Automatic Global Level Set Approach for Lumbar Vertebrae CT Image Segmentation

Table 1

Average DSC (%), MCR(%), MAD (pixel), HD (pixel), and RT (second) with standard deviation of various segmentation results with different levels of noise.

Input images	Criteria	C-V [15]	Lim [21]	Khadidos [30]	Liu [31]	AGLSA
Original images	DSC(%)
	MCR(%)
	MAD(P)
	HD (P)
	RT (S)
Images with 1% salt and pepper noise	DSC(%)
	MCR(%)
	MAD(P)
	HD (P)
	RT (S)
Images with 3% salt and pepper noise	DSC(%)
	MCR(%)
	MAD(P)
	HD (P)
	RT (S)
Images with 5% salt and pepper noise	DSC(%)
	MCR(%)
	MAD(P)
	HD (P)
	RT (S)
Images with 7% salt and pepper noise	DSC(%)
	MCR(%)
	MAD(P)
	HD (P)
	RT (S)