分类： 生物学 >> 生物物理学 >> 神经科学 提交时间： 2016-05-11
Liu, Huilang
Bai, Fan
Ma, Yuanye
Wang, Jianhong
Liu, Huilang
Bai, Fan
Ma, Yuanye
Wang, Jianhong
Liu, Huilang
Bai, Fan
Ma, Yuanye
Wang, Jianhong
Liu, Cirong
Liu, Cirong
Liu, Cirong
Tian, Xiaoguang
Tian, Xiaoguang
Mo, Yin
Zhao, Xudong
Ma, Yuanye
摘要： Early brain development is a complex and rapid process, the disturbance of which may cause the onset of brain disorders. Based on longitudinal imaging data acquired from 6 to 16 months postnatal, we describe a systematic trajectory of monkey brain development during late infancy, and demonstrate the influence of phencyclidine (PCP) on this trajectory. Although the general developmental trajectory of the monkey brain was close to that of the human brain, the development in monkeys was faster and regionally specific. Gray matter volume began to decrease during late infancy in monkeys, much earlier than in humans in whom it occurs in adolescence. Additionally, the decrease of gray matter volume in higher-order association regions (the frontal, parietal and temporal lobes) occurred later than in regions for primary functions (the occipital lobe and cerebellum). White matter volume displayed an increasing trend in most brain regions, but not in the occipital lobe, which had a stable volume. In addition, based on diffusion tensor imaging, we found an increase in fractional anisotropy and a decrease in diffusivity, which may be associated with myelination and axonal changes in white matter tracts. Meanwhile, we tested the influence of 14-day PCP treatment on the developmental trajectories. Such treatment tended to accelerated brain maturation during late infancy, although not statistically significant. These findings provide comparative information for the understanding of primate brain maturation and neurodevelopmental disorders. (C) 2014 Elsevier Inc. All rights reserved.
分类： 生物学 >> 生物物理学 >> 生物物理、生物化学与分子生物学 提交时间： 2016-05-11
摘要： Although the topic of fiducial marker-based alignment in electron tomography (ET) has been widely discussed for decades, alignment without human intervention remains a difficult problem. Specifically, the emergence of subtomogram averaging has increased the demand for batch processing during tomographic reconstruction; fully automatic fiducial marker-based alignment is the main technique in this process. However, the lack of an accurate method for detecting and tracking fiducial markers precludes fully automatic alignment. In this paper, we present a novel, fully automatic alignment scheme for ET. Our scheme has two main contributions: First, we present a series of algorithms to ensure a high recognition rate and precise localization during the detection of fiducial markers. Our proposed solution reduces fiducial marker detection to a sampling and classification problem and further introduces an algorithm to solve the parameter dependence of marker diameter and marker number. Second, we propose a novel algorithm to solve the tracking of fiducial markers by reducing the tracking problem to an incomplete point set registration problem. Because a global optimization of a point set registration occurs, the result of our tracking is independent of the initial image position in the tilt series, allowing for the robust tracking of fiducial markers without pre-alignment. The experimental results indicate that our method can achieve an accurate tracking, almost identical to the current best one in IMOD with half automatic scheme. Furthermore, our scheme is fully automatic, depends on fewer parameters (only requires a gross value of the marker diameter) and does not require any manual interaction, providing the possibility of automatic batch processing of electron tomographic reconstruction. (C) 2015 Elsevier Inc. All rights reserved.
点击量
待评议
点击量
下载量
评论
