End to End Multitask Joint Learning Model for Osteoporosis Classification in CT Images

Table 1

Comparison of joint framework performance at different learning rates.

Learning rate		Accuracy	Sensitivity	Specificity	F1-score
0.1	Normal	0.876	0.910	0.813	0.862
	Osteopenia	0.816	0.818	0.810	0.865
	Osteoporosis	0.935	0.951	0.823	0.963
0.01	Normal	0.936	0.955	0.923	0.925
	Osteopenia	0.898	0.905	0.880	0.927
	Osteoporosis	0.962	0.983	0.829	0.978
0.001	Normal	0.971	0.964	0.976	0.964
	Osteopenia	0.933	0.970	0.836	0.954
	Osteoporosis	0.957	0.962	0.922	0.975
0.0001	Normal	0.880	0.964	0.825	0.866
	Osteopenia	0.811	0.848	0.716	0.866
	Osteoporosis	0.921	0.921	0.922	0.953
0.00001	Normal	0.900	0.941	0.873	0.882
	Osteopenia	0.864	0.868	0.853	0.902
	Osteoporosis	0.964	0.981	0.843	0.980

The bold value indicates that this is the best model results.