New Cornerstone Science Laboratory, SEU-ALLEN Joint Center, Institute for Brain and Intelligence, Southeast University, Nanjing, China.
Guangdong Institute of Intelligence Science and Technology, Hengqin, China.
Nat Methods. 2024 Oct;21(10):1936-1946. doi: 10.1038/s41592-024-02401-8. Epub 2024 Sep 4.
Digital reconstruction of the intricate 3D morphology of individual neurons from microscopic images is a crucial challenge in both individual laboratories and large-scale projects focusing on cell types and brain anatomy. This task often fails in both conventional manual reconstruction and state-of-the-art artificial intelligence (AI)-based automatic reconstruction algorithms. It is also challenging to organize multiple neuroanatomists to generate and cross-validate biologically relevant and mutually agreed upon reconstructions in large-scale data production. Based on collaborative group intelligence augmented by AI, we developed a collaborative augmented reconstruction (CAR) platform for neuron reconstruction at scale. This platform allows for immersive interaction and efficient collaborative editing of neuron anatomy using a variety of devices, such as desktop workstations, virtual reality headsets and mobile phones, enabling users to contribute anytime and anywhere and to take advantage of several AI-based automation tools. We tested CAR's applicability for challenging mouse and human neurons toward scaled and faithful data production.
从微观图像中对单个神经元的复杂 3D 形态进行数字化重建，无论是在个体实验室还是专注于细胞类型和大脑解剖结构的大规模项目中，都是一项至关重要的挑战。在传统的手动重建和最先进的基于人工智能（AI）的自动重建算法中，这一任务常常失败。组织多个神经解剖学家生成和交叉验证大规模数据生成中具有生物学意义且相互一致的重建，也是一项挑战。我们基于人工智能增强的协作群体智能，开发了一个用于大规模神经元重建的协作增强重建（CAR）平台。该平台允许使用各种设备（如台式工作站、虚拟现实耳机和移动电话）进行沉浸式交互和高效的神经元解剖协同编辑，使用户能够随时随地进行协作，并利用多种基于 AI 的自动化工具。我们测试了 CAR 在处理具有挑战性的小鼠和人类神经元时的适用性，以实现规模化和真实数据的生成。
