No No
## 202 No Greater 7.4145642 No No Yes No
## 206 No Greater 6.8293497 No No No No
## 212 No Less 7.2992465 No Yes Yes Yes
## 215 No Less 5.4297358 No Yes Yes Yes
## 216 No Less 0.3320230 No No No No
## 245 No Greater 5.4965278 No Yes Yes Yes
## 251 No Less 2.2718340 No No No No
## 255 Yes Less 1.5496948 No No No No
## 268 No Less 2.3754443 No No No No
## 271 No Same 1.8226215 Yes Yes Yes Yes
## 287 No Less 7.8515610 Yes No No No
## 304 No Same 6.6671755 Yes Yes Yes Yes
## 307 No Same 0.7229242 Yes No No No
## 308 No Less 8.0477422 Yes No No No
## 316 No Same 5.5676163 Yes Yes Yes Yes
## 322 No Same 5.3158170 Yes No No No
## 324 No Same 3.9349309 Yes Yes Yes Yes
## 328 No Greater 7.5961813 Yes No No No
## 332 No Greater 5.8693506 Yes No No No
## 338 No Greater 8.5224982 Yes No No No
## 341 No Greater 1.7254001 Yes Yes Yes Yes
## 342 No Less 3.2335355 Yes Yes Yes Yes
## 349 No Same 2.5541324 Yes Yes Yes Yes
## 359 No Greater 3.1301876 Yes Yes Yes Yes
## 365 No Greater 6.9395759 Yes No No No
## 367 No Greater 7.3925026 Yes No No No
## 369 No Greater 4.1782532 Yes Yes Yes Yes
## 370 No Less 2.4080565 Yes No No No
## 371 No Greater 6.5985888 Yes No Yes No
## 373 No Same 2.3974792 Yes No No No
## 386 No Less 3.4108887 Yes Yes No No
## 394 No Same 2.1137251 Yes No No No
## 401 No Greater 10.4291243 Yes No No No
## 402 No Same 5.7535958 Yes Yes Yes Yes
## 405 No Greater 3.5563597 Yes Yes Yes Yes
## 408 No Same 3.0922428 Yes No No No
## 409 No Less 3.5029658 Yes No No No
## 410 No Same 7.6589157 Yes Yes Yes Yes
## 414 No Greater 4.5321059 Yes No No No
## 417 No Greater 5.1845967 Yes Yes Yes Yes
## 419 No Less 6.2047670 Yes No No No
## 422 No Same 0.1605734 Yes No No No
## 425 No Less 3.7065073 Yes No No No
## 427 No Greater 3.9457952 Yes Yes Yes Yes
## 442 No Less -0.5697186 Yes No No No
## 464 Yes Same 3.9231870 Yes No No No
## 466 No Same 6.1457104 Yes No No No
## 467 No Greater 6.6828235 Yes No No No
## 470 No Same 5.0400135 Yes Yes Yes Yes
## 472 No Same 4.8106546 Yes Yes Yes Yes
## 475 No Same 7.5611009 Yes No No No
## 485 No Same 3.1057034 Yes Yes Yes Yes
## 492 No Greater 5.9810467 Yes Yes Yes Yes
## 498 No Greater 1.4036609 Yes No No No
## 499 No Less 9.6447381 Yes Yes No No
##
## $results
##
## Recursive feature selection
##
## Outer resampling method: Cross-Validated (10 fold)
##
## Resampling performance over subset size:
##
## Variables Accuracy Kappa AccuracySD KappaSD Selected
## 1 1 1 0 0 *
## 2 1 1 0 0
## 3 1 1 0 0
## 4 1 1 0 0
## 5 1 1 0 0
## 14 1 1 0 0
##
## The top 1 variables (out of 1):
## fracture
##
##
## $control
## $control$functions
## $control$functions$summary
## function (data, lev = NULL, model = NULL)
## {
## if (is.character(data$obs))
## data$obs <- factor(data$obs, levels = lev)
## postResample(data[, "pred"], data[, "obs"])
## }
## <bytecode: 0x000002190a4e0df8>
## <environment: namespace:caret>
##
## $control$functions$fit
## function (x, y, first, last, ...)
## {
## loadNamespace("randomForest")
## randomForest::randomForest(x, y, importance = TRUE, ...)
