An optimum algorithm in pathological voice quality assessment using wavelet-packet-based features, linear discriminant analysis and support vector machine

Epilepsy is one of the most common neurological disorders. Epileptic seizures are the manifestation of abnormal hypersynchronous discharges of cortical neurons that impair brain function. Most of the people affected can be treated successfully with drug therapy or neurosurgical procedures. But there is still a large group of epilepsy patients that continues to have frequent seizures. For these patients automated detection of epileptic seizures can be of great clinical importance. Seizure detection can influence daily care or can be used to evaluate treatment effect. Furthermore automated detection can be used to trigger an alarm system during seizures that might be harmful to the patient. This thesis focusses on accelerometry (ACM) based seizure detection. A detailed overview is provided, on the perspectives for long-term epilepsy monitoring and automated seizure detection. The value of accelerometry for seizure detection is shown by means of a clinical evaluation and the first steps are made towards automatic detection of epileptic seizures based on ACM. With accelerometers movements are recorded. A large group of epileptic seizures manifest in specific movement patterns, so called motor seizures. Chapter 2 of this thesis presents an overview of the published literature on available methods for epileptic seizure detection in a long-term monitoring context. Based on this overview recommendations are formulated that should be used in seizure detection research and development. It is shown that for seizure detection in home environments, other sensor modalities besides EEG become more important. The use of alternative sensor modalities (such as ACM) is relatively new and so is the algorithm development for seizure detection based on these measures. It was also found that for both the adaptation of existing techniques and the development of new algorithms, clinical information should be taken more into account. The value of ACM for seizure detection is shown by means of a clinical evaluation in chapter 3. Here 3-D ACM- and EEG/video-recordings of 18 patients with severe epilepsy are visually analyzed. A striking outcome presented in this chapter is the large number of visually detected seizures versus the number of seizures that was expected on forehand and the number of seizures that was observed by the nurses. These results underscore the need for an automatic seizure detection device even more, since in the current situation many seizures are missed and therefore it is possible that patients do not get the right (medical) treatment. It was also observed that 95% of the ACM-patterns during motor seizures are sequences of three elementary patterns: myoclonic, tonic and clonic patterns. These characteristic patterns are a starting point for the development of methods for automated seizure detection based on ACM. It was decided to use a modular approach for the detection methodology and develop algorithms separately for motor activity in general, myoclonic seizures and tonic seizures. Furthermore, clinical information is incorporated in the detection methodology. Therefore in this thesis features were used that are either based on the shape of the patterns of interest as described in clinical practice (chapter 4 and 7), or the features were based on a physiological model with parameters that are related to seizure duration and intensity (chapter 5 and 6). In chapter 4 an algorithm is developed to distinguish periods with and without movement from ACM-data. Hence, when there is no movement there is no motor seizure. The amount of data that needs further analysis for seizure detection is thus reduced. From 15 ACM-signals (measured on five positions on the body), two features are computed, the variance and the jerk. In the resulting 2-D feature space a linear threshold function is used for classification. For training and testing the algorithm ACM data along with video data are used from nocturnal recordings in mentally retarded patients with severe epilepsy. Using this algorithm the amount of data that needs further analysis is reduced considerably. The results also indicate that the algorithm is robust for fluctuations across patients and thus there is no need for training the algorithm for each new patient. For the remaining data it needs to be established whether the detected movement is seizure related or not. To this purpose a model is developed for the accelerometer pattern measured on the arm during a myoclonic seizure (chapter 5). The model consists of a mechanical and an electrophysiological part. This model is used as a matched wavelet filter to detect myoclonic seizures. In chapter 6 the model based wavelet is compared to three other time frequency measures: the short time Fourier transform, the Wigner distribution and the continuous wavelet transform using a Daubechies wavelet. All four time-frequency methods are evaluated in a linear classification setup. Data from mentally retarded patients with severe epilepsy are used for training and evaluation. The results show that both wavelets are useful for detection of myoclonic seizures. On top of that, our model based wavelet has the advantage that it consists of parameters that are related to seizure duration and intensity that are physiological meaningful. Besides myoclonic seizures, the model is also useful for the detection of clonic seizures; physiologically these are repetitive myoclonic seizures. Finally for the detection of tonic seizures, in chapter 7 a set of features is studied that incorporate the mean characteristics of ACM-patterns associated with tonic seizures. Linear discriminant analysis is used for classification in the multi-dimensional feature space. For training and testing the algorithm, again data are used from recordings in mentally retarded patients with severe epilepsy. The results show that our approach is useful for the automated detection of tonic seizures based on 3-D ACM and that it is a promising contribution in a complete multi-sensor seizure detection setup.

Image segmentation is necessary but significant element in less intensity image investigation, pattern recognition, and in robotic systems. It is one of the most complex and demanding tasks in image processing. Image segmentation is the process of separating an image into various regions such that each region is identical. This paper proposes a new medical image segmentation method that integrates multi-resolution wavelet packet decomposition with the watershed transform for MRI image. The wavelet packet transform (WPT) is applied to the input image, creating detail and approximation coefficients. If watershed technique alone is used for segmentation, then over cluster is present. To overcome this, the proposed technique which combines wavelet packet and watershed algorithm is developed. First, the wavelet packet transform is applied to produce multi-resolution images, followed by applying watershed for segmentation to the approximation sub-bands. Finally, Inverse WPT is implemented to obtain the segmented image. Due to wavelet packet decomposition, the quantity of the disturbance can be decreased and leads to a tough segmentation. This proposed work concludes that wavelet packet and watershed transform facilitate to get the elevated precision even in strident images.

