Lee Sanghwa, Kwon Hyunhee, Oh Jeongmin, Kim Kyeong Ryeol, Hwang Joonseup, Kang Suyeon, Lee Kwanhee, Namgoong Jung-Man, Kim Jun Ki
Biomedical Engineering Research Center, Asan Medical Center, Seoul, 05505, Republic of Korea.
Department of Surgery, Asan Medical Center, University of Ulsan College of Medicine, Seoul, 05505, Republic of Korea.
Int J Nanomedicine. 2025 May 27;20:6743-6755. doi: 10.2147/IJN.S497900. eCollection 2025.
While liver transplantation saves lives from irreversible liver damage, it poses challenges such as graft dysfunction due to factors like ischemia-reperfusion (IR) injury, which can lead to significant cellular damage and systemic complications. Current diagnostic tools for detecting IR injury have limitations, necessitating advanced methods for timely intervention. This study explores the integration of surface-enhanced Raman spectroscopy (SERS) with artificial intelligence (AI) to improve diagnostic accuracy for liver IR injury.
IR injury was induced using a mouse model, and histopathological and hepatic functional evaluations were conducted alongside SERS measurements. Raman signals obtained via SERS chips, which selectively filter nano-biomarkers and enhance signals, were analyzed using machine learning algorithms.
The PC-LDA derived from spectra achieved an accuracy of 93.13%, while a machine learning algorithm based on PC-derived PLS-DA improved accuracy to 98.75%.
Our findings emphasize the potential of combining SERS with AI to detect and specifically identify dysfunction due to liver damage early, potentially advancing patient management in liver transplantation.
肝移植虽能挽救因不可逆肝损伤而危及的生命，但它也带来了一些挑战，比如因缺血再灌注（IR）损伤等因素导致的移植物功能障碍，这可能会造成严重的细胞损伤和全身并发症。当前用于检测IR损伤的诊断工具存在局限性，因此需要先进的方法以便及时进行干预。本研究探索将表面增强拉曼光谱（SERS）与人工智能（AI）相结合，以提高对肝脏IR损伤的诊断准确性。
利用小鼠模型诱导IR损伤，并在进行SERS测量的同时进行组织病理学和肝功能评估。通过选择性过滤纳米生物标志物并增强信号的SERS芯片获得的拉曼信号，使用机器学习算法进行分析。
从光谱中得出的主成分线性判别分析（PC-LDA）准确率达到了93.13%，而基于主成分的偏最小二乘判别分析（PC-derived PLS-DA）的机器学习算法将准确率提高到了98.75%。
我们的研究结果强调了将SERS与AI相结合以早期检测并特异性识别肝损伤所致功能障碍的潜力，这可能会推动肝移植患者的管理。
