小惠珠哦

这个屌丝很懒，什么也没留下！

热门标签

【Mamba】医学图像分割的Mamba评估指标参考（更新中...）_基于mamba的图片查重

作者：小惠珠哦 | 2024-07-18 16:28:37

踩

基于mamba的图片查重

医学图像分割的Mamba评估指标参考

All Models in
All Papers in

All Models in

代码汇总：https://github.com/Event-AHU/Mamba_State_Space_Model_Paper_List

VM-UNet&VM-UNetV2

Dataset	Model	mIoU(%)↑	DSC(%)↑	Acc(%)↑	Spe(%)↑	Sen(%)↑
ISIC17	UNet	76.98	86.99	95.65	97.43	86.82
ISIC17	UTNetV2	77.35	87.23	95.84	98.05	84.85
ISIC17	TransFuse	79.21	88.40	96.17	97.98	87.14
ISIC17	MALUNet	78.78	88.13	96.18	98.47	84.78
ISIC17	UNetV2	82.18	90.22	96.78	98.40	88.71
ISIC17	VM-UNet	80.23	89.03	96.29	97.58	89.90
ISIC17	VM-UNetV2	82.34	90.31	96.70	97.67	91.89

Dataset	Model	mIoU(%)↑	DSC(%)↑	Acc(%)↑	Spe(%)↑	Sen(%)↑
ISIC18	UNet	77.86	87.55	94.05	96.69	85.86
ISIC18	UNet++	78.31	87.83	94.02	95.75	88.65
ISIC18	Att-UNet	78.43	87.91	94.13	96.23	87.60
ISIC18	UTNetV2	78.97	88.25	94.32	96.48	87.60
ISIC18	SANet	79.52	88.59	94.39	95.97	89.46
ISIC18	TransFuse	80.63	89.27	94.66	95.74	91.28
ISIC18	MALUNet	80.25	89.04	94.62	96.19	89.74
ISIC18	UNetV2	80.71	89.32	94.86	96.94	88.34
ISIC18	VM-UNet	81.35	89.71	94.91	96.13	91.12
ISIC18	VM-UNetV2	81.37	89.73	95.06	97.13	88.60

Dataset	Model	DSC↑	HD95↓	Aorta	Gallbladder	Kidney(L)	Kidney®	Liver	Pancreas	Spleen	Stomach
Synapse	V-Net [25]	68.81	-	75.34	51.87	77.10	80.75	87.84	40.05	80.56	56.98
Synapse	DARR [4]	69.77	-	74.74	53.77	72.31	73.24	94.08	54.18	89.90	45.96
Synapse	R50 U-Net [10]	74.68	36.87	87.47	66.36	80.60	78.19	93.74	56.90	85.87	74.16
Synapse	UNet [27]	76.85	39.70	89.07	69.72	77.77	68.60	93.43	53.98	86.67	75.58
Synapse	R50 Att-UNet [10]	75.57	36.97	55.92	63.91	79.20	72.71	93.56	49.37	87.19	74.95
Synapse	Att-UNet [26]	77.77	36.02	89.55	68.88	77.98	71.11	93.57	58.04	87.30	75.75
Synapse	R50 ViT [10]	71.29	32.87	73.73	55.13	75.80	72.20	91.51	45.99	81.99	73.95
Synapse	TransUnet [10]	77.48	31.69	87.23	63.13	81.87	77.02	94.08	55.86	85.08	75.62
Synapse	TransNorm [5]	78.40	30.25	86.23	65.10	82.18	78.63	94.22	55.34	89.50	76.01
Synapse	Swin U-Net [9]	79.13	21.55	85.47	66.53	83.28	79.61	94.29	56.58	90.66	76.60
Synapse	TransDeepLab [6]	80.16	21.25	86.04	69.16	84.08	79.88	93.53	61.19	89.00	78.40
Synapse	UCTransNet [32]	78.23	26.75	-	-	-	-	-	-	-	-
Synapse	MT-UNet [33]	78.59	26.59	87.92	64.99	81.47	77.29	93.06	59.46	87.75	76.81
Synapse	MEW-UNet [30]	78.92	16.44	86.68	65.32	82.87	80.02	93.63	58.36	90.19	74.26
Synapse	VM-UNet	81.08	19.21	86.40	69.41	86.16	82.76	94.17	58.80	89.51	81.40

