Samjin Choi

Comparison of envelope extraction algorithms for cardiac sound signal segmentation

This paper describes a comparative study of the envelope extraction algorithms for the cardiac sound signal segmentation. In order to extract the envelope curves based on the time elapses of the first and the second heart sounds of cardiac sound signals, three representative algorithms such as the normalized average Shannon energy, the envelope information of Hilbert transform, and the cardiac sound characteristic waveform (CSCW) are introduced. Performance comparison of the envelope extraction algorithms, and the advantages and disadvantages of the methods are examined by some parameters.

Cardiac sound murmurs classification with autoregressive spectral analysis and multi-support vector machine technique

In this paper, a novel cardiac sound spectral analysis method using the normalized autoregressive power spectral density (NAR-PSD) curve with the support vector machine (SVM) technique is proposed for classifying the cardiac sound murmurs. The 489 cardiac sound signals with 196 normal and 293 abnormal sound cases acquired from six healthy volunteers and 34 patients were tested. Normal sound signals were recorded by our self-produced wireless electric stethoscope system where the subjects are selected who have no the history of other heart complications. Abnormal sound signals were grouped into six heart valvular disorders such as the atrial fibrillation, aortic insufficiency, aortic stenosis, mitral regurgitation, mitral stenosis and split sounds. These abnormal subjects were also not included other coexistent heart valvular disorder. Considering the morphological characteristics of the power spectral density of the heart sounds in frequency domain, we propose two important diagnostic features Fmax and Fwidth, which describe the maximum peak of NAR-PSD curve and the frequency width between the crossed points of NAR-PSD curve on a selected threshold value (THV), respectively. Furthermore, a two-dimensional representation on (Fmax, Fwidth) is introduced. The proposed cardiac sound spectral envelope curve method is validated by some case studies. Then, the SVM technique is employed as a classification tool to identify the cardiac sounds by the extracted diagnostic features. To detect abnormality of heart sound and to discriminate the heart murmurs, the multi-SVM classifiers composed of six SVM modules are considered and designed. A data set was used to validate the classification performances of each multi-SVM module. As a result, the accuracies of six SVM modules used for detection of abnormality and classification of six heart disorders showed 71-98.9% for THVs=10-90% and 81.2-99.6% for THVs=10-50% with respect to each of SVM modules. With the proposed cardiac sound spectral analysis method, the high classification performances were achieved by 99.9% specificity and 99.5% sensitivity in classifying normal and abnormal sounds (heart disorders). Consequently, the proposed method showed relatively very high classification efficiency if the SVM module is designed with considering THV values. And the proposed cardiac sound murmurs classification method with autoregressive spectral analysis and multi-SVM classifiers is validated for the classification of heart valvular disorders.

Detection of valvular heart disorders using wavelet packet decomposition and support vector machine

In this study, the valvular heart disorder (VHD) detection method by the wavelet packet (WP) decomposition and the support vector machine (SVM) techniques are proposed. From considering the truth that the frequency ranges of the normal sound and VHDs are different from each other, the WP decomposition at level 8 is employed to split more elaborate frequency bandwidths of the heart sound signals. And then the WP energy (WPE) with the distribution information of energy throughout the whole frequency range of heart sound signals is calculated. Since the heart sound signals with the frequency range of 20-750Hz are preferred in this study, WPEs at the terminal nodes from (8,1) to (8,47) are selected and two parameters meanWPE and stdWPE as defined by the mean value and standard deviation of the position indices of the terminal nodes with over the weighting value (@z) of the maximum value of WPE are proposed as a feature. Furthermore, the SVM technique is employed as the identification tool to classify between the normal sound and VHDs. Finally, a case study on the normal sound, aortic and mitral VHDs is demonstrated to validate the usefulness and efficiency of the VHD detection using WP decomposition and SVM classifier. The experimental results of the proposed VHD detection method showed high performance like the specificity of over 96% and the sensitivity of 100% for both the training and testing data.

Evaluation of antibiotic effects on Pseudomonas aeruginosa biofilm using Raman spectroscopy and multivariate analysis

