Cheol Soo Park

Urine Analysis in Diluted Situation Using Low-Resolution Raman Spectroscope

We could get valuable information about various diseases by analyzing urine of patients. For the diabetics, it has been very important to know the glucose concentration in their urine. Until now, there have been many methods to estimate the concentration but they were invasive and annoyed patients. The low-resolution Raman spectroscopy (LRRS) could be an alternative method to be non-invasive and unconscious for patients who want to check the amount of their urine components. LRRS is not expensive and smaller than laboratory Raman spectroscope. And we want to attach it to the common toilet bowl. For the reason, we got the spectrum of diluted urine and predicted the glucose concentration. In addition, we tried to find very small amount of glucose. LRRS was adequate for measuring Raman signal of diluted urine and could find very small amount of glucose. This system will realize the observation of ones health condition every day without awareness

Application of independent component analysis (ICA) method to the Raman spectra processing

Independent component analysis (ICA) method was applied as a processing step for Raman spectra. 136 Raman spectra were acquired from urine samples from 18 subjects. Each spectrum was acquired from different sample. 785nm, 100mW (at sample) laser with 2048 element linear silicon TE cooled CCD were used. In order to separate information of glucose, creatinine, urea nitrogen, uric acid and invaluable information from the urine spectrum, ICA by Maximum Likelihood (ML) fast fixed-point estimation algorithm was applied. By looking for maximum likelihood, independent information could be separated from the urine spectra. Among separated information, high frequency noise which could be generated by ambient noise and low frequency noise which contain information of baseline shift were observed. Additionally, peak information of each component was observed. The processing time was very short because fast fixed point algorithm was added to ML estimation method. Before applying ICA, all spectra were mean centered in order to enhance the peak information. In addition, all spectra were pre-processed to have unit variance in order to shorten calculation time. This first study about applying ICA suggested that this algorithm can be used as a pattern recognition algorithm to extract information from Raman spectra. Additionally, because ICA can provide information with statistical independency sufficiently, further studies about ICA which can substitute PCA will be performed.

Machine learning vs. hybrid machine learning model for optimal operation of a chiller

This article compares two modeling approaches for optimal operation of a turbo chiller installed in an office building: (1) a machine learning model developed with artificial neural network (ANN) and (2) a hybrid machine learning model developed with the ANN model and available physical knowledge of the chiller. Before developing the ANN model of the chiller, the authors used Gaussian mixture model in order to check the validity of measured data. Then, the hybrid model was developed by combining the ANN model and physics-based regression equations from the EnergyPlus engineering reference. It was found that both the ANN and hybrid ANN model are satisfactory to predict the chiller’s power consumption: mean bias error (MBE) = −2.63%, coefficient of variation of the root mean square error (CVRMSE) = 8.05% by the ANN model; MBE = −3.99%, CVRMSE = 11.98% by the hybrid ANN model. However, the hybrid model requires fewer inputs (four inputs) than the ANN model (eight inputs). The energy savings of both models are similar coefficient of performance (COP) = 4.32 by the optimal operation of the ANN model; COP = 4.44 by the optimal operation of the hybrid ANN model. In addition, the hybrid ANN model can be applied where the ANN model is unable to provide accurate predictions.