Dataset	Model	mIoU(%)↑	DSC(%)↑	Acc(%)↑	Spe(%)↑	Sen(%)↑
Kvasir-SEG	UNetV2	84.00	91.30	97.47	99.08	88.39
Kvasir-SEG	VMUnet	80.32	89.09	96.80	98.49	87.21
Kvasir-SEG	VMUnetV2	84.15	91.34	97.52	99.25	87.71
ClinicDB	UNetV2	83.85	91.21	98.59	99.16	91.99
ClinicDB	VMUnet	81.95	90.08	98.42	99.18	89.73
ClinicDB	VMUnetV2	89.31	94.35	99.09	99.38	95.64
ColonDB	UNetV2	57.29	72.85	96.19	98.43	68.46
ColonDB	VMUnet	55.28	71.20	96.02	98.45	65.89
ColonDB	VMUnetV2	60.98	75.76	96.54	98.46	72.68
ETIS	UNetV2	71.90	83.65	98.35	98.61	92.96
ETIS	VMUnet	66.41	79.81	98.26	99.33	75.79
ETIS	VMUnetV2	72.29	83.92	98.47	98.96	88.07
CVC-300	UNetV2	82.86	90.63	99.34	99.54	93.82
CVC-300	VMUnet	79.55	88.61	99.20	99.44	92.46
CVC-300	VMUnetV2	89.31	94.35	99.08	99.38	95.6

Model	Input size	Params(M)↓	FLOPs(G)↓	FPS↑
UNetV2	(3, 256, 256)	25.15	5.40	32.74
VM-Unet	(3, 256, 256)	34.62	7.56	20.612
VM-UnetV2	(3, 256, 256)	17.91	4.40	32.58

参考：
https://github.com/JCruan519/VM-UNet
https://github.com/nobodyplayer1/VM-UNetV2

LightM-UNet

Dataset	Models	Params(M)	GFLOPs	Liver (DSC)	Liver (mIoU)	Tumor (DSC)	Tumor (mIoU)	Average (DSC)	Average (mIoU)
LiTS	nnU-Net [8]	88.62	7,240.26	95.77	91.94	68.45	56.34	82.11	74.13
LiTS	SegResNet [15]	18.79	1,158.30	96.11	92.56	71.22	59.76	83.67	76.16
LiTS	UNETR [7]	92.62	1,891.35	94.06	88.95	48.58	37.01	71.32	62.98
LiTS	SwinUNETR [6]	61.99	1,570.32	95.24	91.07	66.67	55.09	80.95	73.08
LiTS	U-Mamba [14]	173.53	18,057.20	95.94	92.33	70.05	58.42	83.00	75.37
LiTS	LightM-UNet	1.87	457.62	96.31	92.92	72.86	62.05	84.58	77.48

Dataset	Models	Params(M)	GFLOPs	DSC(%)	mIoU(%)
Montgomery&Shenzhen	nnU-Net [8]	126.56	5,594.98	96.13	92.66
Montgomery&Shenzhen	SegResNet [15]	6.29	748.96	95.97	92.36
Montgomery&Shenzhen	UNETR [7]	87.51	1,267.53	95.36	91.26
Montgomery&Shenzhen	SwinUNETR [6]	25.12	909.26	95.37	91.31
Montgomery&Shenzhen	U-Mamba [14]	244.10	12,521.27	95.89	92.23
Montgomery&Shenzhen	LightM-UNet	1.09	267.19	96.17	92.74

参考：
https://github.com/MrBlankness/LightM-UNet

UltraLight-VM-UNet

Dataset	Methods	Parameters (Millions)	GFLOPs	DSC	SE	SP	ACC
ISIC17	U-Net	2.009	3.224	0.8989	0.8793	0.9812	0.9613
ISIC17	SCR-Net	0.801	1.567	0.8898	0.8497	0.9853	0.9588
ISIC17	ATTENTION SWIN U-NET	46.910	14.181	0.8859	0.8492	0.9847	0.9591
ISIC17	$C^2SDG$	22.001	7.972	0.8938	0.8859	0.9765	0.9588
ISIC17	VM-UNet	27.427	4.112	0.9070	0.8837	0.9842	0.9645
ISIC17	MALUNet	0.175	0.083	0.8896	0.8824	0.9762	0.9583
ISIC17	LightM-UNet	0.403	0.391	0.9080	0.8839	0.9846	0.9649
ISIC17	UltraLight VM-UNet	0.049	0.060	0.9091	0.9053	0.9790	0.9646

