Time-based compression and classification of heartbeats

Heart function measured by electrocardiograms (ECG) is crucial for patient care. ECG generated waveforms are used to find patterns of irregularities in cardiac cycles in patients. In many cases, irregularities evolve over an extended period of time that requires continuous monitoring. However, this requires wireless ECG recording devices. These devices consist of an enclosed system that includes electrodes, processing circuitry, and a wireless communication block imposing constraints on area, power, bandwidth, and resolution. In order to provide continuous monitoring of cardiac functions for real-time diagnostics, we propose a methodology that combines compression and analysis of heartbeats. The signal encoding scheme is the time-based integrate and fire sampler. The diagnostics can be performed directly on the samples avoiding reconstruction required by the competing finite rate of innovation and compressed sensing. As an added benefit, our scheme provides an efficient hardware implementation and a compressed representation for the ECG recordings, while still preserving discriminative features. We demonstrate the performance of our approach through a heartbeat classification application consisting of normal and irregular heartbeats known as arrhythmia. Our approach that uses simple features extracted from ECG signals is comparable to results in the published literature.     

ECG beat classification using GreyART network

The grey relational grade is a similarity measure. On the basis of the grey relational grade, an adaptive resonant theory (ART) type network, GreyART, has been developed. When the GreyART is used to classify a dataset with varying amount of data, the measurement between two specific data in the dataset may vary since the measurement is affected by new added data. In this case, the grey relational grade is not a global measure. As the measurement varies, in the GreyART, it is hard to use a fixed vigilance threshold value for determining whether the current input data belong to one of the existing clusters or become the template of a new online-created cluster. A method to solve this problem has been proposed and then applied to develop an electrocardiogram (ECG) beat classifier. The proposed ECG beat classification involves two phases. One is the off-line learning phase. With the proposed performance index, the product of the classification accuracy and the partition quality, an optimal value for the vigilance threshold and the corresponding cluster centres from the learning results can be determined. The other is the online examining phase, which classifies the input ECG beats. In this phase, the vigilance threshold value and the initial cluster centres are the optimal ones obtained in the learning phase. Under these conditions, the GreyART network enables real-time classification of ECG beats. Simulation results show that the proposed network achieves a good accuracy with a good computational efficiency for ECG beat classification problems     

Scale space classification using area morphology

We explore the application of area morphology to image classification. From the input image, a scale space is created by successive application of an area morphology operator. The pixels within the scale space corresponding to the same image location form a scale space vector. A scale space vector therefore contains the intensity of a particular pixel for a given set of scales, determined in this approach by image granulometry. Using the standard k-means algorithm or the fuzzy c-means algorithm, the image pixels can be classified by clustering the associated scale space vectors. The scale space classifier presented here is rooted in the novel area open-close and area close-open scale spaces. Unlike other scale generating filters, the area operators affect the image by removing connected components within the image level sets that do not satisfy the minimum area criterion. To show that the area open-close and area close-open scale spaces provide an effective multiscale structure for image classification, we demonstrate the fidelity, causality, and edge localization properties for the scale spaces. The analysis also reveals that the area open-close and area close-open scale spaces improve classification by clustering members of similar objects more effectively than the fixed scale classifier. Experimental results are provided that demonstrate the reduction in intra-region classification error and in overall classification error given by the scale space classifier for classification applications where object scale is important. In both visual and objective comparisons, the scale space approach outperforms the traditional fixed scale clustering algorithms and the parametric Bayesian classifier for classification tasks that depend on object scale.     

Automatic classification of asymptomatic and osteoarthritis knee gait patterns using kinematic data features and the nearest neighbor classifier

The aim of this work is to develop an automatic computer method to distinguish between asymptomatic (AS) and osteoarthritis (OA) knee gait patterns using 3-D ground reaction force (GRF) measurements. GRF features are first extracted from the force vector variations as a function of time and then classified by the nearest neighbor rule. We investigated two different features: the coefficients of a polynomial expansion and the coefficients of a wavelet decomposition. We also analyzed the impact of each GRF component (vertical, anteroposterior, and medial lateral) on classification. The best discrimination rate (91%) was achieved with the wavelet decomposition using the anteroposterior and the medial lateral components. These results demonstrate the validity of the representation and the classifier for automatic classification of AS and OA knee gait patterns. They also highlight the relevance of the anteroposterior and medial lateral force components in gait pattern classification.     

