Reconstructing the Photoacoustic Image with High Quality using the Deep Neural Network Model

Table 1

Compared the performance of the suggested method with other techniques.

Approach	PSNR	Ac	Pn	Rl	AUC measure	F1-measure
KNN	36.4 ± 4.5	92.5	91.5	90	90.4	91
SVM	38.2 ± 4.8	95.5	95.2	94.8	94.4	93
Naïve bayes	37.6 ± 4.4	89.8	89.5	89.2	88.4	88
RNN	35.2 ± 5.2	93.6	94	93.6	93.2	92.7
ANN	35.4 ± 4.2	98	97.2	94	94.5	93
Proposed PS-CNN	39.5 ± 5.4	99.5	99.0	96	98.5	95

Ac, abeats conventional techniques withccuracy; Pn, precision; PSNR, peak signal-to-noise ratio; Rl, recall.