. Author manuscript; available in PMC: 2018 Mar 1.

Published in final edited form as: IEEE Trans Ultrason Ferroelectr Freq Control. 2016 Nov 29;64(3):491–499. doi: 10.1109/TUFFC.2016.2633429

TABLE VI.

Signal to noise ratios TM phantom experiment

ROI	Algorithm	4 slices	6 slices	12 slices	16 slices
bkg	MRF	22.65	23.14	21.72	17.78
	NNB	21.74	21.87	20.24	16.21
	LIN	23.30	23.83	21.56	18.94

inc	MRF	11.14	11.34	10.72	10.47
	NNB	9.03	9.21	8.51	8.22
	LIN	9.27	9.77	10.18	9.55

Signal to noise ratios (SNR) in dB calculated from parallelepiped shaped ROIs in the background and inclusion. Compared to NNB, the MRF reconstruction has a higher SNR in all cases, and better SNR than LIN in the inclusion. (MRF = Markov random field algorithm, NNB = nearest neighbors interpolation, LIN = linear interpolation, bkg = background, inc = inclusion).