Patch-based 3D U-Net and transfer learning for longitudinal piglet brain segmentation on MRI.

BACKGROUND AND OBJECTIVES: In order to study neural plasticity in immature brain following early brain lesion, large animal model are needed. Because of its morphological similarities with the human developmental brain, piglet is a suitable but little used one. Its study from Magnetic Resonance Imaging (MRI) requires the development of automatic algorithms for the segmentation of the different structures and tissues. A crucial preliminary step consists in automatically segmenting the brain. METHODS: We propose a fully automatic brain segmentation method applied to piglets by combining a 3D patch-based U-Net and a post-processing pipeline for spatial regularization and elimination of false positives. Our approach also integrates a transfer-learning strategy for managing an automated longitudinal monitoring evaluated for four developmental stages (2, 6, 10 and 18 weeks), facing the issue of MRI changes resulting from the rapid brain development. It is compared to a 2D approach and the Brain Extraction Tool (BET) as well as techniques adapted to other animals (rodents, macaques). The influence of training patches size and distribution is studied as well as the benefits of spatial regularization. RESULTS: Results show that our approach is efficient in terms of average Dice score (0.952) and Hausdorff distance (8.51), outperforming the use of a 2D U-Net (Dice: 0.919, Hausdorff distance: 11.06) and BET (Dice: 0.764, Hausdorff distance: 25.91). The transfer-learning strategy achieves a good performance on older piglets (Dice of 0.934 at 6 weeks, 0.956 at 10 weeks and 0.958 at 18 weeks) compared to a standard training strategy with few data (Dice of 0.636 at 6 weeks, 0.907 at 10 weeks, not calculable at 18 weeks because of too few training piglets). CONCLUSIONS: In conclusion, we provide a method for longitudinal MRI piglet brain segmentation based on 3D U-Net and transfer learning which can be used for future morphometric studies and applied to other animals.

Effect of field strength on MR images: comparison of the same subject at 0.5, 1.0, and 1.5 T.

To assess the effect of field strength on magnetic resonance (MR) images, the same healthy subject was imaged at three field strengths: 0.5, 1.0, and 1.5 T. Imaging was performed with three similarly equipped MR imagers of the same generation and from the same manufacturer. The same imaging sequences were used with identical parameters and without repetition time correction for field strength. Imaging was performed in four anatomic locations: the brain, lumbar spine, knee, and abdomen. Quantitative image analysis involved calculation of signal-to-noise ratio, contrast-to-noise ratio, and relative contrast; qualitative image analysis was performed by four readers blinded to field strength. The results of all of the examinations were considered to be of diagnostic value. In general, signal-to-noise ratio and contrast-to-noise ratio were lowest at 0.5 T and highest at 1.5 T; relative contrast was not related to field strength. At qualitative analysis, images obtained at 1.0 and 1.5 T were superior to images obtained at 0.5 T; qualitative differences were less important in locations where there is motion or high magnetic susceptibility differences between tissues (e.g., the spine and abdomen). However, excellent image quality was obtained with all three field strengths.