Kwang-Suk Park

Detrended fluctuation analysis of EEG in sleep apnea using MIT/BIH polysomnography data

A number of natural time series including electroencephalogram (EEG) show highly non-stationary characteristics in their behavior. We analyzed the EEG in sleep apnea that typically exhibits non-stationarity and long-range correlations by calculating its scaling exponents. Scaling exponents of the EEG dynamics are obtained by analyzing its fluctuation with detrended fluctuation analysis (DFA), which is suitable for non-stationary time series. We found the mean scaling exponents of EEG is discriminated according to Non-REM, REM (Rapid Eye Movement) and waken stage, and gradually increased from stage 1 to stage 2, 3 and 4.

Air mattress sensor system with balancing tube for unconstrained measurement of respiration and heart beat movements

The cardio-respiratory signal is a fundamental vital sign used for assessment of a patients status. Additionally, the cardio-respiratory signal provides a great deal of information to healthcare providers wishing to monitor healthy individuals. The air mattress sensor system allows the measurement of the respiration and heart beat movements without the use of a harness or sensor on the subjects body, which eliminates the difficulties these pose for long term measurements. In order to increase the sensitivity, a differential measurement technique between two air cells was used. The concept of a balancing tube between two air cells is suggested in order to increase the robustness against postural changes during the measurements. With this balancing tube, the meaningful frequency range could be selected using a pneumatic method. A mathematical model was constructed and validation experiments were performed for step and sinusoidal input signals. This technique was applied to measurements of respiration and heart beat movements in the supine posture on the bed, which showed potential for applications in sleep analysis, unconstrained healthcare monitoring and neonate monitoring.

Anti-inflammatory and Immune-regulatory Effects of Subcutaneous Perillae Fructus Extract Injections on OVA-induced Asthma in Mice

Perillae fructus (perilla seed) is a traditional medicinal herb used to treat bronchial asthma in Oriental medical clinics. ST36 is one of the most widely used acupuncture points, particularly for immune system regulation. Injection of an herbal extract into an acupuncture point (herbal acupuncture) is a therapeutic technique combining both acupuncture and herbal treatment. Perillae fructus extract was injected subcutaneously (Perillae fructus herbal acupuncture; PF-HA) at acupoint ST36 of OVA-induced asthmatic mice. The lung weight, bronchoalveolar fluid (BALF) cell count, the number of CCR3+, CD11b+, CD4+ and CD3e+/CD69+ cells in the lung, and the level of IgE, IL-4, IL-5 and IL-13 in BALF and serum were then measured. RT-PCR was used to measure the mRNA expression of IL-4, IL-5, IL-13 and TNF-α in the lung. Lung sections were analyzed histologically. PF-HA significantly reduced lung weight, the number of inflammatory cells in the lung and BALF, the levels of IgE and Th2 cytokines in BALF and serum, mRNA expression of Th2 cytokines in the lung, and pathological changes in lung tissue. Our results suggest that PF-HA may have an anti-inflammatory and immune-regulatory effect on bronchial allergic asthma by restoring the Th1/Th2 imbalance in the immune system and suppressing eosinophilic inflammation in airways.

Hybrid neural-network and rule-based expert system for automatic sleep stage scoring

In order to increase the performance of automatic sleep stage scoring, we propose a hybrid neural-network and expert system taking advantages of each system. After signal cleaning and feature extraction from polysomnographic signals using several algorithms we suggested, the rule-based expert system classified the sleep states with symbolic reasoning. The neural network supplemented the shortcomings of rule-based system by dealing with exceptions of rules. The result shows that the combination of computational and symbolic intelligence is promising approach to automatic sleep signal analysis.

Chaotic and periodic heart rate dynamics in uncomplicated intrauterine growth restricted fetuses

We studied how chaotic and periodic heart rate dynamics differ between normal fetuses (n = 192) and uncomplicated intrauterine growth restricted fetuses (n = 86), aged 31-42 weeks of gestation. We analyzed each fetal heart rate time series for 25 min. We quantified the chaotic dynamics of each fetal heart rate time series by correlation dimension. The periodic dynamics were analyzed by power spectral analysis. The correlation dimension and, therefore, the complexity, of the heart rate dynamics of the uncomplicated intrauterine growth restricted fetuses was significantly lower than that of the normal fetuses, which was marked at 38-42 weeks of gestation. The low-frequency (0.04-0.15 Hz) component and, therefore, the periodicity of the low-frequency range was significantly higher than that of the normal fetuses during all the gestational weeks. These results mean that, although the intrauterine growth restricted fetuses are not severely compromised, the overall integrity of their cardiovascular control is impaired, especially at term; and sympathetic modulation is increased, both of which may contribute to increased perinatal mortality.

