Scholarlink

<p>This paper will introduce the design of a gold nanorod (AuNR) localized surface plasmon resonance (LSPR) microfluidic biosensor in combination with machine learning to detect sepsis biomarkers rapidly and reliably. High sensitivity (≈180&#xa0;nm/RIU) was also determined by observing the stable longitudinal LSPR peak at 750&#xa0;nm and differences in the refractive index between 1.330 and 1.360, giving a linear shift of wavelength at 750&#xa0;nm to 761&#xa0;nm. The function of surface underwent functionalization with self-assembled monolayer and EDC/NHS chemistry to successfully immobilize the antibody, which leads to an increase in the resonance wavelength, 750&#xa0;nm to 788&#xa0;nm, with biomarker binding and proves the specific interaction of the molecules. Microfluidic integration gave the ability to deliver samples at a controlled rate and optimal results at 10 µL/min were obtained to achieve the highest response rate of 795&#xa0;nm. Drift correction enhanced the stability of signal level by eliminating the variation in wavelength between 750-759&#xa0;nm to 750-752&#xa0;nm over a period of 30&#xa0;min. The sensor detected the concentration of biomarkers ranging between 0.1 and 100&#xa0;ng/mL and gave a measurable wavelength variation of 751 to 798&#xa0;nm which was a good indication of sensitivity and detection limit. The machine learning analysis also increased the accuracy of classification to 95% against 78%, and reliable biomarker identification became possible. Optimization of the system has improved the signal to noise ratio by 15 to 48 which has enhanced detection reliability. The findings of these studies support the fact that the AuNR-LSPR microfluidic system achieved sensitivity, stability and immediate reaction to parameters and has high capability to bi-directional clinical diagnosis and prompt sepsis in real-time.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A spectrotemporal drift-corrected gold nanorod LSPR microfluidic biosensing framework for machine learning–based early sepsis transition detection

  • 期刊论文

  • 发表日期：

    2026年5月20日

查看全文
  • Faris A. Almalki,
  • Kholod D. Alsufiani

摘要

This paper will introduce the design of a gold nanorod (AuNR) localized surface plasmon resonance (LSPR) microfluidic biosensor in combination with machine learning to detect sepsis biomarkers rapidly and reliably. High sensitivity (≈180 nm/RIU) was also determined by observing the stable longitudinal LSPR peak at 750 nm and differences in the refractive index between 1.330 and 1.360, giving a linear shift of wavelength at 750 nm to 761 nm. The function of surface underwent functionalization with self-assembled monolayer and EDC/NHS chemistry to successfully immobilize the antibody, which leads to an increase in the resonance wavelength, 750 nm to 788 nm, with biomarker binding and proves the specific interaction of the molecules. Microfluidic integration gave the ability to deliver samples at a controlled rate and optimal results at 10 µL/min were obtained to achieve the highest response rate of 795 nm. Drift correction enhanced the stability of signal level by eliminating the variation in wavelength between 750-759 nm to 750-752 nm over a period of 30 min. The sensor detected the concentration of biomarkers ranging between 0.1 and 100 ng/mL and gave a measurable wavelength variation of 751 to 798 nm which was a good indication of sensitivity and detection limit. The machine learning analysis also increased the accuracy of classification to 95% against 78%, and reliable biomarker identification became possible. Optimization of the system has improved the signal to noise ratio by 15 to 48 which has enhanced detection reliability. The findings of these studies support the fact that the AuNR-LSPR microfluidic system achieved sensitivity, stability and immediate reaction to parameters and has high capability to bi-directional clinical diagnosis and prompt sepsis in real-time.