Whether in the monitoring of critically ill patients such as shock, respiratory failure, brain injury, or in major anesthesia surgeries, it is necessary to evaluate the patient’s pO<sub>2</sub> and pH. An...Whether in the monitoring of critically ill patients such as shock, respiratory failure, brain injury, or in major anesthesia surgeries, it is necessary to evaluate the patient’s pO<sub>2</sub> and pH. An optical fiber sensor presented is capable of monitoring the presence of oxygen partial pressure (pO<sub>2</sub>) and pH in the real-time. The sensor is based on fluorescence sensing of polymer immobilized in the oxygen/pH-sensitive membranes and covalently attached to the optical fiber probe. The design of this sensor uses LED as light source, which is an excitation light source, inducing specific wavelengths of fluorescence on the oxygen/pH-sensitive membrane. The intensity and lifetime of fluorescence are related to the pO<sub>2</sub> and pH. So the pO<sub>2</sub> and pH can be measured by the relationship between the pO<sub>2</sub>/pH values and the intensity and lifetime of fluorescence. The signal conditioning system based on DSP and STM32 was used to store and process data, and display test values. The response of the sensor for pO<sub>2</sub> and pH monitoring with nitrogen (N<sub>2</sub>) as a balancing gas in the laboratory was performed. Finally, the oxygen/pH sensing scheme presented in this work is intended for using in biological, medical and environmental applications.展开更多
With the development of mid-infrared (MIR) photoelectric devices, mid-infrared spectroscopy has become one of the important methods for non-invasive detection of blood glucose. The mid-infrared region (4000 - 400 cm&l...With the development of mid-infrared (MIR) photoelectric devices, mid-infrared spectroscopy has become one of the important methods for non-invasive detection of blood glucose. The mid-infrared region (4000 - 400 cm<sup>-1</sup>) has the well-known fingerprint region (1200 - 800 cm<sup>-1</sup>) of glucose, which has clearer characteristic absorption peaks and better specificity. There is a lot of molecular information about glucose in the MIR. The non-invasive detection of blood glucose by mid-infrared spectroscopy needs to achieve certain accuracy, and the quantitative model is an important factor affecting the accuracy of glucose detection. In this paper, the samples of imitation solution containing only glucose and the samples of imitation mixed solution are taken as the research objects, and the mid-infrared spectral data of the samples are collected. The full spectrum partial least squares Regression (PLSR) model, SNV + Ctr-PLSR model, MSC + Ctr-PLSR model, and convolutional neural networks (CNN) model of 3000 - 900 cm<sup>-1</sup> band were constructed. Full spectrum PLS model and CNN model of 1200 - 900 cm<sup>-1</sup> band were constructed. The experimental results show that the optimal model of the two bands is CNN, then the correlation coefficient of prediction set (Rp) of 3000 - 900 cm<sup>-1</sup> band is 0.95, and the root mean square error of pre-diction set (RMSEP) value is 22.10. The Rp of 1200 - 900 cm<sup>-1</sup> band is 0.95, and the RMSEP value is 22.54. The research results show that CNN is a promising method, which has higher accuracy than PLSR, and is especially suitable for modeling human complex environment. In addition, the study provides a theoretical and practical basis for CNN in feature selection and model interpretation.展开更多
文摘Whether in the monitoring of critically ill patients such as shock, respiratory failure, brain injury, or in major anesthesia surgeries, it is necessary to evaluate the patient’s pO<sub>2</sub> and pH. An optical fiber sensor presented is capable of monitoring the presence of oxygen partial pressure (pO<sub>2</sub>) and pH in the real-time. The sensor is based on fluorescence sensing of polymer immobilized in the oxygen/pH-sensitive membranes and covalently attached to the optical fiber probe. The design of this sensor uses LED as light source, which is an excitation light source, inducing specific wavelengths of fluorescence on the oxygen/pH-sensitive membrane. The intensity and lifetime of fluorescence are related to the pO<sub>2</sub> and pH. So the pO<sub>2</sub> and pH can be measured by the relationship between the pO<sub>2</sub>/pH values and the intensity and lifetime of fluorescence. The signal conditioning system based on DSP and STM32 was used to store and process data, and display test values. The response of the sensor for pO<sub>2</sub> and pH monitoring with nitrogen (N<sub>2</sub>) as a balancing gas in the laboratory was performed. Finally, the oxygen/pH sensing scheme presented in this work is intended for using in biological, medical and environmental applications.
文摘With the development of mid-infrared (MIR) photoelectric devices, mid-infrared spectroscopy has become one of the important methods for non-invasive detection of blood glucose. The mid-infrared region (4000 - 400 cm<sup>-1</sup>) has the well-known fingerprint region (1200 - 800 cm<sup>-1</sup>) of glucose, which has clearer characteristic absorption peaks and better specificity. There is a lot of molecular information about glucose in the MIR. The non-invasive detection of blood glucose by mid-infrared spectroscopy needs to achieve certain accuracy, and the quantitative model is an important factor affecting the accuracy of glucose detection. In this paper, the samples of imitation solution containing only glucose and the samples of imitation mixed solution are taken as the research objects, and the mid-infrared spectral data of the samples are collected. The full spectrum partial least squares Regression (PLSR) model, SNV + Ctr-PLSR model, MSC + Ctr-PLSR model, and convolutional neural networks (CNN) model of 3000 - 900 cm<sup>-1</sup> band were constructed. Full spectrum PLS model and CNN model of 1200 - 900 cm<sup>-1</sup> band were constructed. The experimental results show that the optimal model of the two bands is CNN, then the correlation coefficient of prediction set (Rp) of 3000 - 900 cm<sup>-1</sup> band is 0.95, and the root mean square error of pre-diction set (RMSEP) value is 22.10. The Rp of 1200 - 900 cm<sup>-1</sup> band is 0.95, and the RMSEP value is 22.54. The research results show that CNN is a promising method, which has higher accuracy than PLSR, and is especially suitable for modeling human complex environment. In addition, the study provides a theoretical and practical basis for CNN in feature selection and model interpretation.
