Gao Siyu, Wang Jingjing, Miao Zeyu, Zhao Xudong, Zhang Ying, Du Wei, Feng Xiaojun, Li Yiwei, Liu Jinzhi, Chen Peng, Liu Bi-Feng
The Key Laboratory for Biomedical Photonics of MOE at Wuhan National Laboratory for Optoelectronics-Hubei Bioinformatics & Molecular Imaging Key Laboratory, Systems Biology Theme, Department of Biomedical Engineering, College of Life Science and Technology, Huazhong University of Science and Technology, Wuhan, 430074, China.
Shenzhen YHLO Biotech Co., Ltd., Shenzhen, Guangdong, 518116, China.
Talanta. 2025 May 15;287:127619. doi: 10.1016/j.talanta.2025.127619. Epub 2025 Jan 23.
Cysteamine (CA) serves as a cystine-depleting agent employed in the management of cystinosis and a range of other medical conditions. Monitoring blood CA levels at the point of care is imperative due to the risk of toxicity associated with elevated CA dosages. An additional significant challenge is presented by the intricate composition of human plasma and the presence of various interfering biological thiols, which possess similar structures or properties. Here, this work proposes an AI-enhanced Lab-on-a-disc system, also termed AI-LOAD, for multiplexed point-of-care testing of cysteamine. The AI-LOAD system incorporates an online whole blood separation mechanism alongside a naked-eye colorimetric detection module, facilitating the rapid and precise visual identification of cysteamine. Remarkably, the system necessitates only 40 μL of whole blood to analyze eight samples within 3-min, achieving a limit of detection as low as 10 μM, which is lower than the physiological toxic concentration of 0.1 mM. By leveraging diverse colorimetric responses generated through interactions between gold nanoparticles of varying sizes and different biological thiols, combined with artificial intelligence methodologies, the system successfully accomplished specific recognition of various biological thiols with 100 % accuracy. The proposed AI-LOAD will drive advancements in centrifugal microfluidics for point-of-care testing, thereby holding potential for broader applications in future biomedical research and in vitro diagnosis.
半胱胺（CA）是一种用于治疗胱氨酸病及一系列其他病症的耗胱氨酸剂。由于高剂量CA存在毒性风险，因此在护理点监测血液中CA水平至关重要。人类血浆的复杂成分以及各种具有相似结构或性质的干扰性生物硫醇的存在带来了另一个重大挑战。在此，这项工作提出了一种人工智能增强的盘上实验室系统，也称为AI-LOAD，用于半胱胺的多重即时检测。AI-LOAD系统结合了在线全血分离机制和肉眼比色检测模块，便于快速、精确地目视识别半胱胺。值得注意的是，该系统仅需40μL全血即可在3分钟内分析8个样本，检测限低至10μM，低于0.1mM的生理毒性浓度。通过利用不同大小的金纳米颗粒与不同生物硫醇之间相互作用产生的多种比色响应，并结合人工智能方法，该系统成功实现了对各种生物硫醇的100%准确特异性识别。所提出的AI-LOAD将推动用于即时检测的离心微流控技术的发展，从而在未来生物医学研究和体外诊断中具有更广泛应用的潜力。
