Results

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.

RankTeam nameSubmission nameSubmission dateScoreSequences usedDuration
1XMU_SmartDSP23D Spatial Weighted U-Net for Multi-modality Brain MRI Segmentation29-08-1833T1; T1_IR; FLAIR10 min
2TailHotMulti-modality aggregation network313-04-1860T1; T1_IR; FLAIR13 sec
3WTA23D Cascade convolutional architecture - Method 2223-05-1876T1; T1_IR; FLAIR2 min
4XMU SmartDSP3D CNN with a Cross-modality Channel Attention Scheme317-08-1882T1; T1_IR; FLAIR10 min
5XLab3D Fully CNN with Multi-Modality Feature Fusion310-08-1892T1; T1_IR; FLAIR2 min
6LIVIA_ETSHyperDenseNet206-02-1899T1; T1_IR; FLAIR4 min
7CU_DL2"3D Deep Learning; voxnet2"28-06-16112T1; T1_IR; FLAIR2 min
8CU_DL"3D Deep Learning; voxnet13"16-06-16120T1; T1_IR; FLAIR2 min
9muyaFusion Weighted Network304-07-19123T1; T1_IR; FLAIR10 min
10LRDEFully Convolutional Network20-12-16129T12 sec
11CVSSCNN with Attention Mechanism312-03-19132T1; T1_IR; FLAIR1 min
12MSL-SKKUDeep Convolutional Neural Network19-06-17136T1; T1-IR; FLAIR1.5 min
13QL111111Multi-modality Aggregation Network with Self-attention and Deep Feature Reconstruction218-03-19159T1; T1_IR; FLAIR10 sec
14MDGRUMulti-Dimensional Gated Recurrent Units327-07-16167T1; T1_IR; FLAIR2 min
15FBI/LMB FreiburgU-Net (3D)01-05-16170T1-1mm; T1-IR; FLAIR2 min
16ASTRI_DLU-Net inspired CNN model with 3D context information225-07-18170T1; T1_IR; FLAIR7 min
17PyraMiD-LSTM2NOCC with rounds323-05-16176T1; T1-IR; FLAIR2 min
18THUityModified U-Net315-08-18178T140 sec
19NISTFully CNN for Multi-Modality Brain Tissue Segmentation14-08-18182T1; T1_IR; FLAIR2 min
20AOCAtlas of Classifiers24-12-17197T16 sec
21IDSIAPyraMiD-LSTM05-06-15204T1; T1_IR; FLAIR2 min
22STHHybrid ANN-based Auto-context method203-06-16218T1; T1-IR; FLAIR5 min
23nanand2DeepLab-v3+ with pre-training on FreeSurfer automated labels316-08-18222T115 sec
24ISI-NeonatologyMulti-stage voxel classification31-05-14226T11.5 hours
25UNC-IDEALINKS:Learning-based multi-source integration09-02-15230T1; T1_IR; FLAIR3 min
26ASTRI_MSARevised VoxResNet27-08-18249T1; T1_IR; FLAIR1 min
27LfBMulti-modality Mix Network17-01-19256T1; T1_IR; FLAIR40 sec
28BCH_CRL_IMAGINE3D patch-wise DenseNet and Patch Fusion224-05-18256T1; T1-IR; FLAIR2 min
29KSOM GHMFASeTs: MAP-Based with Manifold learning13-05-14263T1; T1_IR; FLAIR23 min
30MNAB2Random Forests21-02-14267T1; T1_IR; FLAIR25 min
31WTA3D Cascade convolutional architecture - Method 1315-05-18273T1; T1_IR; FLAIR5 min
32vicorob UdG T1_FMSSEG using T1 + FLAIR (T1-IR skull)14-01-16277T1; IR; FLAIR2 min
33VBM12VBM12_r738 with WMHC=207-10-15281T16 min
34BIGR2Multi-Feature SVM Classification26-09-13291T1; T1_IR; FLAIR35 min
35vicorob UdG T1MSSEG using only T1 (T1-IR skull)21-01-16297T1; IR2 min
36UofL BioImagingMAP-Based Framework26-09-13327T16 sec
37mehran007Tsallis-Entropy Segmentation through MRF04-04-19338T16.5 min
38CMIVModel-guided Level Sets and Skeletons26-09-13346T1; FLAIR3 min
39NarsilSegmentation using Ensemble Trees26-09-13347T1; FLAIR< 2 min
40UB VPML MedMulti-Atlas with Multiway Cut26-09-13349T1; T1_IR; FLAIR30 min
41draiVGG based Fully Convolutional Neural Network20-01-17363T12 sec
42bigr_neuroAutomatically Trained kNN Classifier26-09-13367T1; FLAIR2 hours
43VIBOT10 M-T1+FConvolutional Neural Network27-02-17367T1; FLAIR3 min
44S2_QM2Atlas-based segmentation and AdaBoost28-07-14376T1; T1-IR; FLAIR45 min
45SPM12_T1_F*SPM12 with T1 and FLAIR sequence31-08-15379T1; FLAIR3 min
46RobartsMulti-Atlas with Hierarchical Max-Flow26-09-13384T1-1mm; T1-IR16 min
47KM-MRF-MASMAP on Intensity, and Local and Multi-atlas Priors28-10-16384T110 min
48SJCEFuzzy Clustering12-04-17391T1<1 min
49CSIM-lab_Mqe-MMRFHybrid Modified Entropy-Based Segmentation with Modified MRF25-06-18394T16 min
50SPM12_T1_IR*SPM12 with T1 and IR sequence31-08-15400T1; IR3 min
51SPM12_T1*SPM12 with T1 sequence31-08-15411T13 min
52MNABRandom Decision Forests26-09-13422T1; T1-IR; FLAIR15 min
53SPM12_T1_IR_F*"SPM12 with T1; IR and FLAIR sequence"31-08-15457T1; IR; FLAIR4 min
54FSL-Seg*FSL (v5.0) seg file31-08-15462T110 min
55LUH-TNTLevel-sets with dictionary learning27-10-14468T1< 1 min
56FreeSurfer*FreeSurfer (v5.3.0)31-08-15470T1-1mm1 hour
57FSL-PVSeg*FSL (v5.0) PVSeg file31-08-15474T110 min
58Jedi Mind MeldAutomated Walks using Machine Learning26-09-13488T1; T1-IR; FLAIR27 sec
59mlmi2014ssAdaBoostM228-07-14490T1; T1-IR; FLAIR5 min
60S2_QMBayesian-based Adaptive Mean-Shift26-09-13505T1; T1-IR; FLAIR1.5 hours
61CSIM-LabMultiple Logistic Classification24-10-17518T1
62ISTB UniBeDecision Forests with spatial regularization04-06-14532T1; T1-IR; FLAIR3 min
63LNMBrainsGaussian Intensity Model26-09-13533T1; T1-IR; FLAIR5 min
64ECOESHModified Expectation Maximization06-01-14575T15 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 2nd or 3rd 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.