Table 1 Comparative performance evaluation for the LSTM-AE pipeline on the malaria cohort
From: Overcoming Data Loss in Wearable Disease Detection with GAN-Based Imputation
LSTM-AE performance evaluation (malaria cohort) | ||||||
|---|---|---|---|---|---|---|
Input | Sensitivity | Specificity | Precision | Accuracy | F1-score | ROC AUC |
Raw | 86.00% | 12.00% | 51.00% | 50.00% | 64.00% | 53.00% |
GAN-imputed | 89.45% | 16.35% | 69.75% | 66.29% | 78.38% | 57.67% |
Difference | +3.45% | +4.35% | +18.75% | +16.29% | +14.38% | +4.67% |