Dataset	Methods	Parameters (Millions)	GFLOPs	DSC	SE	SP	ACC
ISIC18	U-Net	2.009	3.224	0.8851	0.8735	0.9744	0.9539
ISIC18	SCR-Net	0.801	1.567	0.8886	0.8892	0.9714	0.9547
ISIC18	ATTENTION SWIN U-NET	46.910	14.181	0.8540	0.8057	0.9826	0.9480
ISIC18	$C^2SDG$	22.001	7.972	0.8806	0.8970	0.9643	0.9506
ISIC18	VM-UNet	27.427	4.112	0.8891	0.8809	0.9743	0.9554
ISIC18	MALUNet	0.175	0.083	0.8931	0.8890	0.9725	0.9548
ISIC18	LightM-UNet	0.403	0.391	0.8898	0.8829	0.9765	0.9555
ISIC18	UltraLight VM-UNet	0.049	0.060	0.8940	0.8680	0.9781	0.9558

Dataset	Methods	Parameters (Millions)	GFLOPs	DSC	SE	SP	ACC
$PH^2$	U-Net	2.009	3.224	0.9060	0.9255	0.9440	0.9381
$PH^2$	SCR-Net	0.801	1.567	0.8989	0.9114	0.9446	0.9339
$PH^2$	ATTENTION SWIN U-NET	46.910	14.181	0.8850	0.8886	0.9363	0.9213
$PH^2$	$C^2SDG$	22.001	7.972	0.9030	0.9137	0.9476	0.9367
$PH^2$	VM-UNet	27.427	4.112	0.9033	0.9131	0.9483	0.9369
$PH^2$	MALUNet	0.175	0.083	0.8865	0.8922	0.9425	0.9263
$PH^2$	LightM-UNet	0.403	0.391	0.9156	0.9129	0.9613	0.9457
$PH^2$	UltraLight VM-UNet	0.049	0.060	0.9265	0.9345	0.9606	0.9521

参考：
https://github.com/wurenkai/UltraLight-VM-UNet

H-vmunet

Dataset	Methods	DSC	SE	SP	ACC
ISIC17	U-Net	0.8159	0.8172	0.9680	0.9164
ISIC17	U-Net v2	0.9149	0.9052	0.9821	0.9670
ISIC17	Att U-Net	0.8082	0.7998	0.9776	0.9145
ISIC17	ATTENTION SWIN U-Net	0.8859	0.8492	0.9847	0.9591
ISIC17	TransNorm	0.8933	0.8535	0.9859	0.9582
ISIC17	MALUNet	0.8896	0.8824	0.9762	0.9583
ISIC17	MSNet	0.9067	0.8771	0.9860	0.9647
ISIC17	SCR-Net	0.8898	0.8497	0.9853	0.9588
ISIC17	META-Unet	0.9068	0.8801	0.9836	0.9639
ISIC17	$C^2SDG$	0.8938	0.8859	0.9765	0.9588
ISIC17	MHorUNet	0.9132	0.8974	0.9834	0.9666
ISIC17	VM-UNet	0.9070	0.8837	0.9842	0.9645
ISIC17	H-vmunet (VSS)	0.9068	0.8897	0.9823	0.9642
ISIC17	H-vmunet (H-VSS)	0.9172	0.9056	0.9831	0.9680

