A subtraction pipeline for automatic detection of new appearing multiple sclerosis lesions in longitudinal studies
Time-series analysis of magnetic resonance images (MRI) is of great value for multiple sclerosis (MS) diagnosis and follow-up. In this paper, we present an unsupervised subtraction approach which incorporates multisequence information to deal with the detection of new MS lesions in longitudinal studies.
The proposed pipeline for detecting new lesions consists of the following steps: skull stripping, bias field correction, histogram matching, registration, white matter masking, image subtraction, automated thresholding, and postprocessing. We also combine the results of PD-w and T2-w images to reduce false positive detections.
Experimental tests are performed in 20 MS patients with two temporal studies separated 12 (12M) or 48 (48M) months in time. The pipeline achieves very good performance obtaining an overall sensitivity of 0.83 and 0.77 with a false discovery rate (FDR) of 0.14 and 0.18 for the 12M and 48M datasets, respectively. The most difficult situation for the pipeline is the detection of very small lesions where the obtained sensitivity is lower and the FDR higher.
Our fully automated approach is robust and accurate, allowing detection of new appearing MS lesions. We believe that the pipeline can be applied to large collections of images and also be easily adapted to monitor other brain pathologies.
KeywordsMultiple sclerosis Brain MRI longitudinal analysis Lesion change detection 3D subtraction
Conflict of interest
We declare that we have no conflict of interest.
- 1.Battaglini M, Rossi F, Grove RA, Stromillo ML, Whitcher B, Matthews PM, De Stefano N (2013) Automated identification of brain new lesions in multiple sclerosis using subtraction images. J Magn Reson Imag to appearGoogle Scholar
- 3.Cabezas M, BachCuadra M, Oliver A, Lladó X, Freixenet J, Vilanova JC, Valls L, Ramió-Torrentà L, Huerga E, Pareto D, Rovira A (2011) A pipeline approach with spatial information for segmenting multiple sclerosis lesions on brain magnetic resonance imaging. In: Proc Europ Comm Treatm Res Mult Scl 381.Google Scholar
- 12.Moraal B, Wattjes MP, Geurts JJG (2010) Improved detection of active multiple sclerosis lesions: 3D subtraction imaging. Radiology 255(1)Google Scholar
- 17.Souplet JC, Lebrun C, Ayache N, Malandain G (2008) An automatic segmentation of T2-FLAIR multiple sclerosis lesions. In: MICCAI—Challenge, 1–11Google Scholar
- 20.Tan IL, van Schijndel RA, van Walderveen MAA, Quist M, Bos R, Pouwels PJW, Desmedt P, Ader HJ, Barkhof F (2002) Magnetic resonance image registration in multiple sclerosis: comparison with repositioning error and observer-based variability. J Magn Reson Imaging 15(5):505–510PubMedCrossRefGoogle Scholar
- 22.Vrenken H, Jenkinson M, Hors eld MA, Battaglini M, Van Schijndel RA, Ros-trup E, Geurts JJG, Fisher E, Zijdenbos A, Ashburner J, Miller DH, Filippi M, Fazekas F, Rovaris M, Rovira A, Barkhof F, De Stefano N (2013) Recommendations to improve imaging and analysis of brain lesion load and atrophy in longitudinal studies of multiple sclerosis. J Neurol 260(10):2458–2471PubMedCentralPubMedCrossRefGoogle Scholar