Background
Methods
Data
Converter MCI
(n = 257)
|
Stable MCI
(n = 462)
|
p-value
| |
---|---|---|---|
Age, years (M ± SD) | 71.7 ± 7.8 | 68.1 ± 8.6 | < 10−8
|
Gender (male/female) | 93/164 | 196/266 | 0.102
#
|
Formal education, years (M ± SD) | 8.9 ± 4.9 | 8.8 ± 4.7 | 0.612 |
Follow-up time, years (M ± SD) | 2.9 ± 2.3 | 3.5 ± 3.0 | 0.007 |
Supervised learning approach using learning examples created with time windows
Creating learning examples using time windows
Learning the prognostic model
Classifier
|
Parameters and respective range
|
---|---|
NB |
Gaussian or Supervised Discrimination or Kernel
|
DT |
Confidence ∈ [0.05,0.5] |
SVM RBF |
Complexity ∈ [10−1, 101] and γ ∈ [10−2, 102] |
SVM Poly |
Complexity ∈ [10−1, 101] and Degree ∈ {1, 2, 3} |
kNN | #Neighbors ∈ [1, 11] |
RF | #Iterations ∈ [5, 30] |
LR |
Ridge ∈ [10−9, 10−6] |
Validating the prognostic model
Using the prognostic model in clinical settings
Results
Creating learning examples using time windows
CV set | Validation set | |||
---|---|---|---|---|
sMCI
|
cMCI
|
sMCI
|
cMCI
| |
FL approach | 377 (62%) | 227 (38%) | 85 (74%) | 30 (26%) |
2-Year window | 280 (75%) | 94 (25%) | 53 (80%) | 13 (20%) |
3-Year window | 206 (60%) | 137 (40%) | 34 (61%) | 22 (39%) |
4-Year window | 146 (47%) | 166 (53%) | 22 (47%) | 25 (53%) |
5-Year window | 106 (36%) | 190 (64%) | 10 (28%) | 26 (72%) |
Learning the prognostic model
Features | FL approach | 2-Year window | 3-Year window | 4-Year window | 5-Year window |
---|---|---|---|---|---|
Age | X | X | X | X | X |
Age of first symptons | X | X | X | X | |
Cancelation Task- A’s time
| X | X | X | ||
Cancelation Task – A’s total
| X | X | |||
Digit Span - Forward
| X | ||||
Digit Span - Backward
| X | X | X | ||
Verbal Paired-Associate Learning – Easy
| X | X | X | ||
Verbal Paired-Associate Learning –Difficult
| X | X | X | X | X |
Verbal Paired-Associate Learning – Total
| X | X | X | X | X |
Logical Memory Immediate A free recall
| X | X | X | X | X |
Logical Memory - A Immediate Cued
| X | X | X | ||
Word Recall – Free recall
| X | X | X | X | X |
Word Recall –Total
| X | X | |||
Logical Memory with Interference-A
| X | X | X | X | |
Orientation (Total)
| X | X | X | X | X |
Orientation – Personal
| X | X | |||
Orientation – Spatial
| X | ||||
Orientation – Temporal
| X | X | X | X | X |
Orientation- MSQ
| X | ||||
Verbal Fluency
| X | X | X | ||
Token Orders (total) | X | ||||
Cube Draw
| X | ||||
Calculation
| X | ||||
Interpretation of Proverbs – (Verbal Abstraction)
| X | X | X | ||
Raven Progressive Matrices | X | X | X | X | X |
Trail Making Test (Part B) - time | X | ||||
CVLT A list (1sttrial) | X | X | |||
CVLT A list (3thtrial) | X | ||||
CVLT A list (4thtrial) | X | X | |||
CVLT A list (five learning trails total) | X | X | X | X | |
CVLT A list (Total intrusions in 5 recalls) | X | ||||
Blessed Dementia Scale (Total of Part 1 - Daily living activities) | X | ||||
Fi_LM_a | X | X | |||
Fi_LM_a_m100 | X | X | |||
