Results | MRBrainS13

When teams submit their segmentation results, the evaluation results will be sent to the team contact person by e-mail and will be listed below.

Rank	Team name	Submission name	Submission date	Score	Sequences used	Duration
1	XMU_SmartDSP2	3D Spatial Weighted U-Net for Multi-modality Brain MRI Segmentation	29-08-18	49	T1; T1_IR; FLAIR	10 min
2	lrhs	Brain Tissue Segmentation based on Transfer Learning	24-03-21	53	T1; T1_IR; FLAIR	34 sec
3	Smartdsp3713	3D weighted U-shape fully convolutional network	26-07-19	59	T1; T1_IR; FLAIR	2 min
4	TailHot	Multi-modality aggregation network³	13-04-18	81	T1; T1_IR; FLAIR	13 sec
5	WTA2	3D Cascade convolutional architecture - Method 2²	23-05-18	101	T1; T1_IR; FLAIR	2 min
6	XMU SmartDSP	3D CNN with a Cross-modality Channel Attention Scheme³	17-08-18	106	T1; T1_IR; FLAIR	10 min
7	3DAPCNet	3D Anisotropic Pyramidal Convolution Network	24-01-21	122	T1; T1_IR; FLAIR	45 sec
8	XLab	3D Fully CNN with Multi-Modality Feature Fusion³	10-08-18	122	T1; T1_IR; FLAIR	2 min
9	LIVIA_ETS	HyperDenseNet²	06-02-18	123	T1; T1_IR; FLAIR	4 min
10	CU_DL2	"3D Deep Learning; voxnet2"	28-06-16	140	T1; T1_IR; FLAIR	2 min
11	CU_DL	"3D Deep Learning; voxnet1³"	16-06-16	149	T1; T1_IR; FLAIR	2 min
12	muya	Fusion Weighted Network³	04-07-19	153	T1; T1_IR; FLAIR	10 min
13	LRDE	Fully Convolutional Network	20-12-16	160	T1	2 sec
14	CVSS	CNN with Attention Mechanism³	12-03-19	163	T1; T1_IR; FLAIR	1 min
15	MSL-SKKU	Deep Convolutional Neural Network	19-06-17	168	T1; T1-IR; FLAIR	1.5 min
16	QL111111	Multi-modality Aggregation Network with Self-attention and Deep Feature Reconstruction²	18-03-19	192	T1; T1_IR; FLAIR	10 sec
17	HUST-CBIB	3D Reversible Spatial Attention Network³	04-01-21	204	T1; T1_IR; FLAIR	20 sec
18	MDGRU	Multi-Dimensional Gated Recurrent Units³	27-07-16	205	T1; T1_IR; FLAIR	2 min
19	FBI/LMB Freiburg	U-Net (3D)	01-05-16	207	T1-1mm; T1-IR; FLAIR	2 min
20	ASTRI_DL	U-Net inspired CNN model with 3D context information²	25-07-18	210	T1; T1_IR; FLAIR	7 min
21	PyraMiD-LSTM2	NOCC with rounds³	23-05-16	213	T1; T1-IR; FLAIR	2 min
22	THUity	Modified U-Net³	15-08-18	217	T1	40 sec
23	NIST	Fully CNN for Multi-Modality Brain Tissue Segmentation	14-08-18	225	T1; T1_IR; FLAIR	2 min
24	AOC	Atlas of Classifiers	24-12-17	244	T1	6 sec
25	IDSIA	PyraMiD-LSTM	05-06-15	249	T1; T1_IR; FLAIR	2 min
26	UDSC_DL	Multi-UDense-Net²	14-09-19	254	T1; T1_IR; FLAIR	4 min
27	STH	Hybrid ANN-based Auto-context method²	03-06-16	267	T1; T1-IR; FLAIR	5 min
28	nanand2	DeepLab-v3+ with pre-training on FreeSurfer automated labels³	16-08-18	268	T1	15 sec
29	holloww	2D Rotation-Driven Convolutions	30-07-20	276	T1; T1_IR; FLAIR	13 sec
30	ISI-Neonatology	Multi-stage voxel classification	31-05-14	278	T1	1.5 hours
31	UNC-IDEA	LINKS:Learning-based multi-source integration	09-02-15	280	T1; T1_IR; FLAIR	3 min
32	CNK	3D-Unet-deep supervision	05-01-21	302	T1; T1_IR; FLAIR	2 min
33	ASTRI_MSA	Revised VoxResNet	27-08-18	304	T1; T1_IR; FLAIR	1 min
34	BCH_CRL_IMAGINE	3D patch-wise DenseNet and Patch Fusion²	24-05-18	308	T1; T1-IR; FLAIR	2 min
35	LfB	Multi-modality Mix Network	17-01-19	315	T1; T1_IR; FLAIR	40 sec
36	medical_imaging_nitdgp	U-Net Inception and Morphological Gradient Channel	24-01-20	318	T1; FLAIR	30 sec
37	KSOM GHMF	ASeTs: MAP-Based with Manifold learning	13-05-14	320	T1; T1_IR; FLAIR	23 min
38	WTA	3D Cascade convolutional architecture - Method 1³	15-05-18	321	T1; T1_IR; FLAIR	5 min
39	MNAB2	Random Forests	21-02-14	325	T1; T1_IR; FLAIR	25 min
40	vicorob UdG T1_F	MSSEG using T1 + FLAIR (T1-IR skull)	14-01-16	328	T1; IR; FLAIR	2 min
41	VBM12	VBM12_r738 with WMHC=2	07-10-15	331	T1	6 min
