MSChallenge/data

The 2015 Longitudinal MS Lesion Segmentation Challenge: Data

2015 Longitudinal MS Lesion Segmentation Challenge
MS Challenge Overview	MS Challenge Data	MS Challenge Evaluation

The Data Set Summary Table (below) includes demographic details for the training data and both test data sets. The top line is the information of the entire data set, while subsequent lines within a section are specific to the patient diagnoses. The codes are RR for relapsing remitting MS, PP for primary progressive MS, and SP for secondary progressive MS. N (M/F) denotes the number of patients and the male/female ratio, respectively. Timepoints is the mean (and standard deviation) of the number of time-points provided to participants. Age is the mean age (and standard deviation), in years, at baseline. Follow-up is the mean (and standard deviation), in years, of the time between follow-up scans.

Data Set Summary Table
Data Set	N (M/F)	Time-Points Mean (SD)	Age Mean (SD)	Follow-Up Mean (SD)
Training	5 (1/4)	4 (±0.55)	43.5 (±10.3)	1.0 (±0.13)
RR	1 (1/3)	4.4 (±0.50)	43.5 (±10.3)	1.0 (±0.14)
PP	1 (0/1)	4.0	57.9	1.0 (±0.04)
Test	14 (3/11)	4.4 (±0.63)	39.3 (±8.9)	1.0 (±0.23)
RR	12 (3/9)	4.4 (±0.67)	39.2 (±9.6)	1.0 (±0.25)
PP	1 (0/1)	4.0	39.0	1.0 (±0.04)
SP	1 (0/1)	4.0	41.7	1.0 (±0.05)

Each scan was imaged and preprocessed in the same manner, with data acquired on a 3.0 Tesla MRI scanner (Philips Medical Systems, Best, The Netherlands) using the following sequences: a T1-weighted (T1-w) magnetization prepared rapid gradient echo (MPRAGE) with TR = 10.3 ms, TE = 6 ms, flip angle = 8°, & 0.82 × 0.82 × 1.17 mm³ voxel size; a double spin echo (DSE) which produces the PD-w and T2-w images with TR = 4177 ms, TE1 = 12.31 ms, TE2 = 80 ms, & 0:82 × 0:82 × 2.2 mm3 voxel size; and a T2-w fluid attenuated inversion recovery (FLAIR) with TI = 835 ms, TE = 68 ms, & 0.82 × 0.82 × 2.2 mm³ voxel size. The imaging protocols were approved by the local institutional review board. Each subject underwent the following preprocessing: the baseline (first time-point) MPRAGE was inhomogeneity-corrected using N4 (Tustison et al., 2010), skull-stripped (Carass et al., 2007, 2010), dura stripped (Shiee et al., 2014), followed by a second N4 inhomogeneity correction, and rigid registration to a 1 mm isotropic MNI template. We have found that running N4 a second time after skull and dura stripping is 25 more effective (relative to a single correction) at reducing any inhomogeneity within the images. Once the baseline MPRAGE is in MNI space, it is used as a target for the remaining images. The remaining images include the baseline T2-w, PD-w, and FLAIR, as well as the scans from each of the follow-up time-points. These images are N4 corrected and 30 then rigidly registered to the 1 mm isotropic baseline MPRAGE in MNI space. Our registration steps are inverse consistent and thus any registration based biases are avoided (Reuter and Fischl, 2011) The skull & dura stripped mask from the baseline MPRAGE is applied to all the subsequent images, which are then N4 corrected again.

For each time-point of every subject’s scans in the Training Set and Test Set, the following data are provided: the original scan images consisting of T1-w MPRAGE, T2-w, PD-w, and FLAIR, as well as the preprocessed images (in MNI space) for each of the scan modalities. The Training Set also included manual delineations by two experts identifying and segmenting WMLs on MR images.

Requesting the Data

The data is available from the Smart Stats Website after the creation of an account. To test the results of your method and be added to the Live Leaderboard just submit the results on the Smart Stats Website.

References

• A. Carass, J. Cuzzocreo, M.B. Wheeler, P.-L. Bazin, S.M. Resnick, and J.L. Prince, "Simple paradigm for extra-cerebral tissue removal: Algorithm and analysis", NeuroImage, 56(4):1982-1992, June 2011. (doi) (PubMed) (PMCID 3105165)
• A. Carass, M.B. Wheeler, J. Cuzzocreo, P.-L. Bazin, S.S. Bassett, and J.L. Prince, "A Joint Registration and Segmentation Approach to Skull Stripping", Fourth IEEE International Symposium on Biomedical Imaging (ISBI 2007), Arlington, VA, April 12-15, 2007. (doi)
• A. Carass, S. Roy, A. Jog, J.L. Cuzzocreo, E. Magrath, A. Gherman, J. Button, J. Nguyen, F. Prados, C.H. Sudre, M.J. Cardoso, N. Cawley, O. Ciccarelli, C.A.M. Wheeler-Kingshott, S. Ourselin, L. Catanese, H. Deshpande, P. Maurel, O. Commowick, C. Barillot, X. Tomas-Fernandez, S.K. Warfield, S. Vaidya, A. Chunduru, R. Muthuganapathy, G. Krishnamurthi, A. Jesson, T. Arbel, O. Maier, H. Handels, L.O. Iheme, D. Unay, S. Jain, D.M. Sima, D. Smeets, M. Ghafoorian, B. Platel, A. Birenbaum, H. Greenspan, P.-L. Bazin, P.A. Calabresi, C.M. Crainiceanu, L.M. Ellingsen, D.S. Reich, J.L. Prince, and D.L. Pham, "Longitudinal Multiple Sclerosis Lesion Segmentation: Resource and Challenge", NeuroImage, 148(C):77-102, 2017. (doi) (PubMed)
• A. Carass, S. Roy, A. Jog, J.L. Cuzzocreo, E. Magrath, A. Gherman, J. Button, J. Nguyen, P.-L. Bazin, P.A. Calabresi, C.M. Crainiceanu, L.M. Ellingsen, D.S. Reich, J.L. Prince, and D.L. Pham, "Longitudinal multiple sclerosis lesion segmentation data resource", Data in Brief, 12:346-350, 2017. (doi) (PubMed)
• M. Reuter and B. Fischl, "Avoiding Asymmetry-Induced Bias in Longitudinal Image Processing", NeuroImage, 57(1):19-21, 2011. (doi)
• N.Shiee, P.-L. Bazin, J.L. Cuzzocreo, C. Ye, B. Kishore, A. Carass, P.A. Calabresi, D.S. Reich, J.L. Prince, and D.L. Pham, "Robust Reconstruction of the Human Brain Cortex in the Presence of the WM Lesions: Method and Validation", Human Brain Mapping, 35(7):3385-3401, 2014. (doi)
• N.J. Tustison, B.B. Avants, P.A. Cook, Y. Zheng, A. Egan, P.A. Yushkevich, and J.C. Gee, "N4ITK: Improved N3 Bias Correction", IEEE Trans. on Medical Imaging, 29:1310-1320, 2010. (doi)