Cancelation Task – A’s total (Z-score) | X | X | |||
Digit Span – Forward (Z-score) | X | X | |||
Digit Span – Backward (Z-score) | X | X | |||
Digit Span – Total (Z-score) | X | X | X | ||
Verbal Paired-Associate Learning (Z-score) | X | X | X | X | X |
Informatio (Z-score) | X | X | X | ||
Orientation (Total) (Z-score) | X | X | X | X | X |
Orientation- MSQ (Z-score) | X | X | X | X | X |
Word Recall –Total (Z-score) | X | X | |||
Verbal Fluency (Z-score) | X | X | X | X | X |
Interpretation of Proverbs – (Verbal Abstraction) (Z-score) | X | X | X | X | |
Raven Progressive Matrices (Z-score) | X | X | X | X | X |
Cancelation task -Toulouse- Pierón (concentration index) (Z-score) | X | ||||
CVLT A list (five learning trails total) (Z-score) | X | X | |||
CVLT A list (5sttrial) (Z-score) | X | X | |||
Logical Memory Immediate A free recall (Z-score) | X | X | X | X | X |
Logical Memory with Interference-A (Z-score) | X | X | X | X |
AUC | Sensitivity | Specificity | |||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
FL
|
2Y
|
3Y
|
4Y
|
5Y
|
FL
|
2Y
|
3Y
|
4Y
|
5Y
|
FL
|
2Y
|
3Y
|
4Y
|
5Y
| |
DT |
0.65 ± 0.02
|
0.71 ± 0.04
|
0.75 ± 0.01
|
0.78 ± 0.02
|
0.79 ± 0.02
|
0.59 ± 0.03
|
0.65 ± 0.05
|
0.75 ± 0.04
|
0.71 ± 0.04
|
0.77 ± 0.02
|
0.68 ± 0.02
|
0.73 ± 0.04
|
0.69 ± 0.03
|
0.77 ± 0.03
|
0.75 ± 0.02
|
kNN |
0.67 ± 0.01
|
0.77 ± 0.01
|
0.82 ± 0.01
|
0.83 ± 0.01
|
0.84 ± 0.01
|
0.60 ± 0.02
|
0.70 ± 0.03
|
0.87 ± 0.01
|
0.69 ± 0.03
|
0.83 ± 0.01
|
0.65 ± 0.01
|
0.71 ± 0.01
|
0.61 ± 0.01
|
0.81 ± 0.02
|
0.72 ± 0.03
|
SVM Poly |
0.63 ± 0.01
|
0.70 ± 0.01
|
0.76 ± 0.01
|
0.79 ± 0.01
|
0.80 ± 0.01
|
0.43 ± 0.02
|
0.55 ± 0.02
|
0.71 ± 0.01
|
0.81 ± 0.02
|
0.86 ± 0.01
|
0.83 ± 0.01
|
0.84 ± 0.01
|
0.81 ± 0.01
|
0.77 ± 0.01
|
0.75 ± 0.02
|
SVM RBF |
0.63 ± 0.01
|
0.64 ± 0.01
|
0.76 ± 0.01
|
0.79 ± 0.01
|
0.80 ± 0.02
|
0.40 ± 0.02
|
0.35 ± 0.02
|
0.72 ± 0.02
|
0.80 ± 0.02
|
0.89 ± 0.01
|
0.86 ± 0.01
|
0.93 ± 0.01
|
0.81 ± 0.01
|
0.78 ± 0.02
|
0.71 ± 0.03
|
NB |
0.74 ± 0.00
|
0.82 ± 0.01
|
0.86 ± 0.00
|
0.87 ± 0.01
|
0.88 ± 0.00
|
0.64 ± 0.02
|
0.66 ± 0.01
|
0.75 ± 0.02
|
0.82 ± 0.01
|
0.88 ± 0.01
|
0.71 ± 0.02
|
0.82 ± 0.01
|
0.79 ± 0.01
|
0.78 ± 0.01
|
0.71 ± 0.01
|
LR |
0.72 ± 0.01
|
0.79 ± 0.01
|
0.84 ± 0.01
|
0.84 ± 0.01
|
0.85 ± 0.01
|
0.47 ± 0.01
|
0.77 ± 0.02
|
0.85 ± 0.03
|
0.74 ± 0.01
|
0.78 ± 0.01
|
0.80 ± 0.01
|
0.66 ± 0.01
|
0.68 ± 0.02
|
0.81 ± 0.02
|
0.78 ± 0.02
|
RF |
0.72 ± 0.01
|
0.79 ± 0.01
|
0.85 ± 0.01
|
0.86 ± 0.01
|
0.87 ± 0.01
|
0.59 ± 0.03
|
0.53 ± 0.04
|
0.75 ± 0.01
|
0.75 ± 0.01
|
0.87 ± 0.01
|
0.71 ± 0.02
|
0.86 ± 0.01
|
0.77 ± 0.02
|
0.81 ± 0.01
|
0.70 ± 0.02
|
Best prognostic model
Validating the prognostic model
AUC | Sensitivity | Specificity | |||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
FL
|
2Y
|
3Y
|
4Y
|
5Y
|
FL
|
2Y
|
3Y
|
4Y
|
5Y
|
FL
|
2Y
|
3Y
|
4Y
|
5Y
| |
Naïve Bayes
|
0.61
|
0.73
|
0.74
|
0.72
|
0.76
|
0.40
|
0.69
|
0.64
|
0.56
|
0.56
|
0.73
|
0.77
|
0.76
|
0.68
|
0.70
|