An image-based AI tool to identify stiffness- or age-related mechanotransduction abnormalities in vascular smooth muscle cells

一种基于图像的人工智能工具，用于识别血管平滑肌细胞中与硬度或年龄相关的机械转导异常

• 批准号: BB/Y513994/1
• 负责人: Sanjay Sinha
• 金额: $ 32.87万
• 依托单位: University of Cambridge
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2024
• 资助国家: 英国
• 起止时间: 2024 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=BB%2FY513994%2F1
• 关键词: image; based; AI; tool; identify

In this project we will build an artificial intelligence (AI)-based image analysis tool called "MechanoAI-tool" trained to identify changes in vascular smooth muscle cell (SMC) shape and structure in response to changes in the stiffness of the surface the cells are plated on.Being able to easily identify these changes in SMC shape is important because vascular SMCs are the cell types that make up the walls of blood vessels and control blood flow to our organs by contracting or relaxing. This contraction or relaxation depends on a number of signals which includes the mechanical factors acting on the cell such as stiffness or stretch. However, as we age, our vessels become stiffer and the way vascular SMCs respond changes, which together can impact blood flow to vital organs.Measuring how vascular SMCs respond to stiffness changes is a good indicator of ageing. However, many methods for studying stiffness responses require complex genetic techniques and expensive microscopes which act as barriers to entry for many research groups.We think that image analysis of vascular SMC responses under simple phase contrast microscopy, which is cheap and widely available, will provide an accessible, and robust way to test whether any given SMCs respond appropriately to stiffness changes or not. However, interpreting the changes on a simple phase contrast microscope is not easy and we will therefore develop an AI-based image analysis tool to identify normal vs abnormal responses quickly and reliably.These findings and the Mechano-AI-tool that we develop could be used to predict biological properties of vascular SMCs by any research group world-wide and have applications in SMC biology, ageing and vascular disease research.

在这个项目中，我们将构建一个名为“MechanoAI-tool”的基于人工智能 (AI) 的图像分析工具，经过训练可识别血管平滑肌细胞 (SMC) 形状和结构的变化，以响应细胞表面硬度的变化能够轻松识别 SMC 形状的这些变化非常重要，因为血管 SMC 是构成血管壁并通过收缩或舒张控制流向我们器官的血液的细胞类型。这种收缩或松弛取决于许多信号，其中包括作用于细胞的机械因素，例如僵硬或拉伸。然而，随着年龄的增长，我们的血管变得更僵硬，血管 SMC 的反应方式也会发生变化，这些因素共同影响流向重要器官的血液。测量血管 SMC 对硬度变化的反应是衰老的一个很好的指标。然而，许多研究僵化反应的方法需要复杂的遗传技术和昂贵的显微镜，这成为许多研究小组进入的障碍。我们认为，在简单的相差显微镜下对血管 SMC 反应进行图像分析，这是便宜且广泛使用的，将提供一种可访问且可靠的方法来测试任何给定的 SMC 是否对刚度变化做出适当的响应。然而，在简单的相差显微镜上解释这些变化并不容易，因此我们将开发一种基于人工智能的图像分析工具，以快速可靠地识别正常反应与异常反应。这些发现和我们开发的 Mechano-AI 工具可以是世界各地的任何研究小组都用来预测血管 SMC 的生物学特性，并在 SMC 生物学、衰老和血管疾病研究中得到应用。