Whole-Organism, Real-time Decision-enabled 3D Tissue Imaging and Recovery for Molecular Analysis

摘要用于分子分析的组织捕获和回收，如免疫染色和基因表达 分析对于从与疾病有关的组织块收集信息是重要的。存在的困难 解决方案是，如果设想组织恢复，则从目标组织切片和恢复薄组织切片 或者器官是不可能的现有的解决方案提供手动组织恢复，需要 操作者在场以确定是否从感兴趣的组织收集切片。解决 面对这一挑战，BioInVision将为市场带来一种用于临床前应用的3D组织成像解决方案， 采用自动，深度学习，从整个生物体中实时识别目标组织， 组织恢复BioInVision开创了CryoVizTM仪器，并成功地将收费 服务CryoVizTM成像超过10年。我们现行的收费服务架构已获广泛采用， 全球超过75家客户和机构的支持。高分辨率、宽荧光支持（可见光 和近红外荧光团），临床前冷冻组织块的高灵敏度块面CryoVizTM成像， 解剖学明场和分子标记荧光3D显微图像体积。基于AI的软件， 深度学习，在遇到彩色解剖结构中的感兴趣组织时实时通知操作员，或 一旦遇到分子荧光，如eGFP癌细胞，能够进一步询问这些 通过超高分辨率组织成像（2µ m尺度）或半自动组织捕获， 组织学分析和免疫染色（我们称之为“图像引导组织学”）。半自动组织捕获 通过速度和温度控制将能够回收组织用于分子分析。示范 概述了一个项目，该项目涉及用荧光报告转移癌细胞表征整个小鼠， 荧光团束缚的癌症靶向成像剂。在这里，我们将研究乳腺癌转移的分布 贯穿整个小鼠，并与免疫细胞或疾病生物标志物共定位。我们的解决方案还将 使荧光示踪剂的药物传递研究成为可能。它将能够追踪多种荧光标记- 细胞类型的标记有助于更好地了解癌症生物学中的肿瘤微环境。它将 减少人工劳动和人员成本，并导致更好的生产量，以实现图像引导的组织学。这 新的解决方案将满足各种应用领域，包括癌症生物学，药物输送，成像 代理和基因表达。

Summary. Tissue capture and recovery for molecular analyses such as immunostaining and gene expression analysis is important for gleaning information from tissue blocks relating to disease. The difficulty with existing solutions is that if tissue recovery is envisioned, sectioning and recovering thin tissue sections from target tissues or organs from an entire organism is not possible. Existing solutions provide manual tissue recovery, requiring the operator to be present to determine whether or not to collect a section from a tissue of interest. To address this challenge, BioInVision will bring to market a 3D tissue imaging solution for preclinical applications that em- ploys automatic, deep-learning, on-the-fly target tissue recognition from a whole organism and semi-automated tissue recovery. BioInVision pioneered the CryoVizTM instrument and has successfully commercialized fee-for- service CryoVizTM imaging for over 10 years. Our existing fee-for-service framework is well-adopted and em- braced by 75+ customers and institutions all over the world. High-resolution, broad-fluorescence-support (visible and NIR fluorophores), high-sensitivity block-face CryoVizTM imaging of preclinical frozen tissue blocks creates anatomical brightfield and molecular marker fluorescence 3D microscopic image volumes. AI-based software for deep-learning that will notify the operator in real-time upon encountering tissues of interest in color anatomy, or upon encountering molecular fluorescence such as eGFP cancer cells, enabling further interrogation of these tissues either through ultra-high-resolution tissue imaging (2µm-scale), or semi-automatic tissue capture for his- tological analyses and immunostaining (what we term “image-guided histology”). Semi-automatic tissue capture through speed and temperature control will enable recovery of tissue for molecular analyses. A demonstration project is outlined that involves characterizing whole mice with fluorescent-reporter metastatic cancer cells and fluorophore-tethered cancer targeted imaging agent. Here, we will study distribution of breast cancer metastases throughout the whole mouse and co-localization with immune cells or disease biomarkers. Our solution will also make possible drug delivery studies with fluorescent tracers. It will enable tracking of multiple fluorescently la- beled markers of cell types helping one better understand the tumor micro-environment in cancer biology. It will reduce manual labor and personnel costs and lead to better throughput to enable image-guided histology. This novel solution will cater to a wide variety of application areas including cancer biology, drug delivery, imaging agents and gene expression.