Table 1 Image quality assessment model training parameters.
From: An AI-powered navigation framework to achieve an automated acquisition of cardiac ultrasound images
Parameters | Value |
|---|---|
No. of training epochs (set) | 200 |
No. of training epochs (true) | 156 |
Batch size | 32 |
Optimiser | Stochastic gradient descent |
Loss | Sparse categorical cross entropy |
Learning rate | 0.00001 |
Momentum | 0.9 |
Early stopping condition | Yes, on no improvement in validation loss during 3 consecutive epochs |
