Table 2.

Regions to be segmented in complex situations of marked mismatch between radiological and nuclear medicine images, like Figs. 2B, C and D. Use a contrast-enhanced CT co-registered with CT of SPECT if necessary (see also Numerical example section in the Appendix)

Region number	Tissue region	VOI name	Segmentation aim and strategy
1	Whole liver	VOICT(whole liver)	CT-based, to quantify volume. Exclude cysts or necrosis from previous treatments.
2	Targeted lesions	VOICT(#1) VOICT(#2) VOICT(#3)	CT-based, including low-perfusion region to quantify volume. Derive respective counts from perfused lesion regions.
3	Non targeted lesion	VOICT (non-targeted lesions)	CT-based, including low-perfusion region
4	Perfused liver	VOISPECT(PL)	SPECT-based, to determine counts
5	Perfused liver	VOICT(PL)	To determine PL volume. To obtain this VOI, Crop 4 within 1
6	Perfused lesion regions	VOISPECT(#1) VOISPECT(#2) VOISPECT(#3)	SPECT-based (aided by CT), excluding low-perfusion region , to determine SPECT lesion volume and counts
7	Perfused non-tumoural liver	VOISPECT(PNTL)	To obtain the number of counts N. 4 minus 6: N{VOISPECT(PL)} − N{∑iVOISPECT(#i)}
8	Perfused non-tumoural liver	VOICT(PNTL)	To obtain volume. 5 minus 2: VOISPECT(PL) − ΣVOICT(#i), limited within 1
9	Non-tumoural whole liver (includes also non-perfused liver)	VOICT(NTL)	Non-tumoural whole liver volume is obtained by subtraction: Volume 1 minus (Volume 2 + Volume 3) Non-tumoural whole liver counts 4 subtracted of counts in 6 identical to 7 Record the non-tumoural liver-treated volume fraction Vf ${\mathit{V}}_{\mathit{f}}\mathit{=}\frac{{\mathbf{VOI}}_{\mathbf{CT}}\left(\mathbf{\text{PNTL}}\right)}{{\mathbf{VOI}}_{\mathbf{CT}}\left(\mathbf{NTL}\right)}$
10	Lung (in case of substantial LSF)	VOICT(Lung)	CT-based, to determine volume and counts.