42	BIGR2	Multi-Feature SVM Classification	26-09-13	345	T1; T1_IR; FLAIR	35 min
43	vicorob UdG T1	MSSEG using only T1 (T1-IR skull)	21-01-16	349	T1; IR	2 min
44	UofL BioImaging	MAP-Based Framework	26-09-13	394	T1	6 sec
45	mehran007	Tsallis-Entropy Segmentation through MRF	04-04-19	396	T1	6.5 min
46	Narsil	Segmentation using Ensemble Trees	26-09-13	404	T1; FLAIR	< 2 min
47	UB VPML Med	Multi-Atlas with Multiway Cut	26-09-13	409	T1; T1_IR; FLAIR	30 min
48	CMIV	Model-guided Level Sets and Skeletons	26-09-13	413	T1; FLAIR	3 min
49	drai	VGG based Fully Convolutional Neural Network	20-01-17	419	T1	2 sec
50	bigr_neuro	Automatically Trained kNN Classifier	26-09-13	430	T1; FLAIR	2 hours
51	VIBOT10 M-T1+F	Convolutional Neural Network	27-02-17	436	T1; FLAIR	3 min
52	SPM12_T1_F*	SPM12 with T1 and FLAIR sequence	31-08-15	442	T1; FLAIR	3 min
53	S2_QM2	Atlas-based segmentation and AdaBoost	28-07-14	445	T1; T1-IR; FLAIR	45 min
54	SJCE	Fuzzy Clustering	12-04-17	449	T1	<1 min
55	Robarts	Multi-Atlas with Hierarchical Max-Flow	26-09-13	451	T1-1mm; T1-IR	16 min
56	KM-MRF-MAS	MAP on Intensity, and Local and Multi-atlas Priors	28-10-16	452	T1	10 min
57	CSIM-lab_Mqe-MMRF	Hybrid Modified Entropy-Based Segmentation with Modified MRF	25-06-18	454	T1	6 min
58	SPM12_T1_IR*	SPM12 with T1 and IR sequence	31-08-15	464	T1; IR	3 min
59	SPM12_T1*	SPM12 with T1 sequence	31-08-15	473	T1	3 min
60	MNAB	Random Decision Forests	26-09-13	492	T1; T1-IR; FLAIR	15 min
61	SPM12_T1_IR_F*	"SPM12 with T1; IR and FLAIR sequence"	31-08-15	528	T1; IR; FLAIR	4 min
62	FSL-Seg*	FSL (v5.0) seg file	31-08-15	532	T1	10 min
63	FreeSurfer*	FreeSurfer (v5.3.0)	31-08-15	539	T1-1mm	1 hour
64	LUH-TNT	Level-sets with dictionary learning	27-10-14	540	T1	< 1 min
65	FSL-PVSeg*	FSL (v5.0) PVSeg file	31-08-15	544	T1	10 min
66	Jedi Mind Meld	Automated Walks using Machine Learning	26-09-13	559	T1; T1-IR; FLAIR	27 sec
67	mlmi2014ss	AdaBoostM2	28-07-14	562	T1; T1-IR; FLAIR	5 min
68	S2_QM	Bayesian-based Adaptive Mean-Shift	26-09-13	576	T1; T1-IR; FLAIR	1.5 hours
69	CSIM-Lab	Multiple Logistic Classification	24-10-17	590	T1
70	ISTB UniBe	Decision Forests with spatial regularization	04-06-14	604	T1; T1-IR; FLAIR	3 min
71	LNMBrains	Gaussian Intensity Model	26-09-13	604	T1; T1-IR; FLAIR	5 min
72	ECOESH	Modified Expectation Maximization	06-01-14	647	T1	5 min

* These methods (SPM, FSL and FreeSurfer) were applied by the MRBrainS organizers as described in the MRBrainS journal paper.
², ³ At the end of the submission name indicates that this is the 2^nd or 3^rd submission (max 3 resubmissions allowed).

Note: As described below, when teams have equal scores, the standard deviations of the three tissue types and the three evaluation measures over all patients are taken into account.

How is the ranking determined?

Ranking is based on the evaluation results of all 15 test datasets and determined as follows: For each evaluation measure (Dice Coefficient (DC), Modified Hausdorff Distance (HD) and Absolute Volume Difference(AVD)), the mean value over all 15 datasets is determined for white matter (WM), gray matter (GM) and cerebrospinal fluid (CSF). Each team receives a rank (1=best) for each tissue type (GM, WM, CSF) and each evaluation measure (DC, HD-95, AVD) based on the mean value of the evaluation measures over all 15 datasets. The final score is determined by adding the ranks of all tissue types and evaluation measures for each team. The team with the lowest score is ranked number 1. In case two teams would have an equal score, the team with the lowest standard deviation over the tissue types will be ranked first.