Prediction of knee osteoarthritis progression using radiological descriptors obtained from bone texture analysis and Siamese neural networks: data from OAI and MOST cohorts

Table 4 Results obtained from training on one cohort (OAI/MOST) and testing on another cohort (MOST/OAI)

Modality	Metric	Train. on OAI and validation on MOST		Train. on MOST and validation on OAI
Modality	Metric	NonQC	QC	NonQC	QC
CR	AUC	0.74 (0.68–0.79)	0.73 (0.67–0.79)	0.7 (0.67–0.74)	0.71 (0.68–0.75)
	DOR	6.82	5.95	3.42	4.04
	ACC	0.74	0.74	0.77	0.73
RG	AUC	0.64 (0.59–0.70)	0.63 (0.56–0.69)	0.69 (0.64–0.74)	0.59 (0.53–0.65)
	DOR	3.01	4.34	3.36	2.51
	ACC	0.62	0.65	0.74	0.78
CR & RG	AUC	0.73 (0.70–0.77)	0.73 (0.69–0.77)	0.71 (0.68–0.74)	0.71 (0.68–0.74)
	DOR	5.82	6.91	4.03	4.04
	ACC	0.7	0.72	0.83	0.82

ISSN: 1478-6362