Monitoring blood hemoglobin level is essential to diagnose anaemia disease. This study aims to evaluate the capability of an artificial neural network (ANN) and Savitzky Golay (SG) pre-processing in predicting the blood hemoglobin level based on the near-infrared spectrum. The effects of the hidden neuron number and different SG pre-processing strategies were examined and discussed. ANN coupled with first order SG derivative and five hidden neurons achieved better prediction performance with root mean square error of prediction of 0.3517 g/dL and coefficient determination of prediction of 0.9849 compared to the previous studies. Results depict that ANN that coupled with first order SG derivative could improve near-infrared spectroscopic analysis in predicting blood hemoglobin level, and the proposed nonlinear model outperforms linear models without variable selections. This finding suggests that the modelling strategy is promising in establishing a better relationship between the blood hemoglobin and near-infrared spectral data.
Near infrared spectroscopy is a non-invasive and optical technology that uses the relative absorption of near infrared light (i.e. 780 to 2500nm). Near infrared spectroscopy has been successfully applied in the evaluation of food quality [1], [2]. Near infrared (NIR) sensing technology has been widely implemented in various areas e.g. medical [3]-[7], agriculture [2], [8]-[14], and industries [15]-[17] replace conventional wet chemistry analysis due to its non-Abstract: Near infrared (NIR) sensing technology has been widely implemented in various areas as an alternative to substitute conventional wet chemistry analysis and sensing applications due to its non-invasive, green, and rapid measurement features. Recent researches indicate that a combination of NIR light emitted diode (LEDs) and photodiodes is promising to reduce the financial barriers to carry out NIR research in various applications. However, there is a challenge to detect and remove unwanted signals particularly ambient light and the changes of surrounding. This is because NIR signals are susceptible to the change of temperatures, moisture, and types of samples. This is worth to highlight that camera technology is feasible to remove unwanted backgrounds and insusceptible to the background for various applications e.g. face recognition, unmanned vehicle systems, and object classification. Therefore, this study aims to investigate the feasibility of a complementary metal-oxidesemiconductor (CMOS) camera in developing a shortwave NIR spectroscopy. Firstly, a slit, a NIR grating, and a CMOS camera were positioned and shielded in a black aluminum chassis. A total of six different parameters of the camera were investigated in this study, i.e. the exposure level, gain, white balance, brightness, sharpness, and saturation. Findings show that the CMOS camera with the optimal values of the exposure level and the gain could produce a good quality of NIR spectrum compared with default settling, in which, the signals were saturated. Thus, CMOS camera is feasible to be used to develop a low-cost NIR spectroscopy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.