Study of features based on nonlinear dynamical modeling in ECG arrhythmia detection and classification

We present a study of the nonlinear dynamics of electrocardiogram (ECG) signals for arrhythmia characterization. The correlation dimension and largest Lyapunov exponent are used to model the chaotic nature of five different classes of ECG signals. The model parameters are evaluated for a large number of real ECG signals within each class and the results are reported. The presented algorithms allow automatic calculation of the features. The statistical analysis of the calculated features indicates that they differ significantly between normal heart rhythm and the different arrhythmia types and, hence, can be rather useful in ECG arrhythmia detection. On the other hand, the results indicate that the discrimination between different arrhythmia types is difficult using such features. The results of this work are supported by statistical analysis that provides a clear outline for the potential uses and limitations of these features.     

Baseline normalisation of ECG signals using empirical mode decomposition and mathematical morphology

An approach to remove baseline wander from electrocardiogram (ECG) signals, based on empirical mode decomposition and mathematical morphology, is described.     

Mass classification of mammograms using fractal dimensions and statistical features

For classification of tumors in mammography, the major features are extracted from the segmented tumor. However, some details of the tumor margin, such as the spiculated parts, are eliminated in the segmentation step. The current study suggests a new approach for extracting the spiculated parts and tumor core. The proposed method segments the tumor by assessing the similarity of the pixels of the tumor core and dissimilarity of the spiculated parts. Then, the spiculated parts and the tumor core are combined to create the final segmentation. Next, the statistical features and fractal dimensions are extracted from the tumor. The fractal dimension is a measure of complexity of the tumor shape that is effective for discriminating between benign and malignant tumors. The simulation results show that the proposed method is more suitable than other methods. The area under the ROC curve and the accuracy of the proposed method on mini-MIAS were 0.9627 and 89.66% and for DDSM were 0.9777 and 93.50%, respectively. The results confirm the efficiency of the proposed method for extracting the mass core and spiculated parts. They also show that use of the fractal dimension increases the accuracy of classification and complements the other shape features.

     

小波变换与模式识别用于自动识别调制模式

本文提出小波变换与模式识别相结合的算法实现通信信号调制模式的自动识别。不同于其他调制模式识别算法，该算法能同时识别模拟调制信号和数字调制信号。采用小波变换估计信号的码速率以区分模拟信号和数字信号。对模拟信号或者数字信号，提取相应的特征参数，识别具体的调制模式。计算机仿真结果表明SNR≥15dB时，该算法艮有很好的性能。     

Automatic classification of brain resting states using fMRI temporal signals

A novel technique is presented for the automatic discrimination between networks of `resting states? of the human brain and physiological fluctuations in functional magnetic resonance imaging (fMRI). The method is based on features identified via a statistical approach to group independent component analysis time courses, which may be extracted from fMRI data. This technique is entirely automatic and, unlike other approaches, uses temporal rather than spatial information. The method achieves 83% accuracy in the identification of resting state networks.     

Automatic modulation classification for cognitive radios using cyclic feature detection

Cognitive radios have become a key research area in communications over the past few years. Automatic modulation classification (AMC) is an important component that improves the overall performance of the cognitive radio. Most modulated signals exhibit the property of cyclostationarity that can be exploited for the purpose of classification. In this paper, AMCs that are based on exploiting the cyclostationarity property of the modulated signals are discussed. Inherent advantages of using cyclostationarity based AMC are also addressed. When the cognitive radio is in a network, distributed sensing methods have the potential to increase the spectral sensing reliability, and decrease the probability of interference to existing radio systems. The use of cyclostationarity based methods for distributed signal detection and classification are presented. Examples are given to illustrate the concepts. The Matlab codes for some of the algorithms described in the paper are available for free download at http://filebox.vt.edu/user/bramkum.     