Effects of phenylephrine on ciliary beat in human nasal respiratory epithelium : Quantitative measurement by video-computerized analysis

The in vitro effects of phenylephrine solution on ciliary beat frequency (CBF) in terms of different concentrations and exposure times were investigated using a video‐computerized analysis technique. Nasal epithelial cells were taken from inferior turbinate of 10 volunteers by scraping the nasal mucosa with a cytology brush. CBF was measured in five different concentrations including 0.125%, 0.25%, 0.5%, 1%, and 2.5%. Each specimen was incubated in different solution for 6 days and CBF was measured at 30 minutes, 1 hour, 2 hours, 4 hours, 8 hours, 12 hours, 1 day, 2 days, 3 days, 4 days, 5 days, and 6 days. CBF decreased significantly after 12‐hour incubation in 0.125% phenylephrine solution and after 8‐hour incubation in 0.25% phenylephrine solution, both clinically used concentrations. There were significant decreases in CBF after incubation in 0.5% phenylephrine for 2 hours, in 1% for 1 hour, and in 2.5% for 30 minutes (P < 0.05, repeated measure analysis of variance [ANOVA]). CBF of the nasal respiratory ciliated cells significantly decreased with increasing concentrations of phenylephrine solution and with increasing incubation times at the same concentration (P < 0.05, repeated measure ANOVA). The results of this study suggest that phenylephrine may inhibit ciliary beat in vitro by its pharmacological effect at lower concentrations than clinically used ones.

Adaptive EMG-driven communication for the disabled

We suggest a communication method for severely disabled persons, who have lost both mobility and speech, and their family using Morse code derived by Masseter muscle EMG. We developed a portable system that comprises EMG amplifier, A/D conversion, text-to-speech module, remote control module and serial communication to the host system. After training, the patient can make speech by composing Morse code by moving his/her chin. Calibration and remote control mode is supported. It also supports the adaptive encoding method for fatigue.

Classification of glucose concentration in diluted urine using the low-resolution Raman spectroscopy and kernel optimization methods

In order to detect minute amounts of glucose in diluted urine, we applied the Raman spectroscopy method. To simulate abnormal diluted urine in a toilet bowl, we diluted normal urine ten-fold with water and added glucose up to 8 mg dl(-1). Data were collected using a low-resolution Raman spectrometer that was preprocessed with the optimizing kernel method. We also applied the neural network algorithm to classify abnormal and normal urine samples according to their glucose concentrations. The kernel optimizing method was very effective in the classification of the tested subjects as it increased the accuracy of classification by 92%. This method suggests the possibility of caring for patients by daily monitoring their urine components in a manner non-invasive to ordinary life.

Directional disorder of ciliary metachronal waves using two-dimensional correlation map

The interrelationship of cilia and the order of wave directions are important factors that determine the effectiveness of cilia to transport materials in mucociliary systems of the respiratory tract. The interrelationship of cilia and the directional disorder of the ciliary metachronal wave were analyzed using digital microscopic images. The degree of synchronization between ciliary beats was determined by the correlation factor between two different spots. To find out the uniphase directions of beating cilia, principal axes of inertia were applied to the two-dimensional correlation map calculated from sequential ciliary images. The standard deviation of determined wave directions in a region of interest (ROI) was defined as a measure of metachronal wave disorder. The pooled mean of metachronal wave disorder was 23.4 /spl plusmn/ 8.79/spl deg/ in ROIs of 8 /spl mu/m /spl times/ 8 /spl mu/m and 25.4 /spl plusmn/ 6.46/spl deg/ in 32 /spl mu/m /spl times/ 24 /spl mu/m from the sphenoid sinus mucosa of five normal subjects. Our result shows that there is a considerable variation in metachronal wave directions of cilia beating on the epithelium.

A study on the elimination of the ECG artifact in the polysomnographic EEG and EOG using AR model

We present the method of the elimination of ECG artifact from the polysomnographic EEG and EGG. The idea of this method is that the ECG-corrupted EEG segment can be detected from ECG R-wave and regarded as a missing segment. After this, we used two interpolations to recover the missing segment. One is the Lagrange polynomial interpolation and the other is the least square error AR interpolation. We also compared the AR-method with the LMS adaptive noise canceling method. Simulations show the AR-method performs better than other methods. To apply to the real EEG and EOG signals practically, we also developed the algorithm to detect whether the artifact level is high or not. If the artifact level is high, then the interpolations are applied.