Table 3 Classification accuracy derived from the SP classification and the relevant structure as the number of hidden units
From: An intelligent Alzheimer’s disease diagnosis method using unsupervised feature learning
Models | Hidden units | Accuracy (ACC) |
|---|---|---|
AD vs. HC | ||
MRI | 500-50-10 | 0.876 ± 0.017 |
PET | 1000-50-30 | 0.879 ± 0.023 |
CSF | 50-3 | 0.845 ± 0.014 |
CONTACT | 500-100-20 | 0.902 ± 0.019 |
MCI vs. HC | ||
MRI | 100-100-20 | 4 ± 0.018 |
PET | 300-50-10 | 0.694 ± 0.014 |
CSF | 10-3 | 0.704 ± 0.019 |
CONTACT | 100-50-20 | 0.752 ± . 0.017 |
AD vs. MCI | ||
MRI | 1000-50-30 | 0.662 ± 0.021 |
PET | 100-50-10 | 0.678 ± 0.020 |
CSF | 10-1 | 0.687 ± 0.011 |
CONTACT | 100-100-20 | 0.712 ± 0.019 |
MCI-C vs. MCI-NC | ||
MRI | 100-100-10 | 0.561 ± 0.030 |
PET | 100-100-10 | 0.614 ± 0.028 |
CSF | 30-2 | 0.601 ± 0.021 |
CONTACT | 500-50-20 | 0.625 ± 0.025 |