Knowledge, Attitude and Practice of Radiologists Regarding Artificial Intelligence in Medical Imaging

Figures & data

Table 1 Participants’ Demographics and KA Score

Variables	N (%)	Knowledge score		Attitude score		Whether you actively understand the relevant knowledge and research progress of AI medical imaging
		Mean ± SD	P	Mean ± SD	P	Yes	P
Total	452	9.01±4.87		48.96±4.90		340 (75.22)
Gender			0.291		0.188		0.943
Male	316(69.91)	8.85±5.08		49.16±5.12		238 (70.00)
Female	136(30.09)	9.38±4.36		48.50±4.32		102 (30.00)
Age			0.002		0.006		0.001
20–35	125(27.65)	8.44±4.64		48.11±4.18		90 (26.47)
36–50	198(43.81)	8.53±5.03		48.80±5.16		138 (40.59)
>50	129(28.54)	10.31±4.63		50.04±4.45		112 (32.94)
Educational level			<0.001		0.750		0.095
Junior college	49(10.84)	7.78±5.49		49.41±5.21		33 (9.71)
Bachelor	269(59.51)	8.49±4.66		48.85±4.82		198 (58.24)
Master or above	134(29.65)	10.51±4.75		49.03±4.98		109 (32.06)
Hospital level			<0.001		0.007		0.201
Class I public hospital	20(4.42)	8.75±4.46		47.15±6.21		12 (3.53)
Class II public hospital	109(24.12)	7.30±4.76		48.40±4.58		82 (24.12)
Class III public hospital	301(66.59)	9.78±4.80		49.46±4.92		232 (68.24)
Private hospital	22(4.87)	7.18±4.39		46.64±3.54		14 (4.12)
Professional title			<0.001		<0.001		<0.001
None	16(3.54)	7.13±4.70		47.63±1.75		7 (2.06)
Junior	97(21.46)	8.95±4.79		48.80±4.80		73 (21.47)
Intermediate	92(20.35)	7.09±4.83		46.49±5.42		52 (15.29)
Associate senior	173(38.27)	9.28±4.82		49.95±4.67		143 (42.06)
Senior	74(16.37)	11.28±4.17		50.23±4.17		65 (19.12)
Working years in radiology			0.016		0.002		0.044
≤5	66(14.60)	8.21±4.44		47.59±4.12		46 (13.53)
(5,10]	70(15.49)	9.19±4.92		48.93±4.72		48 (14.12)
(10,20]	87(19.25)	8.48±5.76		47.91±6.10		60 (17.65)
(20,30]	156(34.51)	8.78±4.46		49.54±4.54		124 (36.47)
>30	73(16.15)	10.71±4.61		50.25±4.41		62 (18.24)
Work in the hospital that has AI image-assisted diagnosis technologies			<0.001		<0.001		0.001
Yes	361(79.87)	9.57±4.83		49.55±4.76		284 (83.53)
No	91(20.13)	6.81±4.44		46.64±4.76		56 (16.47)
Have ever used the AI diagnostic system			<0.001		<0.001		<0.001
Yes	374(82.74)	9.57±4.85		49.60±4.51		301 (88.53)
No	78(17.26)	6.33±4.03		45.90±5.53		39 (11.47)
Have attended training on the principles and operation of AI-assisted diagnosis			<0.001		<0.001		0.005
Yes	205(45.35)	10.74±5.10		50.61±4.38		167 (49.12)
No	247(54.65)	7.58±4.17		47.59±4.89		173 (50.88)
Have participated in AI-assisted diagnosis research			<0.001		0.002		<0.001
Yes	147(32.52)	12.05±4.27		49.99±4.85		135 (39.71)
No	305(67.48)	7.55±4.45		48.47±4.86		340 (60.29)