Dataset	Methods	DSC	SE	SP	ACC
Spleen	U-Net	0.9441	0.9604	0.9989	0.9983
Spleen	U-Net v2	0.9383	0.9244	0.9990	0.9982
Spleen	Att U-Net	0.9321	0.9350	0.9989	0.9979
Spleen	ATTENTION SWIN U-Net	0.7829	0.6662	0.9994	0.9945
Spleen	TransNorm	0.8618	0.7938	0.9992	0.9962
Spleen	MALUNet	0.9310	0.9305	0.9989	0.9979
Spleen	MSNet	0.9521	0.9425	0.9994	0.9986
Spleen	SCR-Net	0.9181	0.9122	0.9988	0.9976
Spleen	META-Unet	0.9233	0.8921	0.9993	0.9978
Spleen	$C^2SDG$	0.9354	0.9263	0.9991	0.9981
Spleen	MHorUNet	0.9424	0.9508	0.9989	0.9982
Spleen	VM-UNet	0.9418	0.9429	0.9991	0.9982
Spleen	H-vmunet (VSS)	0.9403	0.9330	0.9992	0.9982
Spleen	H-vmunet (H-VSS)	0.9571	0.9642	0.9992	0.9987

Dataset	Methods	DSC	SE	SP	ACC
CVC-ClinicDB	U-Net	0.9039	0.8926	0.9914	0.9821
CVC-ClinicDB	U-Net v2	0.9055	0.8861	0.9930	0.9825
CVC-ClinicDB	Att U-Net	0.8697	0.8474	0.9894	0.9760
CVC-ClinicDB	ATTENTION SWIN U-Net	0.7526	0.6503	0.9943	0.9631
CVC-ClinicDB	TransNorm	0.8845	0.8634	0.9918	0.9801
CVC-ClinicDB	MALUNet	0.8562	0.8475	0.9862	0.9731
CVC-ClinicDB	MSNet	0.9050	0.8720	0.9932	0.9832
CVC-ClinicDB	SCR-Net	0.8951	0.8701	0.9922	0.9807
CVC-ClinicDB	META-Unet	0.8975	0.8768	0.9919	0.9811
CVC-ClinicDB	$C^2SDG$	0.8967	0.8724	0.9923	0.9810
CVC-ClinicDB	MHorUNet	0.8930	0.8803	0.9904	0.9801
CVC-ClinicDB	VM-UNet	0.8524	0.8370	0.9867	0.9726
CVC-ClinicDB	H-vmunet (VSS)	0.8984	0.8768	0.9921	0.9813
CVC-ClinicDB	H-vmunet (H-VSS)	0.9087	0.8803	0.9940	0.9833

参考：
https://github.com/wurenkai/H-vmunet

U-Mamba

Methods	Organs in Abdomen CT 3D DSC	Organs in Abdomen CT 3D NSD	Organs in Abdomen MRI 3D DSC	Organs in Abdomen MRI 3D NSD
nnU-Net	0.8615±0.0790	0.8972±0.0824	0.8309±0.0769	0.8996±0.0729
SegResNet	0.7927±0.1162	0.8257±0.1194	0.8146±0.0959	0.8841±0.0917
UNETR	0.6824±0.1506	0.7004±0.1577	0.6867±0.1488	0.7440±0.1627
SwinUNETR	0.7594±0.1095	0.7663±0.1190	0.7565±0.1394	0.8218±0.1409
U-Mamba_Bot	0.8683±0.0808	0.9049±0.0821	0.8453±0.0673	0.9121±0.0634
U-Mamba_Enc	0.8638±0.0908	0.8980±0.0921	0.8501±0.0732	0.9171±0.0689

Methods	Organs in Abdomen CT 2D DSC	Organs in Abdomen CT 2D NSD	Organs in Abdomen MRI 2D DSC	Organs in Abdomen MRI 2D NSD
nnU-Net	0.7450±0.1117	0.8153±0.1145	0.6264±0.3024	0.6412±0.3074
SegResNet	0.7317±0.1379	0.8034±0.1386	0.5820±0.3268	0.5968±0.3303
UNETR	0.5747±0.1672	0.6309±0.1858	0.5017±0.3201	0.5168±0.3235
SwinUNETR	0.7028±0.1348	0.7669±0.1442	0.5528±0.3089	0.5683±0.3123
U-Mamba_Bot	0.7588±0.1051	0.8285±0.1074	0.6540±0.3008	0.6692±0.3050
U-Mamba_Enc	0.7625±0.1082	0.8327±0.1087	0.6303±0.3067	0.6451±0.3104

