Table 1.

Data partitions used to train, validate, and test ResNet18 to predict radiographic hip osteoarthritis (rHOA). Model-1 and Model-2 were trained with unprocessed and downsampled CHECK X-ray images, respectively. Model-3 and Model-4 were trained on a combination of the CT-AP and X-ray images where the X-ray images were similar to the ones used in Model-1 and Model-2, respectively. Model-5 was trained solely on the CT-AP images. The overall images are the total of Cohort Hip and Cohort Knee (CHECK) X-ray and CT-AP images

Data partitions	CHECK		CT-AP		Combined		Overall (%)
	Model-1 and Model-2		Model-5		Model-3 and Model-4
	rHOA	no-rHOA	rHOA	no-rHOA	rHOA	no-rHOA
Training	1093	2019	29	17	1122	2036	3158 (55%)
Validation	597	1101	15	9	612	1110	1722 (30%)
Test	298	551	15	9	313	560	873 (15%)
Total	1988	3671	59	35	2047	3706	5753 (100%)