Automatic registration of CT and MR brain images using correlation of geometrical features

Describes an automated approach to register CT and MR brain images. Differential operators in scale space are applied to each type of image data, so as to produce feature images depicting "ridgeness". The resulting CT and MR feature images show similarities which can be used for matching. No segmentation is needed and the method is devoid of human interaction. The matching is accomplished by hierarchical correlation techniques. Results of 2-D and 3-D matching experiments are presented. The correlation function ensures an accurate match even if the scanned volumes to be matched do not completely overlap, or if some of the features in the images are not similar.     

Automatic segmentation and classification of ionic-channel signals

Identification of ionic-channel types and their selectivity depends critically on the open channel current that can be resolved. In this paper, an automatic channel detection algorithm is proposed that is based on sequential minimization of an index which is usually used in cluster analysis. The algorithm consists of two stages, namely segmentation and classification. In the first stage, the signal samples are segmented based on the assumption that the samples in each segment should be sequentially connected. In the second stage, the resultant segments are classified with no regard to their connectivities. Results on synthetic and real channel currents are very encouraging and they suggest that this algorithm will substantially increase the productivity of many laboratories involved in ionic-channel research.     

Automatic motion and noise artifact detection in Holter ECG data using empirical mode decomposition and statistical approaches

We present a real-time method for the detection of motion and noise (MN) artifacts, which frequently interferes with accurate rhythm assessment when ECG signals are collected from Holter monitors. Our MN artifact detection approach involves two stages. The first stage involves the use of the first-order intrinsic mode function (F-IMF) from the empirical mode decomposition to isolate the artifacts' dynamics as they are largely concentrated in the higher frequencies. The second stage of our approach uses three statistical measures on the F-IMF time series to look for characteristics of randomness and variability, which are hallmark signatures of MN artifacts: the Shannon entropy, mean, and variance. We then use the receiver-operator characteristics curve on Holter data from 15 healthy subjects to derive threshold values associated with these statistical measures to separate between the clean and MN artifacts' data segments. With threshold values derived from 15 training data sets, we tested our algorithms on 30 additional healthy subjects. Our results show that our algorithms are able to detect the presence of MN artifacts with sensitivity and specificity of 96.63% and 94.73%, respectively. In addition, when we applied our previously developed algorithm for atrial fibrillation (AF) detection on those segments that have been labeled to be free from MN artifacts, the specificity increased from 73.66% to 85.04% without loss of sensitivity (74.48%-74.62%) on six subjects diagnosed with AF. Finally, the computation time was less than 0.2?s using a MATLAB code, indicating that real-time application of the algorithms is possible for Holter monitoring.     

Non-contact heartbeat and respiration detector using capacitive sensor with Colpitts oscillator

A capacitive sensor for detecting the heartbeat rate of a human without direct contact with the skin is investigated. Precordial movement changes the capacitance between patch electrodes and modulates the frequency of a Colpitts oscillator. Heartbeat and respiration information can be obtained by demodulating the oscillating signal. Heartbeat signal obtained by bandpass filtering the harmonics of heartbeat frequency is separated from the demodulated signal.     

Automatic classification of high frequency signals

The discovery of the possibility to transmit information by electromagnetic waves led to the appreciation of the importance of such wireless transmissions and subsequently to the requirement for monitoring these radio emissions. Radio monitoring is particularly important in the high frequency range (3 MHz … 30 MHz) because no fixed channel pattern exists in the short wave band. One task of a radio monitoring device detecting signals transmitted by short wave radio stations could well be the automatic determination of the transmitters modulation mode. One may consider this task to represent a pattern recognition problem and use of methods of digital signal processingand pattern recognition for its solution.     