We investigate the mode of action and classification of antibiotic agents (ceftazidime, patulin, and epigallocatechin gallate; EGCG) on Pseudomonas aeruginosa (P. aeruginosa) biofilm using Raman spectroscopy with multivariate analysis, including support vector machine (SVM) and principal component analysis (PCA). This method allows for quantitative, label-free, non-invasive and rapid monitoring of biochemical changes in complex biofilm matrices with high sensitivity and specificity. In this study, the biofilms were grown and treated with various agents in the microfluidic device, and then transferred onto gold-coated substrates for Raman measurement. Here, we show changes in biochemical properties, and this technology can be used to distinguish between changes induced in P. aeruginosa biofilms using three antibiotic agents. The Raman band intensities associated with DNA and proteins were decreased, compared to control biofilms, when the biofilms were treated with antibiotics. Unlike with exposure to ceftazidime and patulin, the Raman spectrum of biofilms exposed to EGCG showed a shift in the spectral position of the CH deformation stretch band from 1313 cm(-1) to 1333 cm(-1), and there was no difference in the band intensity at 1530 cm(-1) (C = C stretching, carotenoids). The PCA-SVM analysis results show that antibiotic-treated biofilms can be detected with high sensitivity of 93.33%, a specificity of 100% and an accuracy of 98.33%. This method also discriminated the three antibiotic agents based on the cellular biochemical and structural changes induced by antibiotics with high sensitivity and specificity of 100%. This study suggests that Raman spectroscopy with PCA-SVM is potentially useful for the rapid identification and classification of clinically-relevant antibiotics of bacteria biofilm. Furthermore, this method could be a powerful approach for the development and screening of new antibiotics.

Selection of wavelet packet measures for insufficiency murmur identification

This paper presents a new analysis method for aortic and mitral insufficiency murmurs using wavelet packet (WP) decomposition. We proposed four diagnostic features including the maximum peak frequency, the position index of the WP coefficient corresponding to the maximum peak frequency, and the ratios of the wavelet energy and entropy information to achieve greater accuracy for detection of heart murmurs. The proposed WP-based insufficiency murmur analysis method was validated by some case studies. We employed a thresholding scheme to discriminate between insufficiency murmurs and control sounds. Three hundred and thirty-two heart sounds with 126 control and 206 murmur cases were acquired from four healthy volunteers and 47 patients who suffered from heart defects. Control sounds were recorded by applying a wireless electric stethoscope system to subjects with no history of other heart complications. Insufficiency murmurs were grouped into two valvular heart defect categories, aortic and mitral. These murmur subjects had no other coexistent valvular defects. The proposed insufficiency murmur detection method yielded a high classification efficiency of 99.78% specificity and 99.43% sensitivity.

Facile Fabrication of a Silver Nanoparticle Immersed, Surface-Enhanced Raman Scattering Imposed Paper Platform through Successive Ionic Layer Absorption and Reaction for On-Site Bioassays

We introduce a novel, facile, rapid, low-cost, highly reproducible, and power-free synthesizable fabrication method of paper-based silver nanoparticle (AgNP) immersed surface-enhanced Raman scattering (SERS) platform, known as the successive ionic layer absorption and reaction (SILAR) method. The rough and porous properties of the paper led to direct synthesis of AgNPs on the surface as well as in the paper due to capillary effects, resulting in improved plasmon coupling with interparticles and interlayers. The proposed SERS platform showed an enhancement factor of 1.1 × 10(9), high reproducibility (relative standard deviation of 4.2%), and 10(-12) M rhodamine B highly sensitive detection limit by optimizing the SILAR conditions including the concentration of the reactive solution (20/20 mM/mM AgNO3/NaBH4) and the number of SILAR cycles (six). The applicability of the SERS platform was evaluated using two samples including human cervical fluid for clinical diagnosis of human papillomavirus (HPV) infection, associated with cervical cancer, and a malachite green (MG) solution for fungicide and parasiticide in aquaculture, associated with human carcinogenesis. The AgNP-immersed SERS-functionalized platform using the SILAR technique allowed for high chemical structure sensitivity without additional tagging or chemical modification, making it a good alternative for early clinical diagnosis of HPV infection and detection of MG-activated human carcinogenesis.

Neuroprotective effects of magnesium-sulfate on ischemic injury mediated by modulating the release of glutamate and reduced of hyperreperfusion.

This study examined the neuroprotective effects of magnesium-sulfate (MgSO(4)) on the cerebral blood flow (CBF) and extracellular glutamate concentration in an eleven vessel occlusion (11VO) rat model. Twenty-one male Sprague-Dawley rats (250-350g) were used for the 11VO ischemic model, which was induced by a 10-min transient occlusion. The animals were divided into 3 groups, including ischemic-induced animals (ischemia group), ischemic-induced and MgSO(4) treated animals (MgSO(4) group), and sham animals for comparison. The real-time extracellular glutamate concentration was measured using a microdialysis biosensor, and the CBF was monitored by laser Doppler flowmetry. Neuronal cell death in the hippocampal region was observed 72h after ischemia by several stains (Nissl, DAPI, NeuN, and cleaved caspase3). A significant decrease in %CBF was observed in both the ischemia and MgSO(4) groups, such as ~10% during the ischemic period. However, the MgSO(4) group showed a significant decrease in the initial reperfusion %CBF compared to the ischemia group. A significantly lower level of glutamate release was observed in the MgSO(4) group than in the ischemia group during the ischemic and reperfusion episode. Our staining results revealed a significant decrease in neuronal cell death in the hippocampus in the MgSO(4) group compared to the ischemia group. These results suggest that MgSO(4) is responsible for the protection of neuronal cells by suppressing the release of extracellular glutamate under ischemic conditions and the CBF response during the initial reperfusion period.