## }
## <bytecode: 0x000002190a4e00d8>
## <environment: namespace:caret>
##
## $control$functions$pred
## function (object, x)
## {
## tmp <- predict(object, x)
## if (is.factor(object$y)) {
## out <- cbind(data.frame(pred = tmp), as.data.frame(predict(object,
## x, type = "prob"), stringsAsFactors = TRUE))
## }
## else out <- tmp
## out
## }
## <bytecode: 0x000002190a4dfc08>
## <environment: namespace:caret>
##
## $control$functions$rank
## function (object, x, y)
## {
## vimp <- varImp(object)
## if (is.factor(y)) {
## if (all(levels(y) %in% colnames(vimp))) {
## avImp <- apply(vimp[, levels(y), drop = TRUE], 1,
## mean)
## vimp$Overall <- avImp
## }
## }
## vimp <- vimp[order(vimp$Overall, decreasing = TRUE), , drop = FALSE]
## if (ncol(x) == 1) {
## vimp$var <- colnames(x)
## }
## else vimp$var <- rownames(vimp)
## vimp
## }
## <bytecode: 0x000002190a4e2d60>
## <environment: namespace:caret>
##
## $control$functions$selectSize
## function (x, metric, maximize)
## {
## best <- if (maximize)
## which.max(x[, metric])
## else which.min(x[, metric])
## min(x[best, "Variables"])
## }
## <bytecode: 0x000002190a4e4980>
## <environment: namespace:caret>
##
## $control$functions$selectVar
## function (y, size)
## {
## finalImp <- ddply(y[, c("Overall", "var")], .(var), function(x) mean(x$Overall,
## na.rm = TRUE))
## names(finalImp)[2] <- "Overall"
## finalImp <- finalImp[order(finalImp$Overall, decreasing = TRUE),
## ]
## as.character(finalImp$var[1:size])
## }
## <bytecode: 0x000002190a4e4248>
## <environment: namespace:caret>
##
##
## $control$rerank
## [1] FALSE
##
## $control$method
## [1] "cv"
##
## $control$saveDetails
## [1] TRUE
##
## $control$number
## [1] 10
##
## $control$repeats
## [1] 1
##
## $control$returnResamp
## [1] "all"
##
## $control$verbose
## [1] FALSE
##
## $control$p
## [1] 0.75
##
## $control$index
## NULL
##
## $control$indexOut
## NULL
##
## $control$timingSamps
## [1] 0
##
## $control$seeds
## [1] NA
##
## $control$allowParallel
## [1] TRUE# Set up parallel processing
library(doParallel)
cl <- makeCluster(detectCores() - 1) # Use one less than the total number of cores
registerDoParallel(cl)
# Suppress warnings to clean up model training output
options(warn = -1)
# Train models
rf_model <- train(fracture ~ ., data = train_set, method = "rf", trControl = fit_control)## Aggregating results
## Selecting tuning parameters
## Fitting mtry = 2 on full training setknn_model <- train(fracture ~ ., data = train_set, method = "knn", trControl = fit_control)## Aggregating results
## Selecting tuning parameters
## Fitting k = 7 on full training settree_model <- train(fracture ~ ., data = train_set, method = "rpart", trControl = fit_control)## Aggregating results
## Selecting tuning parameters
## Fitting cp = 0.0473 on full training set# Train XGBoost model with a comprehensive tuning grid
xgb_model <- train(
fracture ~ .,
data = train_set,
method = "xgbTree",
trControl = fit_control,
tuneGrid = expand.grid(
nrounds = 100,
max_depth = c(3, 6, 9),
eta = c(0.01, 0.1, 0.3),
gamma = c(0, 0.1, 0.2),
colsample_bytree = c(0.5, 0.75, 1),
min_child_weight = c(1, 3, 5),
subsample = c(0.5, 0.75, 1)
),
verbose = FALSE
)## Aggregating results
## Selecting tuning parameters
## Fitting nrounds = 100, max_depth = 9, eta = 0.01, gamma = 0.1, colsample_bytree = 0.75, min_child_weight = 1, subsample = 1 on full training set# Stop parallel processing and reset options
stopCluster(cl)
registerDoSEQ()
options(warn = 0) # Reset warning level# Define function to extract and print model metrics
extract_metrics <- function(model, data, outcome_col) {
predictions <- predict(model, newdata = data)
prob_predictions <- predict(model, newdata = data, type = "prob")
confusion <- confusionMatrix(predictions, data[[outcome_col]])
roc_result <- roc(response = data[[outcome_col]], predictor = prob_predictions[,2])
list(
Sensitivity = confusion$byClass['Sensitivity'],
Specificity = confusion$byClass['Specificity'],