参考：
https://github.com/bowang-lab/U-Mamba

LMa-UNet

Methods	Organs in Abdomen CT 3D DSC	Organs in Abdomen CT 3D NSD	Organs in Abdomen MRI 2D DSC	Organs in Abdomen MRI 2D NSD
SegResNet	79.27	82.57	73.17	80.34
nnUNet	86.15	89.72	74.50	81.53
UNETR	68.24	70.04	57.47	63.09
SwinUNETR	75.94	76.63	70.28	76.69
nnFormer	78.34	81.45	72.79	79.63
U-Mamba	86.38	89.80	76.25	83.27
LMa-UNet (Ours)	86.82	90.02	77.35	83.80

参考：
https://github.com/wjh892521292/LMa-UNet

Mamba-UNet

Dataset	Methods	Dice ↑	IoU ↑	Acc ↑	Pre ↑	Sen ↑	Spe ↑	HD ↓	ASD ↓
ACDC MRI Cardiac	UNet	0.9248	0.8645	0.9969	0.9157	0.9364	0.9883	2.7655	0.8180
ACDC MRI Cardiac	AttentionUNet	0.9249	0.8647	0.9970	0.9239	0.9260	0.9858	3.4156	0.9765
ACDC MRI Cardiac	TransUNet	0.9196	0.8561	0.9968	0.9187	0.9207	0.9846	2.7742	0.8324
ACDC MRI Cardiac	Swin-UNet	0.9188	0.8545	0.9968	0.9151	0.9231	0.9857	3.1817	0.9932
ACDC MRI Cardiac	Mamba-UNet	0.9281	0.8698	0.9972	0.9275	0.9289	0.9859	2.4645	0.7677

Dataset	Methods	Dice ↑	IoU ↑	Acc ↑	Pre ↑	Sen ↑	Spe ↑	HD ↓	ASD ↓
Synapse CT abdominal	UNet	0.6299	0.5198	0.9969	0.6224	0.6541	0.9889	37.7342	10.0725
Synapse CT abdominal	AttentionUNet	0.6069	0.4964	0.9962	0.5959	0.6394	0.9890	74.2449	25.8229
Synapse CT abdominal	TransUNet	0.6092	0.5017	0.9970	0.6027	0.6318	0.9858	33.8126	10.8979
Synapse CT abdominal	Swin-UNet	0.6178	0.5121	0.9972	0.6126	0.6357	0.9859	30.5414	9.2504
Synapse CT abdominal	Mamba-UNet	0.6429	0.5405	0.9975	0.6452	0.6603	0.9890	24.4725	6.4717

参考：
https://github.com/ziyangwang007/Mamba-UNet

SegMamba

Dataset	Methods	Dice ↑ （WT）	HD95 ↓ （WT）	Dice ↑ （TC）	HD95 ↓ （TC）	Dice ↑（ET）	HD95 ↓ （ET）	Dice ↑ (Avg)	HD95 ↓ (Avg)
BraTS2023	SegresNet	92.02	4.07	89.10	4.08	83.66	3.88	88.26	4.01
BraTS2023	UX-Net	93.13	4.56	90.03	5.68	85.91	4.19	89.69	4.81
BraTS2023	MedNeXt	92.41	4.98	87.75	4.67	83.96	4.51	88.04	4.72
BraTS2023	UNETR	92.19	6.17	86.39	5.29	84.48	5.03	87.68	5.49
BraTS2023	SwinUNETR	92.71	5.22	87.79	4.42	84.21	4.48	88.23	4.70
BraTS2023	SwinUNETR-V2	93.35	5.01	89.65	4.41	85.17	4.41	89.39	4.51
BraTS2023	SegMamba	93.61	3.37	92.65	3.85	87.71	3.48	91.32	3.56

Dataset	Dataset	IoU ↑ ( Airway Tree)	DLR ↑ ( Airway Tree)	DBR ↑ ( Airway Tree)
AIIB2023	SegresNet	87.49	65.07	53.91
AIIB2023	UX-Net	87.55	65.56	54.04
AIIB2023	MedNeXt	85.81	57.43	47.34
AIIB2023	UNETR	83.22	48.03	38.73
AIIB2023	SwinUNETR	87.11	63.31	52.15
AIIB2023	SwinUNETR-V2	87.51	64.68	53.19
AIIB2023	SegMamba	88.59	70.21	61.33