Development of ECG beat segmentation method by combining lowpass filter and irregular R-R interval checkup strategy

We have developed a long-term cardiorespiratory sensor system that includes a wearable sensor probe with adaptive hardware filters and data processing algorithms (Choi & Jiang, 2006, 2008). However, the data processing algorithm proposed for the R-R interval (RRI) information extraction did not work well in the case of ECG signals with baseline shifts or muscle artifacts. Furthermore, many false ECG beats were extracted due to a weak decision-making scheme. Then, those false beats produced irregular RRI information and erroneous heart rate variability results. Modification of data processing algorithm was strongly needed. Therefore, this work presented an efficient ECG beat segmentation method using an irregular RRI checkup strategy into five sequential RRI patterns. This algorithm was comprised of signal processing stage and ECG beat detector stage. The signal processing included the wavelet denoising, the baseline shift elimination by 20Hz lowpass filter and the envelope curve extraction by a single degree of freedom analytical model. The ECG beat detector included the candidate ECG beat detection and segmentation by one threshold and by irregular RRI checkup strategy, respectively. In particular, four abnormal RRI patterns were proposed to find out false ECG beats. The MIT-BIH arrhythmia database was selected as the dataset for testing the proposed algorithm. The proposed irregular RRI checkup strategy estimated 5463 beats to the suspected false beats and succeeded in segmenting 96.19% (5255 beats) of them. The performance results showed that our algorithm had very good results such as the detection error of 0.54%, sensitivity of 99.66% and positive predictivity of 99.80%. Furthermore, our algorithm showed very high accuracy as the mean time error between the beat annotations of the database and our obtained beat occurence times was 7.75ms.

Instrument-Free Synthesizable Fabrication of Label-Free Optical Biosensing Paper Strips for the Early Detection of Infectious Keratoconjunctivitides

We introduce a surface-enhanced Raman scattering (SERS)-functionalized, gold nanoparticle (GNP)-deposited paper strip capable of label-free biofluid sensing for the early detection of infectious eye diseases. The GNP biosensing paper strip was fabricated by the direct synthesis and deposition of GNPs on wax-divided hydrophilic areas of a permeable porous substrate through a facile, power-free synthesizable, and highly reproducible successive ionic layer absorption and reaction (SILAR) technique. To maximize localized surface plasmon resonance-generated SERS activity, the concentration of the reactive solution and number of SILAR cycles were optimized by controlling the size and gap distance of GNPs and verified by computational modeling with geometrical hypotheses of Gaussian-estimated metallic nanoparticles. The responses of our SERS-functionalized GNP paper strip to Raman intensities exhibited an enhancement factor of 7.8 × 10(8), high reproducibility (relative standard deviation of 7.5%), and 1 pM 2-naphthalenethiol highly sensitive detection limit with a correlation coefficient of 0.99, achieved by optimized SILAR conditions including a 10/10 mM/mM HAuCl4/NaBH4 concentration and six SILAR cycles. The SERS-functionalized GNP paper is supported by a multivariate statistics-preprocessed machine learning-judged bioclassification system to provide excellent label-free chemical structure sensitivity for identifying infectious keratoconjunctivitis. The power-free synthesizable fabrication, label-free, rapid analysis, and high sensitivity feature of the SILAR-fabricated SERS-functionalized GNP biosensing paper strip makes it an excellent alternative in point-of-care applications for the early detection of various infectious diseases.

A wearable cardiorespiratory sensor system for analyzing the sleep condition

This paper describes a long-term cardiorespiratory sensor system, which is supposed to be used for monitoring sleep condition at home environment. This system consists of belt sensor probe, data acquisition and communication devices, and the data processing algorithms. New wearable sensor probe with a couple of conductive fabric sheets material and a PVDF film material is developed. To obtain clear cardiorespiratory responses from the belt sensor probe, adaptive hardware filters with signal conditioners are designed and further software data processing algorithms are proposed for extraction of the relative cardiorespiratory information, such as respiratory cycle (RC) and RR interval (RRI). These simple and powerful data processing algorithms for extraction of the proposed RC and RRI information are described and testified in detail. And two commercial sensor devices such as thermistor-type pneumography sensor and 3-lead electrocardiogram sensor are used simultaneously to validate the performance and efficiency of the proposed sensor system. Furthermore, the belt type sensor demonstrated that PVDF film and conductive fabric sensors can complement each other, especially in RRI extraction. Finally, for a case study, the sleep conditions are estimated experimentally by analyzing the heart rate variability (HRV) from the extracted RRI information.