notesum.ai

Published at November 19

DiM: $f$-Divergence Minimization Guided Sharpness-Aware Optimization for Semi-supervised Medical Image Segmentation

cs.CV

cs.AI

Released Date: November 19, 2024

Authors: Bingli Wang¹, Houcheng Su², Nan Yin³, Mengzhu Wang⁴, Li Shen⁵

Aff.: ¹Tsinghua University (SZ); ²University of Macau; ³Zayed University of Artificial Intelligence, United Arab Emirates; ⁴Hebei University of Technology; ⁵Sun Yat-sen University

Arxiv: http://arxiv.org/abs/2411.12350v1

Task			Optic Cup / Disc Segmentation
Method		#L	DC $\uparrow$				DC $\uparrow$	JC $\uparrow$	HD $\downarrow$	ASD $\downarrow$
Method		#L	Domain 1	Domain 2	Domain 3	Domain 4	Avg.	Avg.	Avg.	Avg.
U-Net		20	59.54 / 73.89	71.28 / 74.23	50.87 / 64.29	35.61 / 63.30	61.63	52.65	48.28	28.86
UA-MT	MICCAI’19	20	59.35 / 78.46	63.08 / 74.45	35.24 / 47.73	36.18 / 55.43	56.24	47.00	48.64	31.35
FDA	CVPR’20	20	76.99 / 89.94	77.69 / 89.63	78.27 / 90.96	64.52 / 74.29	80.29	71.05	16.23	8.44
SIFA	TMI’20	20	50.67 / 75.30	64.44 / 80.69	61.67 / 83.77	55.07 / 70.67	67.78	54.77	20.16	10.93
FixMatch	NeurIPS’20	20	81.18 / 91.29	72.04 / 87.60	80.41 / 92.95	74.58 / 87.07	83.39	73.48	11.77	5.60
CPS	CVPR’21	20	64.53 / 86.25	70.26 / 86.97	42.92 / 54.94	36.98 / 46.70	61.19	52.69	34.44	26.79
CoraNet	TMI’21	20	61.64 / 87.32	65.56 / 87.05	66.12 / 83.54	49.01 / 77.73	72.25	60.50	20.52	10.44
UDA-VAE++	CVPR’22	20	55.01 / 80.76	68.87 / 85.94	63.23 / 84.92	68.42 / 80.89	73.51	61.40	17.60	9.86
SS-Net	MICCAI’22	20	59.42 / 78.15	67.32 / 85.05	45.69 / 69.91	38.76 / 61.13	63.18	53.49	44.90	25.73
BCP	CVPR’23	20	71.65 / 91.10	77.19 / 92.00	72.63 / 90.77	77.67 / 91.42	83.05	73.66	11.05	5.80
CauSSL	ICCV’23	20	63.38 / 80.60	67.52 / 80.72	49.53 / 63.88	39.43 / 49.43	61.81	51.80	41.25	23.94
MiDSS	CVPR’24	20	83.39 / 92.96	73.12 / 88.88	83.50 / 92.97	78.63 / 93.38	85.85	76.95	9.06	4.40
DiM		20	84.68 / 92.91	78.16 / 90.49	84.82 / 93.36	81.63 / 92.18	87.28	78.53	7.83	3.82
Upper bound		*	85.53 / 93.41	80.55 / 90.90	85.44 / 93.04	85.61 / 93.21	88.46	80.35	7.41	3.70