Dataset	Methods	Dice ↑	HD95 ↓
CRC-500	SegresNet	46.10	34.97
CRC-500	UX-Net	45.73	49.73
CRC-500	MedNeXt	35.93	52.54
CRC-500	UNETR	33.70	61.51
CRC-500	SwinUNETR	38.36	55.05
CRC-500	SwinUNETR-V2	41.76	58.05
CRC-500	SegMamba	48.02	30.89

参考：
https://github.com/ge-xing/SegMamba

nnMamba

Dataset	Methods	WT (Dice)	TC (Dice)	ET (Dice)	Average (Dice)	WT (HD95)	TC (HD95)	ET (HD95)	Average (HD95)
BraTS 2023	UniMiss[36]	93.48	90.06	84.40	89.31	4.55	6.80	13.70	8.38
BraTS 2023	DIT[37]	93.49	90.22	84.38	89.36	4.21	5.27	13.64	7.71
BraTS 2023	UNETR[21]	93.33	89.89	85.19	89.47	4.76	7.27	12.78	8.27
BraTS 2023	nnUNet[5]	93.31	90.24	85.18	89.58	4.49	4.95	11.91	7.12
BraTS 2023	nnMamba	93.46	90.74	85.72	89.97	4.18	5.12	10.31	6.53

Dataset	Methods	Parameter (M)	Flops (G)	mDice （CT-Test）	mNSD （CT-Test）	mDice（MRI-Test）	mNSD（MRI-Test）
AMOS2022	nnUNet[5]	31.18	680.31	89.04	78.32	67.63	59.02
AMOS2022	VNet[38]	45.65	849.96	82.92	67.56	65.64	57.37
AMOS2022	CoTr [39]	41.87	668.15	80.86	66.31	60.49	51.18
AMOS2022	nnFormer[22]	150.14	425.78	85.61	72.48	62.92	54.06
AMOS2022	UNETR[21]	93.02	177.51	79.43	60.84	57.91	47.25
AMOS2022	SwinUNetr[40]	62.83	668.15	86.32	73.83	57.50	47.04
AMOS2022	nnMamba	15.55	141.14	89.63	79.73	73.98	65.13

分类：

Dataset	Methods	ACC （NC VS AD）	F1 （NC VS AD）	AUC （NC VS AD）	ACC （sMCI VS pMCI）	F1 （sMCI VS pMCI）	AUC （sMCI VS pMCI）
ANDI	ResNet	88.40±3.41	88.00±2.81	94.93±0.72	67.96±1.50	52.14±1.51	74.94±2.18
ANDI	DenseNet	87.95±0.70	86.93±0.87	94.86±0.40	73.12±3.10	53.30±2.99	76.31±3.09
ANDI	ViT	88.85±1.17	87.66±1.72	94.12±1.29	67.16±3.16	51.68±5.72	75.08±6.88
ANDI	CRATE	84.69±2.53	82.66±3.47	91.42±1.43	70.63±2.60	53.41±2.53	76.06±2.98
ANDI	nnMamba	89.41±0.85	88.68±0.77	95.81±0.59	75.79±1.79	56.55±2.37	76.84±0.84

参考：
https://github.com/lhaof/nnMamba

T-Mamba

Method	FLOPs(G)↓	Params(M)↓	HD(mm)↓	ASSD(mm)↓	IoU(%)↑	SO(%)↑	DSC(%)↑
UNet3D	2223.03	16.32	113.79	22.40	70.62	70.72	36.67
DenseVNet	23.73	0.87	8.21	1.14	84.57	94.88	91.15
AttentionUNet3D	1804	2720.79	94.48	147.10	61.10	52.52	42.49
DenseVoxelNet	402.32	1.78	41.18	3.88	81.51	92.50	89.58
MultiResUNet3D	1505.38	17.93	74.06	8.17	76.19	81.70	65.45
UNETR	229.19	93.08	107.89	17.95	74.30	73.14	81.84
SwinUNETR	912.35	62.19	82.71	7.50	83.10	86.80	89.74
TransBTS	306.80	33.15	29.03	4.10	82.94	90.68	39.32
nnFormer	583.49	149.25	51.28	5.08	83.54	90.89	90.66
3D UX-Net	1754.79	53.01	108.52	19.69	75.40	73.48	84.89
PMFSNet3D	15.14	0.63	5.57	0.79	84.68	95.10	91.30
T-Mamba (Ours)	-	-	1.18	0.42	88.31	97.53	93.60

