期刊
ANALYTICAL METHODS
卷 13, 期 29, 页码 3249-3255出版社
ROYAL SOC CHEMISTRY
DOI: 10.1039/d1ay00491c
关键词
-
资金
- National Research Foundation of Korea [2019R1A2C2089239, 2021R1A2C4002700]
- National Research Foundation of Korea [2021R1A2C4002700, 2019R1A2C2089239] Funding Source: Korea Institute of Science & Technology Information (KISTI), National Science & Technology Information Service (NTIS)
This study introduces a label-free, simple, and high-efficiency breast cancer detection platform with multimodal biomarker analytic algorithms, demonstrating the potential for real-time diagnosis of breast cancer during surgery.
We report the development of a label-free, simple, and high efficiency breast cancer detection platform with multimodal biomarker analytic algorithms on a portable 785 nm Raman setup with an endoscopic Raman-lensed fiber optic probe. We propose a multimodal biomarker extraction algorithm (PCMA) implemented by combining a multivariate statistics principal component analysis (PCA) algorithm and a multivariate curve resolution-alternating least squares (MCR-ALS) computational model for extraction of the biomarker information hidden in Raman spectrochemical data. We show that the six Raman spectrochemical peaks at 1009, 1270, 1305/1443, 1658, and 1750 cm(-1) assigned to phenylalanine, amide III in proteins, CH2 deformation in lipids, amide I in proteins, and carbonyl, respectively, can be used as a biomarker for breast cancer diagnosis using the biomarker-dominated PCMA spectrochemical spectra of breast tissues. From 20 human breast tissues, the PCMA-linear discriminant analysis (PCMA-LDA) identification method achieved high classification performance with a sensitivity and specificity >99% along with an improvement of approximately 4.5% compared to the performance without the PCMA mixture analysis algorithm. Our label-free breast cancer detection method has the potential for clinical application to diagnose breast cancer in real-time during surgery.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据