Keo-Sik Kim

Non-invasive algorithm for bowel motility estimation using a back-propagation neural network model of bowel sounds

BackgroundRadiological scoring methods such as colon transit time (CTT) have been widely used for the assessment of bowel motility. However, these radiograph-based methods need cumbersome radiological instruments and their frequent exposure to radiation. Therefore, a non-invasive estimation algorithm of bowel motility, based on a back-propagation neural network (BPNN) model of bowel sounds (BS) obtained by an auscultation, was devised.MethodsTwelve healthy males (age: 24.8 ± 2.7 years) and 6 patients with spinal cord injury (6 males, age: 55.3 ± 7.1 years) were examined. BS signals generated during the digestive process were recorded from 3 colonic segments (ascending, descending and sigmoid colon), and then, the acoustical features (jitter and shimmer) of the individual BS segment were obtained. Only 6 features (J1, 3, J3, 3, S1, 2, S2, 1, S2, 2, S3, 2 ), which are highly correlated to the CTTs measured by the conventional method, were used as the features of the input vector for the BPNN.ResultsAs a results, both the jitters and shimmers of the normal subjects were relatively higher than those of the patients, whereas the CTTs of the normal subjects were relatively lower than those of the patients (p < 0.01). Also, through k-fold cross validation, the correlation coefficient and mean average error between the CTTs measured by a conventional radiograph and the values estimated by our algorithm were 0.89 and 10.6 hours, respectively.ConclusionsThe jitter and shimmer of the BS signals generated during the peristalsis could be clinically useful for the discriminative parameters of bowel motility. Also, the devised algorithm showed good potential for the continuous monitoring and estimation of bowel motility, instead of conventional radiography, and thus, it could be used as a complementary tool for the non-invasive measurement of bowel motility.

A new bio-impedance sensor technique for leg movement analysis

The paper provides a new approach for detection using bio-impedance. This impedance is measured by the four-electrode method. As impedance changes resulting from ankle, knee, and hip movements depended heavily on electrode placement, we determined the optimal electrode configurations for those movements by searching for high correlation coefficients, large impedance changes, and minimum interferences in ten subjects (age: 20/spl plusmn/4 years). Our optimal electrode configurations showed very strong relationships between the ankle joint angle and ankle impedance (/spl gamma/= -0.913/spl plusmn/0.03), between the knee joint angle and knee impedance (/spl gamma/= -0.944/spl plusmn/0.02), and between the hip joint angle and hip impedance (/spl gamma/=0.823/spl plusmn/0.08). The study showed the possibility that lower leg movement can be easily measured by an impedance measurement system with two pairs of skin-electrodes.

Awareness system for bowel motility estimation based on artificial neural network of bowel sounds

Awareness system of bowel motility estimation based on an artificial neural network (ANN) model of bowel sounds obtained by an auscultation was devised. Twelve healthy males and 6 patients with delayed bowel motility were examined. BS signals generated during the digestive process were recorded from 3 colonic segments (ascending, descending and sigmoid colon), and then, the acoustical features (jitter and shimmer) of the individual BS segment were obtained. Only 6 features (J1,3, J3,3, S1,2, S2,1, S2,2, S3,2) highly correlated to the conventional colon transit time (CTT) were used as the features. Through k-fold cross validation, the correlation coefficient and mean average error between the CTTs and the values estimated by our algorithm were 0.89 and 10.6 hours, respectively. The devised system showed good potential for the continuous monitoring and estimating the bowel motility, instead of conventional radiography, and thus, it could be used as an awareness tool for the non-invasive measurement of bowel motility.

An efficient lgorithm to improve feature extraction and classification of knee joint sound

Vibroarthrographic (VAG) signals, generated during the active flexion and extension of the leg, represent acoustic signals caused by joint vibration and can be used as useful indicators of the roughness, softening, breakdown, or the state of lubrication of the articular cartilage surface. This paper describes an efficient algorithm in order to improve the classification accuracy of the features obtained from the time-frequency distribution (TFD) of normal and abnormal VAG signals. VAG signals were correctly segmented by the dynamic time warping and the noise within the TFD of the segmented VAG signals was diminished by the singular value decomposition method. The classification of the knees as normal or abnormal was evaluated using a back-propagation neural network (BPNN). 1408 VAG segments (normal 1031, abnormal 377) were used for evaluating our devised algorithm by a BPNN and, consequently, the mean accuracy was 91.4±1.7%. This algorithm could help to enhance the performance of the feature extraction and classification of VAG signal.

