Mori Susumu, Onda Kengo, Fujita Shohei, Suzuki Toshiaki, Ikeda Mikimasa, Zay Yar Myint Khin, Hikage Jun, Abe Osamu, Tomimoto Hidekazu, Oishi Kenichi, Taguchi Junichi
Department of Radiology, Johns Hopkins University, School of Medicine, 330 Traylor Bldg, 217 Rutland Ave, Baltimore, MD 21205, USA.
Tokyo Medical and Dental University, 1 Chome-5-45 Yushima, Bunkyo City, Tokyo 113-0034, Japan.
Brain Commun. 2022 Aug 22;4(4):fcac211. doi: 10.1093/braincomms/fcac211. eCollection 2022.
Although health screening plays a key role in the management of chronic diseases associated with lifestyle choices, brain health is not generally monitored, remaining a black box prior to the manifestation of clinical symptoms. Japan is unique in this regard, as brain MRI scans have been widely performed for more than two decades as part of Brain Dock, a comprehensive health screening programme. A vast number of stored images (well over a million) of longitudinal scans and extensive health data are available, offering a valuable resource for investigating the prevalence of various types of brain-related health conditions occurring throughout adulthood. In this paper, we report on the findings of our preliminary quantitative analysis of T-weighted MRIs of the brain obtained from 13 980 subjects from three participating sites during the period 2015-19. We applied automated segmentation analysis and observed age-dependent volume loss of various brain structures. We subsequently investigated the effects of scan protocols and the feasibility of calibration for pooling the data. Last, the degree of brain atrophy was correlated with four known risk factors of dementia; blood glucose level, hypertension, obesity, and alcohol consumption. In this initial analysis, we identified brain ventricular volume as an effective marker of age-dependent brain atrophy, being highly sensitive to ageing and evidencing strong robustness against protocol variability. We established the normal range of ventricular volumes at each age, which is an essential first step for establishing criteria used to interpret data obtained for individual participants. We identified a subgroup of individuals at midlife with ventricles that substantially exceeded the average size. The correlation studies revealed that all four risk factors were associated with greater ventricular volumes at midlife, some of which reached highly significant sizes. This study demonstrates the feasibility of conducting a large-scale quantitative analysis of existing Brain Dock data in Japan. It will importantly guide future efforts to investigate the prevalence of large ventricles at midlife and the potential reduction of this prevalence, and hence of dementia risk, through lifestyle changes.
尽管健康筛查在与生活方式选择相关的慢性病管理中起着关键作用，但大脑健康通常未得到监测，在临床症状出现之前仍是一个未知领域。日本在这方面独具特色，因为作为综合健康筛查项目“脑库”的一部分，脑部磁共振成像扫描已广泛开展了二十多年。现已有大量存储的纵向扫描图像（超过一百万张）和广泛的健康数据，为研究成年期各种脑部相关健康状况的患病率提供了宝贵资源。在本文中，我们报告了对2015年至2019年期间来自三个参与站点的13980名受试者的脑部T加权磁共振成像进行初步定量分析的结果。我们应用了自动分割分析，观察到各种脑结构存在年龄依赖性体积减少。随后，我们研究了扫描协议的影响以及合并数据校准的可行性。最后，将脑萎缩程度与四种已知的痴呆风险因素进行了关联；血糖水平、高血压、肥胖和饮酒量。在这项初步分析中，我们确定脑室体积是年龄依赖性脑萎缩的有效标志物，对衰老高度敏感，并且在应对协议变异性方面具有很强的稳健性。我们确定了每个年龄脑室体积的正常范围，这是建立用于解释个体参与者数据的标准的重要第一步。我们确定了中年时期脑室体积大幅超过平均大小的一个亚组个体。相关性研究表明，所有四个风险因素都与中年时期更大的脑室体积相关，其中一些达到了高度显著的程度。这项研究证明了对日本现有的“脑库”数据进行大规模定量分析的可行性。它将重要地指导未来的研究工作，以调查中年时期大脑室的患病率以及通过改变生活方式降低这种患病率从而降低痴呆风险的可能性。