PPV = confusion$byClass['Pos Pred Value'],
NPV = confusion$byClass['Neg Pred Value'],
Accuracy = confusion$overall['Accuracy'],
AUROC = auc(roc_result)
)
}
# Evaluate models
rf_metrics <- extract_metrics(rf_model, validation_set, "fracture")## Setting levels: control = No, case = Yes## Setting direction: controls < casesknn_metrics <- extract_metrics(knn_model, validation_set, "fracture")## Setting levels: control = No, case = Yes## Setting direction: controls > casestree_metrics <- extract_metrics(tree_model, validation_set, "fracture")## Setting levels: control = No, case = Yes## Setting direction: controls < casesxgb_metrics <- extract_metrics(xgb_model, validation_set, "fracture")## Setting levels: control = No, case = Yes
## Setting direction: controls < cases# Print metrics
print("Random Forest Metrics:")## [1] "Random Forest Metrics:"print(rf_metrics)## $Sensitivity
## Sensitivity
## 0.7209302
##
## $Specificity
## Specificity
## 0.6388889
##
## $PPV
## Pos Pred Value
## 0.7045455
##
## $NPV
## Neg Pred Value
## 0.6571429
##
## $Accuracy
## Accuracy
## 0.6835443
##
## $AUROC
## Area under the curve: 0.7574print("KNN Metrics:")## [1] "KNN Metrics:"print(knn_metrics)## $Sensitivity
## Sensitivity
## 0.6744186
##
## $Specificity
## Specificity
## 0.2222222
##
## $PPV
## Pos Pred Value
## 0.5087719
##
## $NPV
## Neg Pred Value
## 0.3636364
##
## $Accuracy
## Accuracy
## 0.4683544
##
## $AUROC
## Area under the curve: 0.5252print("Decision Tree Metrics:")## [1] "Decision Tree Metrics:"print(tree_metrics)## $Sensitivity
## Sensitivity
## 0.6744186
##
## $Specificity
## Specificity
## 0.6111111
##
## $PPV
## Pos Pred Value
## 0.6744186
##
## $NPV
## Neg Pred Value
## 0.6111111
##
## $Accuracy
## Accuracy
## 0.6455696
##
## $AUROC
## Area under the curve: 0.6376print("XGBoost Metrics:")## [1] "XGBoost Metrics:"print(xgb_metrics)## $Sensitivity
## Sensitivity
## 0.7209302
##
## $Specificity
## Specificity
## 0.5277778
##
## $PPV
## Pos Pred Value
## 0.6458333
##
## $NPV
## Neg Pred Value
## 0.6129032
##
## $Accuracy
## Accuracy
## 0.6329114
##
## $AUROC
## Area under the curve: 0.6789# Feature Importance Plot for Random Forest
importance <- varImp(rf_model, scale = FALSE)
plot(importance)
# Correlation matrix of the model predictions to compare model agreement
predictions_rf <- predict(rf_model, validation_set, type = "prob")
predictions_knn <- predict(knn_model, validation_set, type = "prob")
predictions_tree <- predict(tree_model, validation_set, type = "prob")
predictions_xgb <- predict(xgb_model, validation_set, type = "prob")
# Assuming binary classification and interested in positive class probabilities
cor_matrix <- cor(cbind(predictions_rf[,2], predictions_knn[,2], predictions_tree[,2], predictions_xgb[,2]),
method = "pearson")
print(cor_matrix)## [,1] [,2] [,3] [,4]
## [1,] 1.0000000 0.4698662 0.8070576 0.8628819
## [2,] 0.4698662 1.0000000 0.5212532 0.4023063
## [3,] 0.8070576 0.5212532 1.0000000 0.6969978
## [4,] 0.8628819 0.4023063 0.6969978 1.0000000# Evaluate models on the test set using the metrics already defined
test_metrics_rf <- extract_metrics(rf_model, test_set, "fracture")## Setting levels: control = No, case = Yes