Comparison of bio-impedance changes and EMG activity during daily events

A conventional method for measuring abdominal pressure involves the use of a fluid-filled rectal catheter. However, this method has some drawbacks, so it is limited to ambulatory urodynamics monitoring study. In this study, we proposed a novel method for estimating abdominal pressure in non invasive manner by changes in bio-impedance and electromyographic (EMG) signals. As a preliminary, we compared the bio-impedance changes and EMG activity during daily events, such as cough, sneeze and lumbar movement, while the abdominal pressure increased. The correlation coefficients between changes in bio-impedance and EMG signals, according to increases in abdominal contractions, were 0.72, 0.96, 0.90, 0.84 and 0.78 for the weak, strong, stronger, vigorous and maximal contraction, respectively. Also, daily activities such as coughing, sneezing and conversation, were sensitively monitored by measuring the change in amplitude of the bio-impedance signals, whilst EMG signals could not be used to detect these activities; therefore, the bio-impedance method is a more useful means of non-invasively measuring the changes in the abdominal pressure for urodynamics monitoring.

An new approach of ambulatory urodynamic system for measuring the abdominal pressure

We developed a system to perform fully ambulatory monitoring studies of the bladder and a new method of measuring the abdominal pressure noninvasively using the bio-impedance method. To verify this system, bladder and abdominal pressures and surface EMG signals were recorded during artificial filling cystometry. The system was evaluated in 10 AUM studies, whose mean duration was 11 minutes 42 seconds(standard deviation plusmn 4 minutes 16 seconds). Visual inspection of the recorded data showed that the subtraction of the abdominal trace from the bladder trace provided a useful detrusor trace when the subjects were either active or resting. Also, the amplitude of the bio-impedance signal was much higher when the subject coughed, sneezed or talked than when he or she was resting(p<0.05). Therefore, the bio-impedance method is a useful tool to measure the change in the abdominal pressure in an AUM study. We concluded that the proposed system can provide good quality recordings and should prove useful in future evaluations of natural filling urodynamics studies.

Enhanced optical coherence tomography imaging using a histogram-based denoising algorithm

Abstract. A histogram-based denoising algorithm was developed to effectively reduce ghost artifact noise and enhance the quality of an optical coherence tomography (OCT) imaging system used to guide surgical instruments. The noise signal is iteratively detected by comparing the histogram of the ensemble average of all A-scans, and the ghost artifacts included in the noisy signal are removed separately from the raw signals using the polynomial curve fitting method. The devised algorithm was simulated with various noisy OCT images, and >87% of the ghost artifact noise was removed despite different locations. Our results show the feasibility of selectively and effectively removing ghost artifact noise.

Enhancement of Common-path Fourier-domain Optical Coherence Tomography using Active Surface Tracking Algorithm

Optical coherence tomography(OCT) can provide real-time and non-invasive subsurface imaging with ultra-high resolution of micrometer scale. However, conventional OCT systems generally have a limited imaging depth range within a depth of only 1-2 mm. To overcome the limitation, we have proposed an active surface tracking algorithm used in common-path Fourier-domain OCT system in order to extend the imaging depth range. The surface tracking algorithm based on the threshold and Savitzky-Golay filter of A-scan data was applied to real-time tracking. The algorithm has controlled a moving stage according to the sample`s surface variance in real time. An OCT image obtained by the algorithm clearly show an extended imaging depth range. Consequently, the proposed algorithm demonstrated the potential for improving the conventional OCT systems with limitary depth range.

Continuous Real-Time Ambulatory Urodynamic Monitoring Using Personal Digital Assistance

Conventional voiding cystometry involves artificially filling the bladder with saline and reproduces their symptoms, while making precise measurements in order to identify their underlying causes. However, it is difficult to evaluate physiological functions during storage and voiding of the bladder. In this study, we designed a portable cystometry device based on personal digital assistance (PDA) device and compared its clinical utilities while applying this device to both natural and artificial filling cystometry study. The range of errors of the pressure signals measured our constructed device was less than 1 cmH2O, and the reproducibility of two pressure channels were 2.32±2.97 and 3.67 ±5.31 %, respectively. Also, the clinical assessment of our device showed that the system has the capability to accurately monitor the types of information that contribute to the diagnosis and management of patient

Feature extraction of knee joint sound for non-invasive diagnosis of articular pathology

The aim of this paper is to classify the vibroarthrographic (VAG) signals according to the pathological condition using the characteristic parameters extracted by the time-frequency transform, and to evaluate the classification accuracy. VAG and knee angle signals, recorded simultaneously during one flexion and one extension of the knee, were segmented and normalized at 0.5 Hz by the dynamic time warping method. Also, the noise within the time-frequency distribution (TFD) of the segmented VAG signals was reduced by the singular value decomposition algorithm, and a back-propagation neural network (BPNN) was used to classify the normal and abnormal VAG signals. A total of 1408 segments (normal 1031, patient 377) were used for training and evaluating the BPNN. As a result, the average classification accuracy was 92.3 plusmn 0.9 %. The proposed method showed good potential for the non-invasive diagnosis and monitoring of joint disorders.