High-Resolution, Parallelized Imaging of Freely Swimming Zebrafish with a Gigapixel Microscope

意义：高通量光学显微镜目前正在改变遗传学、药物发现等研究领域 和神经科学。大型光学分析现在通常使用数千张高分辨率图像来提供关键的见解 进入人体，我们的大脑和影响我们的疾病。然而，今天的光学显微镜还远远不够理想。 由于大型镜头设计的挑战，没有一台标准显微镜可以捕捉到超过5000万像素的每张图像快照， 这使得不可能在多厘米的可视区域(视场， FOV)。对于在体内筛选和监测斑马鱼，这种分辨率/视野的权衡是一个关键的瓶颈：每个生物体 必须被限制或瘫痪才能以高分辨率成像，自由游泳的生物体只能在低分辨率下观看， 目前还没有设备能够以细胞分辨率同时监测多条游动的斑马鱼。建议：光波前 在实验室，LLC(OWL)开发了一种新的显微镜，克服了这些限制。其第一阶段“微型摄像机阵列” 显微镜“原型机(MCAM-1)由24个微型照相机单元和相关电子设备组成，用于捕捉亚细胞 整个大型培养皿上的分辨率图像(0.24亿像素图像)。在第二阶段，OWL将生产一款准备上市的 产品MCAM-2，具有改进的规格和软件，用于获取明亮视场和荧光视频。这个 MCAM-2将显著提高高通量显微镜筛选的效率，降低电流的复杂性 设置，并实现全新的生物实验(例如SA3)。 SA1：优化MCAM-2硬件：OWL将创建一款可达到6微米分辨率的市场就绪的MCAM-2设备 以8帧/秒(Fps)的速度在120平方厘米的视野内成像。软件选项将使视频成像速率接近24fps 面积缩小了。MCAM-2提供的每幅图像的像素(0.3亿像素)比竞争对手的顶级显微镜多15-20倍。 SA2：开发用于高速数字跟踪的电子设备和软件：与哈佛大学恩格特实验室合作， OWL将大大减少MCAM使用自动数字跟踪保存的数据量。这款新软件 将从图像中分割每个幼虫并丢弃所有剩余像素，从而将内存需求减少100倍，并促进 30帧/秒的单生物视频跟踪。此外，OWL还将在其当前的Python中添加几个图像分析功能 软件界面(例如3D位置、眼睛位置、尾部曲率)提供最先进的精度(&lt；5%的误差，3-10分钟)。 SA3：展示神经活动的荧光成像：与杜克大学诺曼实验室合作， OWL将提高MCAM的荧光检测的灵敏度和准确性。专用硬件附加组件(激励 源和发射滤光片阵列)将在静止和自由地提供15-25的荧光图像信噪比 移动中的转基因幼虫。具有泛神经元GCaMP6s表达的自由游泳转基因幼虫的校准视频将 验证MCAM-2可以在自然相互作用期间同时非侵入性地测量&gt；10生物体的神经活动。 SA4：进行用户试验并收集反馈：OWL将向5个研究小组提供MCAM-2原型 在3个月的试用期内通过问卷提供详细的反馈。然后，OWL将把评论纳入最终产品中。 这一第二阶段项目的成果将是一个旗舰的MCAM-2设备和软件，准备好进行中等规模的生产。

Significance: High-throughput optical microscopy is currently transforming the research fields of genetics, drug discovery and neuroscience. Large-scale optical assays now routinely use thousands of high-resolution images to offer critical insights into the human body, our brain and the diseases that affect us. Today's optical microscopes, however, are still far from ideal. Due to challenges with large lens design, no standard microscope can capture more than 50 megapixels per image snapshot, which makes it impossible to simultaneously image at cellular-resolution over a multi-centimeter viewing area (field of view, FOV). For screening and monitoring zebrafish in vivo, this resolution/FOV tradeoff is a critical bottleneck: each organism must be constrained or paralyzed to image at high resolution, freely swimming organisms can only be viewed at low resolution, and no setups yet can monitor multiple swimming zebrafish at cellular resolution in parallel. Proposal: Optical Wavefront Laboratories, LLC (OWL) has developed a new microscope that overcomes these limitations. Its Phase I “micro-camera array microscope” prototype (the MCAM-1) consists of 24 micro-camera units and associated electronics to capture sub-cellular resolution images over an entire large petri dish (0.24 gigapixel images). In Phase II, OWL will produce a market-ready product, the MCAM-2, with improved specifications and software for acquiring both bright-field and fluorescence videos. The MCAM-2 will significantly improve the efficiency of high-throughput microscope screening, reduce the complexity of current setups, and enable completely new biological experiments (e.g., SA3). SA1: Optimize MCAM-2 hardware: OWL will create a market-ready MCAM-2 device that achieves 6 µm resolution imaging across an 120 cm2 FOV at 8 frames/sec (fps). Software options will allow video imaging rates to approach 24 fps over a reduced area. The MCAM-2 offers 15-20X more pixels per image (0.3 gigapixels) than top competing microscopes. SA2: Develop electronics and software for high-speed digital tracking: Working with the Engert Lab at Harvard, OWL will dramatically reduce the amount of data saved by the MCAM using automated digital tracking. This new software will segment each larva from images and discard all residual pixels, decreasing memory requirements by 100X and facilitating 30 fps single-organism video tracking. In addition, OWL will add several image analysis functions to its current Python software interface (e.g. 3D position, eye position, tail curvature) offering state-of-the-art accuracy (<5% error, 3-10 min.). SA3: Demonstrate fluorescence imaging of neural activity: Working with the Naumann Lab at Duke University, OWL will improve the MCAM's sensitivity and accuracy of fluorescence detection. Dedicated hardware add-ons (an excitation source and emission filter array) will provide a fluorescence image signal-to-noise ratio of 15-25 in stationary and freely moving transgenic larvae. Calibrated videos of freely swimming transgenic larvae with pan-neuronal GCaMP6s expression will verify the MCAM-2 can non-invasively measure neural activity in >10 organisms simultaneously during natural interactions. SA4: Conduct user trials and gather feedback: OWL will provide MCAM-2 prototypes to 5 research groups for detailed feedback via questionnaires over a 3-month trial. OWL will then incorporate comments into a finalized product. The outcome of this Phase II project will be a flagship MCAM-2 device and software ready for medium-scale production.