Fig. 8: Schematic of the A-line-optimized spectral undersampling method. | Light: Science & Applications

Fig. 8: Schematic of the A-line-optimized spectral undersampling method.

From: Neural network-based image reconstruction in swept-source optical coherence tomography using undersampled spectral data

Fig. 8

A continuous, trainable vector was initialized and then binarized by a rounding function with a threshold of T = 0.5. The pointwise multiplication of the binarized vector and a regular 2× undersampling spectral grid forms the A-line-optimized undersampling grid, which was then applied to raw OCT fringes to generate undersampled fringes with Nspec < 640. After the training process, the converged A-line optimized undersampling grid is shown in Supplementary Fig. S5.

Back to article page