Massively parallel high-speed 3D functional photoacoustic computed tomography of the adult human brain

摘要(30行) 大脑倡议(RFA-EB-19-002)呼吁开发全新的或下一代 非侵入性人脑成像工具和方法将导致我们的 对人脑的理解。超高场功能磁共振成像(FMRI)已经取得了巨大的进步 在时空分辨率上，允许在大脑皮层和大脑皮层的水平上研究大脑功能。 然而，功能磁共振成像通常被认为具有较低的敏感性和较强的组织背景来检测 功能。正电子发射断层扫描通过放射性示踪剂提供强大的代谢成像，但 空间分辨率低，漫反射光学层析成像也是如此，尽管它在速度、成本和 可移植性。只有超声波成像不能成像成人的大脑，因为超声波是 由于往返传播，颅骨有两次衰减和像差。 为了解决这些问题，我们建议为FAST开发3D光声计算机层析成像(PACT) 以及人类大脑中的超高速大规模神经活动成像。PACT特别适合于检测 血流动力学改变与神经活动有关。它提供了类似的空间分辨率，但可以制作得更多 比功能磁共振成像还快。它对低组织背景的氧合和脱氧血红蛋白都直接敏感。 与功能磁共振成像相比，PACT的其他潜在优势包括开放式成像平台、最低限度的站点要求、安静和 床边操作，无磁铁环境，系统维护低。 在过去的二十年里，我们在多个空间尺度上开发了光声技术，范围从 从微观(亚细胞和细胞)到宏观(整个啮齿动物、整个人类乳房、体外成人) 并初步进行了单通道2D和通道3D活体成人脑成像。我们有 揭示了啮齿类动物大脑对胡须或电刺激的血流动力学反应，并绘制了休眠状态图- 大鼠大脑深部丘脑的功能连接状态。我们还开发了复杂的 光声波在非均匀介质中传播的数值模拟方法及发展 声学非均匀介质中图像重建的框架。此外，我们还成功地 通过成人头骨演示了体外PACT，并获得了人头的初步图像 活着。我们建议通过两个具体目标将PACT中的这些进展转化为人脑成像： 目标1：为活体快速和超快功能人脑开发大规模并行高速3D PACT 成像。 目的2：通过与超高场7T功能磁共振成像的比较，验证在成人体内的功能PACT。

ABSTRACT (30 Lines) The BRAIN initiative (RFA-EB-19-002) has called for the development of entirely new or next-generation noninvasive human brain imaging tools and methods that will lead to transformative advances in our understanding of the human brain. Functional MRI (fMRI) at ultrahigh fields has made tremendous improvements in spatiotemporal resolution, allowing brain function to be studied on the level of cortical layers and columns. However, fMRI is generally considered to have a low sensitivity and strong tissue background for detection of function. Positron emission tomography provides powerful metabolic imaging through radioactive tracers but suffers low spatial resolution, as is diffuse optical tomography despite its advantages in speed, cost, and portability. Ultrasound-only imaging cannot image adult human brains because the ultrasonic waves are attenuated and aberrated twice by the skull due to the round-trip propagation. To address these issues, we propose to develop 3D photoacoustic computed tomography (PACT) for fast and ultrafast large-scale neural activity imaging in human brains. PACT is especially well suited for detecting hemodynamic changes related to neural activities. It offers comparable spatial resolution but can be made much faster than fMRI. It is directly sensitive to both oxy- and deoxy-hemoglobin linearly with a low tissue background. Other potential benefits of PACT over fMRI include open imaging platforms, minimal site requirements, quiet and bedside operation, magnet-free environment, and low system maintenance. In the last two decades, we have developed photoacoustic technology at multiple spatial scales ranging from microscopic (subcellular and cellular) to macroscopic (whole rodent, whole human breast, ex vivo adult human skull, and preliminary single-channel 2D and 64-channel 3D in vivo adult human brain) imaging. We have revealed hemodynamic response in the rodent brain to whisker or electrical stimulation and mapped the resting- state functional connectivity of the rat brain in the deep thalamic region. We have also developed sophisticated numerical methods for simulating photoacoustic wave propagation in heterogeneous media and developed frameworks for image reconstruction in acoustically heterogeneous media. Further, we have successfully demonstrated ex vivo PACT through adult human skulls and acquired preliminary images of human heads in vivo. We propose to translate these advances in PACT to human brain imaging through two specific aims: Aim 1: Develop massively parallel high-speed 3D PACT for in vivo fast and ultrafast functional human brain imaging. Aim 2: Validate functional PACT in adult humans in vivo by comparing with ultrahigh-field 7 T fMRI.