Table 2 Multivariate Analysis of Sufficient Knowledge

	OR (95% CI)	P
Educational level
Junior college	1.189 (0.485 2.911)	0.705
Bachelor	Ref.
Master or above	1.877 (1.086 3.245)	0.024
Level of hospital
Class I public hospital	1.552 (0.494 4.873)	0.452
Class II public hospital	0.729 (0.381 1.396)	0.341
Class III public hospital	Ref.
Private hospital	0.790 (0.233 3.245)	0.705
Professional title
None	0.563 (0.114 2.789)	0.482
Junior	Ref.
Intermediate	0.599 (0.216 1.662)	0.325
Associate senior	1.066 (0.346 3.284)	0.911
Senior	2.062 (0.564 7.539)	0.274
Working years in radiology
≤5	Ref.
(5,10]	3.481 (1.345 9.013)	0.010
(10,20]	3.439 (0.966 12.242)	0.057
(20,30]	2.637 (0.709 9.803)	0.148
>30	3.189 (0.772 13.172)	0.109
Work in the hospital that has AI image-assisted diagnosis
Yes	0.883 (0.402 1.938)	0.757
No	Ref.
Have ever used the AI diagnostic system
Yes	1.772 (0.758 4.141)	0.186
No	Ref.
Have attended training on the principles and operation of AI-assisted diagnosis
Yes	2.915 (1.776 4.786)	<0.001
No	Ref.
Have participated in AI-assisted diagnosis research
Yes	3.178 (1.881 5.371)	<0.001
No	Ref.

Table 3 Multivariate Analysis of Positive Attitude

	OR (95% CI)	P
Knowledge score	1.013 (0.967 1.062)	0.579
Age
20–35	Ref.
36–50	1.820 (0.714 4.643)	0.210
>50	2.835 (0.874 9.193)	0.083
Educational level
Junior college	2.139 (1.084 4.220)	0.028
Bachelor	Ref.
Master or above	1.028 (0.616 1.715)	0.917
Working years in radiology
≤5	Ref.
(5,10]	2.462 (1.013 5.983)	0.047
(10,20]	1.838 (0.607 5.561)	0.281
(20,30]	1.705 (0.545 5.341)	0.359
>30	1.682 (0.444 6.383)	0.444
Have ever used the AI diagnostic system
Yes	1.293 (0.726 2.304)	0.382
No	Ref.
Have attended training on the principles and operation of AI-assisted diagnosis
Yes	2.264 (1.430 3.586)	<0.001
No	Ref.

Figure 1 Distribution of practice dimension. (A) Who do you believe more when your own judgment is consistent with that of AI; (B) Who do you believe more when your own judgment is inconsistent with that of AI; (C) Whether you actively understand the relevant knowledge and research progress of AI medical imaging; (D) Understand the information channels related to AI-assisted diagnosis; (E) Factors influencing imaging physicians’ learning of AI-assisted diagnosis related knowledge. AI, artificial intelligence.

Table 4 Multivariate Analysis of Proactive Practice

	OR (95% CI)	P
Knowledge score	5.240 (2.391 11.486)	<0.001
Attitude score	1.509 (0.829 2.746)	0.179
Age
20–35
36–50	0.318 (0.081 1.250)	0.101
>50	0.886 (0.160 4.902)	0.890
Professional title
None	0.171 (0.042 0.704)	0.014
Junior	Ref.
Intermediate	1.791 (0.591 5.4240	0.303
Associate senior	4.267 (1.187 15.337)	0.026
Senior	4.342 (0.984 19.165)	0.053
Working years in radiology
≤5	Ref.
(5,10]	0.344 (0.121 0.974)	0.044
(10,20]	0.634 (0.135 2.979)	0.564
(20,30]	0.966 (0.195 4.772)	0.966
>30	0.491 (0.069 3.474)	0.476
Work in the hospital that has AI image-assisted diagnosis
Yes	0.698 (0.345 1.411)	0.317
No	Ref.
Have ever used the AI diagnostic system
Yes	3.643 (1.739 7.631)	0.001
No	Ref.
Have attended training on the principles and operation of AI-assisted diagnosis
Yes	0.743 (0.413 1.336)	0.321
No	Ref.
Have participated in AI-assisted diagnosis research
Yes	6.382 (2.949 13.812)	<0.001
No	Ref.

Table 5 Results of SEM

			Estimate	P
Attitude	<---	Knowledge	0.481	<0.001
Practice	<---	Attitude	0.135	<0.001
Practice	<---	Knowledge	0.412	<0.001

Figure 2 SEM for KAP.