Automatic identification of retinal arteries and veins from dual-wavelength images using structural and functional features

This paper presents an automated method to identify arteries and veins in dual-wavelength retinal fundus images recorded at 570 and 600 nm. Dual-wavelength imaging provides both structural and functional features that can be exploited for identification. The processing begins with automated tracing of the vessels from the 570-nm image. The 600-nm image is registered to this image, and structural and functional features are computed for each vessel segment. We use the relative strength of the vessel central reflex as the structural feature. The central reflex phenomenon, caused by light reflection from vessel surfaces that are parallel to the incident light, is especially pronounced at longer wavelengths for arteries compared to veins. We use a dual-Gaussian to model the cross-sectional intensity profile of vessels. The model parameters are estimated using a robust -estimator, and the relative strength of the central reflex is computed from these parameters. The functional feature exploits the fact that arterial blood is more oxygenated relative to that in veins. This motivates use of the ratio of the vessel optical densities (ODs) from images at oxygen-sensitive and oxygen-insensitive wavelengths () as a functional indicator. Finally, the structural and functional features are combined in a classifier to identify the type of the vessel. We experimented with four different classifiers and the best result was given by a support vector machine (SVM) classifier. With the SVM classifier, the proposed algorithm achieved true positive rates of 97% for the arteries and 90% for the veins, when applied to a set of 251 vessel segments obtained from 25 dual wavelength images. The ability to identify the vessel type is useful in applications such as automated retinal vessel oximetry and automated analysis of vascular changes without manual intervention.     

Automatic Detection of Respiration Rate From Ambulatory Single-Lead ECG

Ambulatory electrocardiography is increasingly being used in clinical practice to detect abnormal electrical behavior of the heart during ordinary daily activities. The utility of this monitoring can be improved by deriving respiration, which previously has been based on overnight apnea studies where patients are stationary, or the use of multilead ECG systems for stress testing. We compared six respiratory measures derived from a single-lead portable ECG monitor with simultaneously measured respiration air flow obtained from an ambulatory nasal cannula respiratory monitor. Ten controlled 1-h recordings were performed covering activities of daily living (lying, sitting, standing, walking, jogging, running, and stair climbing) and six overnight studies. The best method was an average of a 0.2–0.8 Hz bandpass filter and RR technique based on lengthening and shortening of the RR interval. Mean error rates with the reference gold standard were $pm$4 breaths per minute (bpm) (all activities), $pm$2 bpm (lying and sitting), and $pm$1 breath per minute (overnight studies). Statistically similar results were obtained using heart rate information alone (RR technique) compared to the best technique derived from the full ECG waveform that simplifies data collection procedures. The study shows that respiration can be derived under dynamic activities from a single-lead ECG without significant differences from traditional methods.      

融合纹理特征和形状特征的医学图像分类方法

针对近些年来,女性人群中乳腺恶性肿瘤患者逐年增长的趋势,提出了一种融合纹理特征和形状特征的乳腺超声肿瘤图像的识别方法。该方法不但结合了临床医生的诊断经验,而且有效利用了数字图像特征提取技术,提取出的特征能反映出良性肿瘤和恶性肿瘤的本质区别,将样本进行特征提取并通过支持向量机(SVM)技术分类,该方法取得了良好的分类效果。实验结果证明文中的特征融合方法十分有效。     

基于心电与呼吸信号的睡眠分期算法研究

睡眠是人类的一项重要生理活动,其质量与人体健康状态密切相关,对睡眠情况进行分析有助于许多疾病的预防和监测.传统的睡眠分期黄金标准是多导睡眠图,并包含脑电、眼电、肌电、呼吸、血氧、运动等多种信号,操作复杂且对测试者睡眠有影响.因此,基于可穿戴设备和有限类型数据进行睡眠分即已成为当前研究热点.本文仅采用了心电和呼吸信号进行特征提取,使用前向序列选择方法进行特征选择,分别采用支持向量机、随机森林和AdaBoost等方法进行分类,在睡眠呼吸障碍患者数据库中进行WAKE-REM-NREM分期上获得最高71.9％的准确率(Kappa=0.36).实验表明心电与呼吸信号可在一定程度上代替多导睡眠仪应用于睡眠呼吸障碍患者的分析,有助于睡眠呼吸类疾病的诊断和评价,为相关设备的便携化提供了算法基础.