EURASIP Journal on Advances in Signal Processing

Table 4 Experiment results based on abdominal scans under setting 1

From: A dynamic few-shot learning framework for medical image stream mining based on self-training

Method	Dataset-abdomen—MRI					Dataset-abdomen—CT
Method	Left kidney	Right kidney	Spleen	Liver	Mean	Left kidney	Right kidney	Spleen	Liver	Mean
SE-Net [37]	46.48	47.23	47.79	28.82	42.58	23.51	14.40	42.59	35.34	28.96
Vanilla PANet [38]	31.11	32.78	41.74	50.40	39.01	20.56	20.91	36.30	49.55	31.83
ALPNet [39,40,41]	44.73	48.66	49.81	63.14	51.58	29.12	30.23	41.01	64.70	41.26
DARCNN [13]	57.54	60.41	61.32	72.15	62.85	56.37	51.24	55.72	61.68	56.25
MRNet [42]	61.28	63.83	63.99	72.46	65.39	65.59	68.33	69.14	70.01	68.26
Ours	82.61	85.68	72.26	77.48	79.50	73.96	72.18	70.94	78.68	73.94

Back to article page