参考：
https://github.com/isbrycee/T-Mamba

ViM-UNet

表格无

参考：
https://github.com/constantinpape/torch-em/tree/main/torch_em/model

TM-UNet

Model	Dataset	mIoU(%)↑	DSC(%)↑	Acc(%)↑	Spe(%)↑	Sen(%)↑
UNet	ISIC17	76.98	86.99	95.65	97.43	86.82
UTNetV2	ISIC17	77.35	87.23	95.84	98.05	84.85
TransFuse	ISIC17	79.21	88.40	96.17	97.98	87.14
MALUNet	ISIC17	78.78	88.13	96.18	98.47	84.78
VM-UNet	ISIC17	79.82	88.52	96.34	97.96	87.65
VM-UNetV2	ISIC17	79.73	88.14	96.09	97.78	87.89
TM-UNet	ISIC17	80.51	89.20	96.46	98.28	87.37
UNet	ISIC18	77.86	87.55	94.05	96.69	85.86
Unet++	ISIC18	78.31	87.83	94.02	95.75	88.65
Att-Unet	ISIC18	78.43	87.91	94.13	96.23	87.60
UTNetV2	ISIC18	78.97	88.25	94.32	96.48	87.60
SANet	ISIC18	79.52	88.59	94.39	95.97	89.46
TransFuse	ISIC18	80.63	89.27	94.66	95.74	91.28
MALUNet	ISIC18	80.25	89.04	94.62	96.19	89.74
UNetV2	ISIC18	80.71	89.32	94.86	96.94	88.34
VM-Unet	ISIC18	80.21	88.69	94.57	96.15	90.32
VM-UNetV2	ISIC18	80.45	88.73	94.76	96.51	88.57
TM-UNet	ISIC18	81.55	89.84	95.08	96.68	89.98

Dataset	Model	mIoU(%)↑	DSC(%)↑	Acc(%)↑	Spe(%)↑	Sen(%)↑
VM-UNet	Kvasir-Instrument	88.67	93.99	98.90	99.55	92.51
VM-UNetV2	Kvasir-Instrument	87.30	93.22	98.75	99.41	92.33
TM-UNet	Kvasir-Instrument	89.57	94.50	98.97	99.44	94.48
VM-UNet	Kvasir-Instrument	78.66	88.05	96.61	98.96	83.27
VM-UNetV2	Kvasir-SEG	76.99	87.00	96.12	97.81	86.55
TM-UNet	Kvasir-SEG	77.83	87.53	96.37	98.37	85.02
VM-UNet	Kvasir-SEG	76.53	86.71	97.92	99.05	84.89
VM-UNetV2	ClinicDB	83.80	91.19	98.59	99.22	91.35
TM-UNet	ClinicDB	83.82	91.20	98.59	99.25	91.10
VM-UNet	ClinicDB	51.17	67.70	95.55	98.19	62.67
VM-UNetV2	ColonDB	51.89	68.32	95.31	97.52	67.85
TM-UNet	ColonDB	57.23	72.80	96.26	98.58	98.37
VM-UNet	ColonDB	68.57	81.35	98.68	99.12	85.89
VM-UNetV2	CVC-300	72.30	83.92	98.81	99.03	92.51
TM-UNet	CVC-300	77.79	87.51	99.11	99.32	92.94

参考：
无代码
论文名：Rotate to Scan: UNet-like Mamba with Triplet SSM Module for Medical Image Segmentation

All Papers in

文章汇总：https://github.com/Event-AHU/Mamba_State_Space_Model_Paper_List

声明：本文内容由网友自发贡献，不代表【wpsshop博客】立场，版权归原作者所有，本站不承担相应法律责任。如您发现有侵权的内容，请联系我们。转载请注明出处：https://www.wpsshop.cn/w/小惠珠哦/article/detail/846678