The effects of various penalty parameter values in Q.Clear algorithm for rectal cancer detection on 18F-FDG images using a BGO-based PET/CT scanner: a phantom and clinical study

EJNMMI Physics

Table 2 The relative difference in SNR among distinct β values in Q.Clear reconstruction for hot lesions measuring 10 mm, 22 mm, and 37 mm in diameter, filled with LBRs of 2:1, 4:1, and 8:1 (Above)

LBR	Lesion size	Relative difference of SNR(%)
LBR	Lesion size	\(\Delta {\mathrm{SNR}}_{\left(200-100\right)}\)	\(\Delta {\mathrm{SNR}}_{(300-200)}\)	\(\Delta {\mathrm{SNR}}_{\left(400-300\right)}\)	\(\Delta {\mathrm{SNR}}_{\left(500-400\right)}\)	\(\Delta {\mathrm{SNR}}_{\left(500-100\right)}\)	\(\Delta {\mathrm{SNR}}_{\left(\mathrm{OSEM}-500\right)}\)
2:1	10 mm	26.5	16	11.4	8.8	49.1	− 4.9
	22 mm	24.9	12.8	6.9	3.9	42.8	− 0.9
	37 mm	18.3	8.4	4.7	2.9	30.8	1.2
	p	S	S	S	S	S	S
4:1	10 mm	26.1	14.8	9.5	7.3	45.1	− 10.2
	22 mm	24.6	12.2	6.8	6.9	40.5	− 7.3
	37 mm	21.7	10.6	6.4	4.7	37.4	− 3.1
	p	S	S	S	NS	S	S
8:1	10 mm	25.2	13.6	8.7	6	43.7	− 30.6
	22 mm	24.2	13.5	8.3	5.6	40.7	− 20.4
	37 mm	22.1	12.5	8	5.4	44.0	− 18.8
	p	S	S	S	NS	S	S
Relative difference of LE(%)
8:1	Lung insert	\(\Delta {\mathrm{LE}}_{\left(200-100\right)}\)	\(\Delta {\mathrm{LE}}_{\left(300-200\right)}\)	\(\Delta {\mathrm{LE}}_{(400-300)}\)	\(\Delta {\mathrm{LE}}_{\left(500-400\right)}\)	\(\Delta {\mathrm{LE}}_{\left(500-100\right)}\)	\(\Delta {\mathrm{LE}}_{\left(\mathrm{OSEM}-500\right)}\)
	Lung insert	− 5.3	− 4.1	− 3.8	− 3.2	− 13.1	54.5
	p	S	S	NS	NS	S	S

Additionally, the relative difference in lung residual error between OSEM and Q.Clear in the lung insert of the NEMA IEC phantom at the LBR of 8:1 is presented (Below). P values were deduced using a paired t test. ‘S’ denotes statistically significant differences (p value ≤ 0.05), whereas 'NS' indicates no significant variation (p value > 0.05).