Higher‐resolution quantification of white matter hypointensities by large‐scale transfer learning from 2D images on the JPSC‐AD cohort. Issue 13 (7th May 2022)

• Record Type: Journal Article
• Authors: Thyreau, Benjamin; Tatewaki, Yasuko; Chen, Liying; Takano, Yuji; Hirabayashi, Naoki; Furuta, Yoshihiko; Hata, Jun; Nakaji, Shigeyuki; Maeda, Tetsuya; Noguchi‐Shinohara, Moeko; Mimura, Masaru; Nakashima, Kenji; Mori, Takaaki; Takebayashi, Minoru; Ninomiya, Toshiharu; Taki, Yasuyuki
• Is Part Of: Human brain mapping. Volume 43:Issue 13(2022)
• Page Start: 3998
• Page End: 4012
• Publication Date: 2022-05-07
• Subjects: Brain mapping -- Periodicals; 611.81
• DOI: 10.1002/hbm.25899
• Languages: English
• ISSNs: 1065-9471

Abstract

Abstract: White matter lesions (WML) commonly occur in older brains and are quantifiable on MRI, often used as a biomarker in Aging research. Although algorithms are regularly proposed that identify these lesions from T2‐fluid‐attenuated inversion recovery (FLAIR) sequences, none so far can estimate lesions directly from T1‐weighted images with acceptable accuracy. Since 3D T1 is a polyvalent and higher‐resolution sequence, it could be beneficial to obtain the distribution of WML directly from it. However a serious difficulty, both for algorithms and human, can be found in the ambiguities of brain signal intensity in T1 images. This manuscript shows that a cross‐domain ConvNet (Convolutional Neural Network) approach can help solve this problem. Still, this is non‐trivial, as it would appear to require a large and varied dataset (for robustness) labelled at the same high resolution (for spatial accuracy). Instead, our model was taught from two‐dimensional FLAIR images with a loss function designed to handle the super‐resolution need. And crucially, we leveraged a very large training set for this task, the recently assembled, multi‐sites Japan Prospective Studies Collaboration for Aging and Dementia (JPSC‐AD) cohort. We describe the two‐step procedure that we followed to handle such a large number of imperfectly labeled samples. A large‐scale accuracy evaluation conducted against FreeSurfer 7, and a further visual expert rating revealed that WML segmentation from our ConvNet was consistently better. Finally, we made a directly usable software program based on that trained ConvNet model, available at https://github.com/bthyreau/deep-T1-WMH . Abstract : It is possible to identify White Matter Lesions on the high‐resolution, morphometry‐friendly, T1‐weigthed sequence directly with good accuracy. We leverage a new, large, Eldery brain cohort to train and evaluate a ConvNet identification model, solving various specific technical challenge. Two trained Radiologists judged the outcome much above FreeSurfer and even FLAIR‐based segmentation.