Direct determination of glucose, lactate and triglycerides in blood serum by a tunable quantum cascade laser-based mid-IR sensor

Direct determination of glucose, lactate and triglycerides in blood serum by a tunable quantum cascade laser-based mid-IR sensor
复制标题

DOI:
10.1007/s00340-012-5080-z
复制
发表时间:
2013-02-01
影响因子:
2.1
通讯作者:
Lendl, B.
Lendl, B.
中科院分区:
物理与天体物理3区
文献类型:
--
作者:
Brandstetter, M.;Volgger, L.;Lendl, B.

文献摘要

被引文献

相似文献

这项工作报告了一种紧凑型传感器,基于可调谐(1030-1230 cm(-1))外腔量子级联激光器(EC-QCL),可快速且无需试剂地即时测定血清中的葡萄糖、乳酸和甘油三酯。为了简单而稳健的操作,设计了单光束装置,并且所使用的激光器和探测器仅使用热电冷却。全面的计算机分析控制(包括液体处理和数据分析)促进了常规测量。高光程(> 100 μm)是临床实践中可靠测量的先决条件。因此,从理论和实验上考虑了水溶液中传输测量的最佳光程长度。实验确定的 QCL 血液传感器的最大信噪比 (SNR) 约为 140 μm,而采用液氮冷却碲化汞镉 (MCT) 检测器的标准 FT-IR 光谱仪则约为 50 μm。通过使用偏最小二乘 (PLS) 回归分析,使用单一吸收光谱同时计算分析物浓度。在 42 名不同患者中测定血清中的葡萄糖,预测误差 (RMSEP) 为 6.9 mg/dl,甘油三酯的交叉验证误差 (RMSECV) 为 17.5 mg/dl。在加标血清样品中,乳酸浓度可通过 RMSECV 8.9 mg/dl 测定。
This work reports on a compact sensor for fast and reagent-free point-of-care determination of glucose, lactate and triglycerides in blood serum based on a tunable (1030-1230 cm(-1)) external-cavity quantum cascade laser (EC-QCL). For simple and robust operation a single beam set-up was designed and only thermoelectric cooling was used for the employed laser and detector. Full computer control of analysis including liquid handling and data analysis facilitated routine measurements. A high optical pathlength (> 100 mu m) is a prerequisite for robust measurements in clinical practice. Hence, the optimum optical pathlength for transmission measurements in aqueous solution was considered in theory and experiment. The experimentally determined maximum signal-to-noise ratio (SNR) was around 140 mu m for the QCL blood sensor and around 50 mu m for a standard FT-IR spectrometer employing a liquid nitrogen cooled mercury cadmium telluride (MCT) detector. A single absorption spectrum was used to calculate the analyte concentrations simultaneously by using a partial-least-squares (PLS) regression analysis. Glucose was determined in blood serum with a prediction error (RMSEP) of 6.9 mg/dl and triglycerides with an error of cross-validation (RMSECV) of 17.5 mg/dl in a set of 42 different patients. In spiked serum samples the lactate concentration could be determined with an RMSECV of 8.9 mg/dl.