## 3 observations missing `PIB` have been removed. To include these observations, use `forcats::fct_explicit_na()` on `PIB` column before passing to `tbl_summary()`.
| Characteristic | negative, N = 61 | positive, N = 141 |
|---|---|---|
| MMSE | 27.50 (25.25, 29.00) | 24.00 (21.25, 25.75) |
| CDR | 0.75 (0.50, 1.00) | 0.50 (0.50, 1.00) |
| FDG | ||
| AD pattern | 3 (50%) | 12 (86%) |
| Others | 2 (33%) | 2 (14%) |
| Vascular | 1 (17%) | 0 (0%) |
| Memoria alterada | 5 (83%) | 11 (79%) |
| Lenguaje alterado | 3 (50%) | 9 (64%) |
| 1 Median (IQR); n (%) | ||
Aca testeamos el poder diagnostico de obtener un PIB positivo representado por el area abajo de la curva roc (AUC)
## Setting levels: control = 0, case = 1
## Setting direction: controls < cases
##
## Call:
## roc.default(response = MCI_CUDIM$DX, predictor = MCI_CUDIM$`Recuerdo de palabras - - puntaje`)
##
## Data: MCI_CUDIM$`Recuerdo de palabras - - puntaje` in 6 controls (MCI_CUDIM$DX 0) < 14 cases (MCI_CUDIM$DX 1).
## Area under the curve: 0.6429
## Setting levels: control = 0, case = 1
## Setting direction: controls > cases
##
## Call:
## roc.default(response = MCI_CUDIM$DX, predictor = MCI_CUDIM$`Praxia constructiva - puntaje`)
##
## Data: MCI_CUDIM$`Praxia constructiva - puntaje` in 6 controls (MCI_CUDIM$DX 0) > 14 cases (MCI_CUDIM$DX 1).
## Area under the curve: 0.5774
## Setting levels: control = 0, case = 1
## Setting direction: controls < cases
##
## Call:
## roc.default(response = MCI_CUDIM$DX, predictor = MCI_CUDIM$`Recuerdo de palabras - puntaje`)
##
## Data: MCI_CUDIM$`Recuerdo de palabras - puntaje` in 6 controls (MCI_CUDIM$DX 0) < 14 cases (MCI_CUDIM$DX 1).
## Area under the curve: 0.631
## Setting levels: control = 0, case = 1
## Setting direction: controls < cases
##
## Call:
## roc.default(response = MCI_CUDIM$DX, predictor = MCI_CUDIM$`Reconocimiento - puntaje`)
##
## Data: MCI_CUDIM$`Reconocimiento - puntaje` in 6 controls (MCI_CUDIM$DX 0) < 14 cases (MCI_CUDIM$DX 1).
## Area under the curve: 0.7202
## Setting levels: control = 0, case = 1
## Setting direction: controls > cases
##
## Call:
## roc.default(response = MCI_CUDIM$DX, predictor = MCI_CUDIM$`Fluencia verbal semántica`)
##
## Data: MCI_CUDIM$`Fluencia verbal semántica` in 6 controls (MCI_CUDIM$DX 0) > 14 cases (MCI_CUDIM$DX 1).
## Area under the curve: 0.5536
## Setting levels: control = 0, case = 1
## Setting direction: controls > cases
##
## Call:
## roc.default(response = MCI_CUDIM$DX, predictor = MCI_CUDIM$RecordTotalCorrect)
##
## Data: MCI_CUDIM$RecordTotalCorrect in 6 controls (MCI_CUDIM$DX 0) > 11 cases (MCI_CUDIM$DX 1).
## Area under the curve: 0.6212
## Setting levels: control = 0, case = 1
## Setting direction: controls < cases
##
## Call:
## roc.default(response = MCI_CUDIM$DX, predictor = MCI_CUDIM$`Recuerdo de palabras - - puntaje`, percent = TRUE, plot = TRUE, legacy.axes = FALSE, col = "salmon", lwd = 2, print.auc = TRUE)
##
## Data: MCI_CUDIM$`Recuerdo de palabras - - puntaje` in 6 controls (MCI_CUDIM$DX 0) < 14 cases (MCI_CUDIM$DX 1).
## Area under the curve: 64.29%
## Setting levels: control = 0, case = 1
## Setting direction: controls < cases
## Setting levels: control = 0, case = 1
## Setting direction: controls < cases
## Setting levels: control = 0, case = 1
## Setting direction: controls > cases
En estos resulatos encontraremos el punto de corte optimo con su sensibilidad y especificidad, en este caso el pto de corte es 13.
##
## Attaching package: 'cutpointr'
## The following objects are masked from 'package:pROC':
##
## auc, roc
## Assuming the positive class is 1
## Assuming the positive class has higher x values
## Running bootstrap...
## Method: oc_mean
## Predictor: Reconocimiento - puntaje
## Outcome: DX
## Direction: >=
## Nr. of bootstraps: 100
##
## AUC n n_pos n_neg
## 0.7202 20 14 6
##
## optimal_cutpoint sum_sens_spec acc sensitivity specificity tp fn fp tn
## 13.1 1.4048 0.65 0.5714 0.8333 8 6 1 5
##
## Predictor summary:
## Data Min. 5% 1st Qu. Median Mean 3rd Qu. 95% Max. SD NAs
## Overall 0 1.90 8.75 12.0 13.100000 19.25 23.05 24 7.304649 0
## 0 2 2.75 6.00 9.5 9.166667 10.75 16.25 18 5.492419 0
## 1 0 2.60 9.25 17.5 14.785714 20.75 23.35 24 7.495420 0
##
## Bootstrap summary:
## Variable Min. 5% 1st Qu. Median Mean 3rd Qu. 95% Max. SD
## optimal_cutpoint 10.80 11.50 12.57 13.45 13.48 14.10 15.70 17.10 1.28
## AUC_b 0.43 0.53 0.68 0.73 0.73 0.81 0.89 0.98 0.11
## AUC_oob 0.25 0.40 0.60 0.71 0.71 0.83 1.00 1.00 0.17
## sum_sens_spec_b 0.92 1.07 1.33 1.47 1.44 1.60 1.71 1.85 0.21
## sum_sens_spec_oob 0.50 0.89 1.25 1.40 1.40 1.62 1.81 2.00 0.30
## acc_b 0.40 0.55 0.60 0.65 0.68 0.75 0.80 0.90 0.09
## acc_oob 0.29 0.43 0.57 0.67 0.65 0.75 0.88 1.00 0.14
## sensitivity_b 0.31 0.46 0.56 0.62 0.62 0.67 0.80 0.86 0.10
## sensitivity_oob 0.20 0.25 0.50 0.59 0.57 0.67 0.83 1.00 0.18
## specificity_b 0.33 0.50 0.75 0.83 0.82 1.00 1.00 1.00 0.17
## specificity_oob 0.00 0.50 0.67 1.00 0.83 1.00 1.00 1.00 0.24
## cohens_kappa_b -0.05 0.06 0.24 0.36 0.36 0.50 0.61 0.80 0.18
## cohens_kappa_oob -0.29 -0.08 0.16 0.33 0.32 0.46 0.71 1.00 0.26
## NAs
## 0
## 0
## 3
## 0
## 3
## 0
## 0
## 0
## 0
## 0
## 3
## 0
## 0
## Warning: Removed 3 rows containing non-finite values (stat_density).