Table 5. Item-wise correlation between reporting status and online publication year.

CLAIM items1	Pre- and post-publication of CLAIM			Post-publication of CLAIM
	rho	P	flag2	rho	P	flag2
Item#1 (AI methodology and technology type in title)	−0.097	0.046	*	−0.074	0.281	-
Item#2 (Structured study summary)	0.034	0.491	-	0.022	0.748	-
Item#3 (Background and clinical role of AI)	−0.038	0.435	-	0.071	0.300	-
Item#4 (Study objectives and hypotheses)	−0.131	0.007	**	−0.162	0.018	*
Item#5 (Prospective or retrospective design)	0.092	0.060	-	0.017	0.806	-
Item#6 (Study goal, e.g., model creation, feasibility)	−0.098	0.045	*	0.046	0.502	-
Item#7 (Data sources)	0.024	0.626	-	−0.009	0.899	-
Item#8 (Eligibility criteria, e.g., inclusion/exclusion)	−0.055	0.261	-	−0.014	0.842	-
Item#9 (Data pre-processing)	−0.086	0.078	-	−0.217	0.001	**
Item#10 (Data subset selection, if applicable)	0.191	<0.001	***	0.225	<0.001	***
Item#11 (Definitions of data elements)	−0.220	<0.001	***	−0.057	0.405	-
Item#12 (De-identification methods)	0.099	0.042	*	0.068	0.322	-
Item#13 (Handling of missing data)	0.134	0.006	**	0.110	0.110	-
Item#14 (Ground truth definition)	−0.057	0.240	-	−0.137	0.045	*
Item#15 (Rationale for reference standard)	−0.205	<0.001	***	−0.069	0.312	-
Item#16 (Source and qualifications of annotators)	−0.078	0.111	-	−0.153	0.025	*
Item#17 (Annotation tools)	−0.244	<0.001	***	−0.092	0.180	-
Item#18 [Variability assessment (inter/intra-rater)]	−0.211	<0.001	***	−0.121	0.078	-
Item#19 (Sample size determination)	0.220	<0.001	***	0.250	<0.001	***
Item#20 (Data partitioning method)	0.140	0.004	**	−0.112	0.102	-
Item#21 (Partition level, e.g., image, patient)	0.345	<0.001	***	0.116	0.091	-
Item#22 [Model description (inputs, outputs, layers)]	0.036	0.462	-	−0.109	0.110	-
Item#23 (Software and frameworks used)	−0.127	0.009	**	−0.134	0.050	-
Item#24 (Model parameter initialization)	−0.124	0.011	*	−0.176	0.010	*
Item#25 (Training details, e.g., augmentation, hyperparameters)	0.141	0.004	**	−0.123	0.073	-
Item#26 (Final model selection)	−0.057	0.246	-	−0.117	0.088	-
Item#27 (Ensemble techniques, if applicable)	0.186	<0.001	***	0.167	0.014	*
Item#28 (Model performance metrics)	−0.076	0.119	-	−0.129	0.060	-
Item#29 (Statistical significance and uncertainty)	0.026	0.594	-	0.060	0.386	-
Item#30 (Robustness/sensitivity analysis)	0.022	0.656	-	0.015	0.831	-
Item#31 (Explainability methods, e.g., saliency maps)	0.222	<0.001	***	−0.002	0.982	-
Item#32 (External validation/testing)	0.009	0.846	-	0.098	0.152	-
Item#33 (Participant flow diagram)	0.356	<0.001	***	0.202	0.003	**
Item#34 (Demographic/clinical characteristics by partition)	0.195	<0.001	***	0.126	0.065	-
Item#35 (Performance metrics for optimal model)	−0.020	0.684	-	−0.097	0.158	-
Item#36 (Diagnostic accuracy estimates)	0.101	0.039	*	0.172	0.012	*
Item#37 (Failure analysis)	0.075	0.125	-	−0.009	0.892	-
Item#38 [Study limitations (bias, uncertainty, generalizability)]	0.173	<0.001	***	0.107	0.119	-
Item#39 (Practice implications and clinical role)	−0.270	<0.001	***	−0.298	<0.001	***
Item#40 (Registration number and registry name)	−0.141	0.004	**	−0.093	0.174	-
Item#41 (Study protocol access)	0.160	0.001	**	−0.075	0.275	-
Item#42 (Funding sources and funder roles)	0.207	<0.001	***	0.092	0.178	-

¹ Note that item names have been abbreviated; ²* P < 0.05; ** P < 0.01; *** P < 0.001. CLAIM, Checklist for Artificial Intelligence in Medical Imaging.