. Author manuscript; available in PMC: 2015 Feb 24.

Published in final edited form as: Med Image Comput Comput Assist Interv. 2013;16(0 1):259–266. doi: 10.1007/978-3-642-40811-3_33

Table 1.

Prediction and registration results. Prediction is based on the method in [3], and we use SSR to evaluate the prediction results. Here, MD denotes our proposed multimodal dictionary learning method and ST denotes the dictionary learning method in [3]. The registrations use Sum of Squared Differences (SSD) and mutual information (MI) similarity measures. We report the results of mean and standard deviation of the absolute error of corresponding landmarks in micron (0.069 micron = 1 pixel). The p-value is computed using a paired t-test.

	Metric	Method	mean	std	p-value
Prediction	SSR	MD	6.28 × 10⁴	3.61 × 10³
Prediction	SSR	ST	7.43 × 10⁴	4.72 × 10³

Registration	SSD	MD	0.760	0.124	0.0004
	SSD	ST	0.801	0.139	0.0004
	MI	MD	0.754	0.127	0.0005
	MI	ST	0.795	0.140	0.0005