Canada-UK AI 2019 : Self-guided microrobots for automated brain dissection

神经前体细胞是神经干细胞及其后代，是多能、自我更新的细胞，负责在胚胎发育期间构建脊椎动物中枢神经系统的基本组成部分。值得注意的是，神经前体细胞作为非常罕见的细胞持续存在于成熟的大脑中，在生物体的整个生命周期中继续产生新的神经元。对这些珍贵细胞的深入研究可能会导致神经生物学领域的分水岭时刻，以及治疗阿尔茨海默病、痴呆症和创伤性脑损伤的再生疗法的发展。然而，我们利用鼻咽癌治疗潜力的能力受到识别和收集它们进行分析的实际困难的限制。目前的方法需要对神经组织进行艰苦的手工显微解剖，然后采用细胞分选方法，这种方法除了需要大量样本(因此需要大量组织)外，还往往无法区分神经前体细胞和其他表达相同生物标志物的细胞。 这个高度跨学科的项目汇集了人工智能(AI)、机器人学、工程学和神经科学方面的专业知识的研究人员，以创建一个新的自动化显微外科平台，该平台将用于识别和收集脑组织中的NPC。通过利用深度学习方法在复杂数据集中检测特征的能力，我们将开发两种新的图像引导显微手术工具：能够仔细切除发现NPC的脑组织(SVZ)小区域的微型机器人切割器，以及能够使用协调的光电微型机器人阵列高效地采集单个细胞以进行进一步研究和分析的微型机器人细胞收集器。然后，旨在将这两个系统集成到一个单一的、独特的机器人显微手术平台中。 新系统将用于收集和分析不同脑组织中的npc。特别是，我们建议分析鼻咽癌表型的性别依赖差异，这在以前的研究中已经报道过。除了改变我们收集神经干细胞的能力，同样支持新的再生疗法的开发，我们相信这种方法将为广泛的显微外科应用提供一种强大的新方法。总之，我们将利用这些结果来模拟相关工程界、科学界和更广泛的公众之间的辩论，以形成一个新的国际多学科网络，专注于再生医学的挑战，即为政府提供新的案例研究。政策和对人工智能驱动的医疗应用的进一步投资。

Neural precursor cells (NPCs) are neural stem cells and their progeny, which are multipotent, self-renewing cells that are responsible for building the fundamental components of the vertebrate central nervous system during embryonic development. Notably, NPCs persist into the mature brain as very rare cells that continue to generate new neurons throughout the lifetime of the organism. Intensive study of these precious cells could lead to a watershed moment in the field of neurobiology and the development of regenerative therapies to treat Alzheimer's disease, dementia, and traumatic brain injury. However, our ability to harness the therapeutic potential of NPCs is limited by the practical difficulty of identifying and collecting them for analysis. Current methods require painstaking manual microdissection of neural tissue followed by cell sorting methods which, in addition to requiring large samples sizes (and thus large quantities of tissue) also often fail to distinguish between NPCs and other cells expressing the same biomarkers. This highly interdisciplinary project brings together researchers with expertise in artificial intelligence (AI), robotics, engineering and neuroscience to create a new automated microsurgical platform which will be applied to identify and collect NPCs from brain tissues. By exploiting the capacity of deep learning approaches to detect features in complex datasets we will develop two new image-guided microsurgery tools: a microrobotic resector capable of carefully excising the small region of brain tissue (the SVZ) within which NPCs are found and microrobotic cell collector, capable of efficiently harvesting individual cells for further study and analysis using a coordinated array of optoelectronic microrobots. When then aim to integrate these two systems into a single, unique robotic microsurgery platform. The new system will be applied to collect and analyse NPCs from different brain tissues. In particular, we propose to analyse the sex-dependent differences in NPCs phenotypes, which have been reported in previous research. As well as transforming our capacity for collection of NPCs and likewise supporting the development of new regenerative therapies, we believe this approach will provide a powerful new method for a wide range of microsurgery applications. Altogether, we will use these results to simulate debate among the related engineering, scientific and wider public communities to shape anew international multi-disciplinary network focusing on challenges with regenerative medicine i.e. providing new case studies for govt. policy and further investment for AI-driven healthcare applications.