. Author manuscript; available in PMC: 2019 Sep 5.

Published in final edited form as: IEEE Trans Med Imaging. 2019 Jan 23;38(9):2151–2164. doi: 10.1109/TMI.2019.2894322

Table VI.

Comparison of clinical measures derived from SSLLN-LR+SR and manual segmentations on 20 pairs of low-resolution and high-resolution volumetric images from Dataset 2. SSLLN-LR+SR segmented 20 low-resolution volumes into high-resolution models, whilst manual segmentation was performed on 20 high-resolution cardiac volumes directly. The 4th column shows absolute difference between SSLLN-LR+SR and manual measures.

	SSLLN-LR+SR^†	Manual	† vs Manual	p-value
LVV (ml)	120.098±20.822	114.815±25.099	10.42±9.378	p ≈0.119
LVM (gram)	125.989±34.639	124.237±25.271	8.032±9.614	p ≈0.855
RVV (ml)	221.514±64.534	204.293±58.534	18.69±13.55	p ≈0.001
RVM (gram)	51.621±14.938	49.877±14.166	3.857±2.558	p ≈0.501