## Setting direction: controls < casestest_metrics_knn <- extract_metrics(knn_model, test_set, "fracture")## Setting levels: control = No, case = Yes
## Setting direction: controls < casestest_metrics_tree <- extract_metrics(tree_model, test_set, "fracture")## Setting levels: control = No, case = Yes
## Setting direction: controls < casestest_metrics_xgb <- extract_metrics(xgb_model, test_set, "fracture")## Setting levels: control = No, case = Yes
## Setting direction: controls < cases# Print test metrics for each model
print("Test Metrics - Random Forest:")## [1] "Test Metrics - Random Forest:"print(test_metrics_rf)## $Sensitivity
## Sensitivity
## 0.6851852
##
## $Specificity
## Specificity
## 0.5652174
##
## $PPV
## Pos Pred Value
## 0.6491228
##
## $NPV
## Neg Pred Value
## 0.6046512
##
## $Accuracy
## Accuracy
## 0.63
##
## $AUROC
## Area under the curve: 0.7206print("Test Metrics - KNN:")## [1] "Test Metrics - KNN:"print(test_metrics_knn)## $Sensitivity
## Sensitivity
## 0.7037037
##
## $Specificity
## Specificity
## 0.3043478
##
## $PPV
## Pos Pred Value
## 0.5428571
##
## $NPV
## Neg Pred Value
## 0.4666667
##
## $Accuracy
## Accuracy
## 0.52
##
## $AUROC
## Area under the curve: 0.5215print("Test Metrics - Decision Tree:")## [1] "Test Metrics - Decision Tree:"print(test_metrics_tree)## $Sensitivity
## Sensitivity
## 0.6851852
##
## $Specificity
## Specificity
## 0.4565217
##
## $PPV
## Pos Pred Value
## 0.5967742
##
## $NPV
## Neg Pred Value
## 0.5526316
##
## $Accuracy
## Accuracy
## 0.58
##
## $AUROC
## Area under the curve: 0.5773print("Test Metrics - XGBoost:")## [1] "Test Metrics - XGBoost:"print(test_metrics_xgb)## $Sensitivity
## Sensitivity
## 0.6666667
##
## $Specificity
## Specificity
## 0.4565217
##
## $PPV
## Pos Pred Value
## 0.5901639
##
## $NPV
## Neg Pred Value
## 0.5384615
##
## $Accuracy
## Accuracy
## 0.57
##
## $AUROC
## Area under the curve: 0.6135# Plot ROC curves for each model using the test set predictions
library(pROC)
roc_rf <- roc(response = test_set$fracture, predictor = predict(rf_model, test_set, type = "prob")[,2])## Setting levels: control = No, case = Yes## Setting direction: controls < casesroc_knn <- roc(response = test_set$fracture, predictor = predict(knn_model, test_set, type = "prob")[,2])## Setting levels: control = No, case = Yes
## Setting direction: controls < casesroc_tree <- roc(response = test_set$fracture, predictor = predict(tree_model, test_set, type = "prob")[,2])## Setting levels: control = No, case = Yes
## Setting direction: controls < casesroc_xgb <- roc(response = test_set$fracture, predictor = predict(xgb_model, test_set, type = "prob")[,2])## Setting levels: control = No, case = Yes
## Setting direction: controls < casesplot(roc_rf, col="#1F77B4", lwd=2, main="ROC Curves for Models")
plot(roc_knn, col="#FF7F0E", lwd=2, add=TRUE)
plot(roc_tree, col="#2CA02C", lwd=2, add=TRUE)
plot(roc_xgb, col="#D62728", lwd=2, add=TRUE)
legend("bottomright", legend=c("Random Forest", "KNN", "Decision Tree", "XGBoost"),
col=c("#1F77B4", "#FF7F0E", "#2CA02C", "#D62728"), lwd=2)
```{compare-and-analyze, cache=TRUE} # Comparison visualization and analysis # Combine AUC and other metrics into a single data frame for comparison aucs <- data.frame( Model = c(“Random Forest”, “KNN”, “Decision Tree”, “XGBoost”), AUC = c(auc(roc_rf), auc(roc_knn), auc(roc_tree), auc(roc_xgb)), Accuracy = c(test_metrics_rf\(Accuracy, test_metrics_knn\)Accuracy, test_metrics_tree\(Accuracy, test_metrics_xgb\)Accuracy) )
Bar plot of AUCs using ggplot2
library(ggplot2) ggplot(aucs, aes(x=Model, y=AUC, fill=Model)) + geom_bar(stat=“identity”, color=“black”) + theme_minimal() + labs(title=“Comparison of Model AUCs”, x=“Model”, y=“AUC Value”) + scale_fill_brewer(palette=“Set1”)
```