生物医学工程研究
生物醫學工程研究
생물의학공정연구
JOURNAL OF BIOMEDICAL ENGINEERING RESEARCH
2013年
3期
153-157
,共5页
近红外光谱%偏最小二乘%卷积平滑%交叉验证%预测残差平方和
近紅外光譜%偏最小二乘%捲積平滑%交扠驗證%預測殘差平方和
근홍외광보%편최소이승%권적평활%교차험증%예측잔차평방화
Near-infrared spectroscopy%Partial least squares%Convolution smoothing%Cross-validation%Predicted Residual Error Sum of Squares
葡萄糖浓度测量是无创检测人体血糖浓度的基础研究,具有重要的科学意义。本研究利用近红外光谱技术(NIRS)和回归定标方法相结合测定葡萄糖水溶液浓度。实验采用傅里叶光谱仪对葡萄糖溶液的近红外透射光谱进行采集,利用光谱数据建立回归模型,讨论偏最小二乘法(PLS )校正模型中主因子数的确定方法,并对不同预处理方法对PLS模型建立影响进行了研究。采用Savitzky-Golay 卷积平滑对溶液样品吸光度光谱进行预处理,通过交叉验证方法求取预测残差平方和(PRESS),葡萄糖溶液浓度预测值的最大相对误差限制在了8%以内。研究结果表明,近红外透射法用于葡萄糖水溶液浓度测量能够达到一定的测试精度,具有进一步的研究价值。
葡萄糖濃度測量是無創檢測人體血糖濃度的基礎研究,具有重要的科學意義。本研究利用近紅外光譜技術(NIRS)和迴歸定標方法相結閤測定葡萄糖水溶液濃度。實驗採用傅裏葉光譜儀對葡萄糖溶液的近紅外透射光譜進行採集,利用光譜數據建立迴歸模型,討論偏最小二乘法(PLS )校正模型中主因子數的確定方法,併對不同預處理方法對PLS模型建立影響進行瞭研究。採用Savitzky-Golay 捲積平滑對溶液樣品吸光度光譜進行預處理,通過交扠驗證方法求取預測殘差平方和(PRESS),葡萄糖溶液濃度預測值的最大相對誤差限製在瞭8%以內。研究結果錶明,近紅外透射法用于葡萄糖水溶液濃度測量能夠達到一定的測試精度,具有進一步的研究價值。
포도당농도측량시무창검측인체혈당농도적기출연구,구유중요적과학의의。본연구이용근홍외광보기술(NIRS)화회귀정표방법상결합측정포도당수용액농도。실험채용부리협광보의대포도당용액적근홍외투사광보진행채집,이용광보수거건립회귀모형,토론편최소이승법(PLS )교정모형중주인자수적학정방법,병대불동예처리방법대PLS모형건립영향진행료연구。채용Savitzky-Golay 권적평활대용액양품흡광도광보진행예처리,통과교차험증방법구취예측잔차평방화(PRESS),포도당용액농도예측치적최대상대오차한제재료8%이내。연구결과표명,근홍외투사법용우포도당수용액농도측량능구체도일정적측시정도,구유진일보적연구개치。
The measurement of glucose concentration is the basis for the noninvasive detection of blood glucose concentration, which is significant in scientific research.In the present study,Near Infrared Spectrum(NIRS)and regression analysis were combined to measure the glucose concentration. The spectrum of glucose solutions was obtained with the Fourier Transformed Infrared Spectrometer,and the data was used for further regression analysis.In this study,the methodology of determining partial least squares (PLS)components and spectral preprocessing were discussed.During PLS modeling,using Savitzky -Golay and the Prediction Residual Error Sum of Squares (PRESS),we achieved the maximum relative error within 8%.The experiment results demonstrat that NIRS has the potential to the measurement of glucose solution and noninvasive detecting of blood glucose.