MRI-based brain tumor segmentation using FPGA-accelerated neural network

Table 2 Performance of proposed method

Dataset	Method	DSC	Execution time	Energy Consumption
BraTS19	GPU method before quantization	0.873	0.78 s	96 W
	GPU method after quantization	0.871	0.64 s	72 W
	FPGA method	0.871	0.15 s	45 W
BraTS20	GPU method before quantization	0.885	0.78 s	97 W
	GPU method after quantization	0.882	0.65 s	74 W
	FPGA method	0.882	0.15 s	45 W