In this paper, a new feature extraction method utilizing ant colony optimization in the selection of wavelet packet transform (WPT) best basis is presented and adopted in classifying biomedical signals. The new algorithm, termed intelligent artificial ants (IAA), searches the wavelet packet tree for subsets of features that best interact together to produce high classification accuracies. While traversing the WPT tree, the IAA takes into account existing correlation between features thus avoiding information redundancy. The IAA method is a mixture of filter and wrapper approaches in feature subset selection. The pheromone that the ants lay down is updated by means of an estimation of the information contents of a single feature or feature subset. The significance of the subsets selected by the ants is measured using linear discriminant analysis (LDA) classifier. The IAA method is tested on one of the most important biosignal driven applications, which is the brain computer interface (BCI) problem with 56 EEG channels. Practical results indicate the significance of the proposed method achieving a maximum accuracy of 83%.

Automatic age estimation relying on human facial images is a key technology of many real-world applications, which is still a challenging task in the computer vision field. There are three cascade modules for facial age estimation: facial aging feature extraction, dimension reduction (or feature selection) and estimation method. Many existing literatures focus on the first or last module while for an age estimation system, it's also important to construct a reasonable framework. Our work focuses on creating an effective framework by selecting methods for these modules reasonably. Firstly, a BIM (bio-inspired model) is employed to extract facial aging features because it can not only capture discriminative local and global features, but also overcome interferences of some 2D deformations to some extent. Then, LDA (linear discriminant analysis) is used for reducing the BIF (bio-inspired features) to lower dimensions and extracting more discriminative information at the same time. Finally, CS-OHRank (cost-sensitive ordinal hyperplane rank), which tackles with sparse data well and reflects the cumulative attributes of aging, is applied as the estimation method. Experimental results on benchmark dataset FG-NET show that our framework combining BIF, LDA and CS-OHRank is competitive among the state of the art, with MAE (mean absolute error) = 4.72 years..

In this paper, a Hybrid Wavelet Packet and Directional transform (HW&DT) method is proposed to compress an image effectively. The image is pixel de-correlated using Daubechies Wavelet Packet transform and set of wavelet packet coefficients are transformed using Directional transform. The Directional transform pairs DCT-II, DCT-V, DCT-VIII and DST-VII are useful in retrieving the texture information in different directions. Then the coefficients of the hybrid transform are uniformly quantized and entropy coded using Huffman coding, generating the Bit-Stream. The performance evaluation of the proposed work is done using the compression parameters like Structural Similarity Index (SSIM), Bit-Saving (%), Peak Signal to Noise ratio (PSNR) and Mean Square Error (MSE). The experimental results confirm the improvement in Bit-Saving for the image.KeywordsDiscrete cosine transform (DCT)Wavelet packet transform (WPT)Uniform quantizationHuffman coding

Fault degradation assessment of water hydraulic motor by impulse vibration signal with Wavelet Packet Analysis and Kolmogorov–Smirnov Test

This paper addresses a new compression algorithm for Doppler signal in miss distance measurement. In order to compact as much energy of the original Doppler signal as possible into as few significant coefficients as possible, the new compression algorithm takes advantage of a near adaptive hybrid time-frequency representation, i.e. it employs the short time Fourier transform (STFT) while the Doppler signal is quasi-stationary and the wavelet packet transform (WPT) while the Doppler signal becomes non-stationary. Meanwhile, the decision criterion to determine the non-stationarity of Doppler signal is proposed, and a coefficient classification method based on clustering in time-frequency domain is also given. Experimental results fully show the superiority of the hybrid compression algorithm over that employing the STFT or WPT solely

Aiming reliable detection and localization of cerebral blood flow and emboli, embolic signals were added to simulated middle cerebral artery Doppler signals and analysed. Short-time Fourier transform (STFT) and continuous wavelet transform (CWT) were used in the evaluation. The following parameters were used in this study: the powers of the embolic signals added were 5, 6, 6.5, 7, 7.5, 8 and 9 dB; the mother wavelets for CWT analysis were Morlet, Mexican hat, Meyer, Gaussian (order 4) and Daubechies (orders 4 and 8); and the thresholds for detection (equated in terms of false positive, false negative and sensitivity) were 2 and 3.5 dB for the CWT and STFT, respectively. The results indicate that although the STFT allows accurately detecting emboli, better time localization can be achieved with the CWT. Among the CWT, the current best overall results were obtained with Mexican Hat mother wavelet, with optimal results for sensitivity (100% detection rate) for nearly all emboli power values studied.

Artificial intelligence (AI) has shown potential in analyzing electromyography (EMG) signals, but clinical applicability remains limited by studies based on small, curated datasets, and variable accuracy. This study evaluated AI performance in classifying needle electromyography (EMG) signals as muscle activity versus background/noise/artifact, and then in distinguishing three clinical categories: amyotrophic lateral sclerosis (ALS), myopathy, and non-disease controls. Data from the Cleveland Clinic Foundation EMG Database (CCFDB), a large clinically acquired EMG dataset was utilized for this study. A two-step classification approach was used: a convolutional neural network (CNN) to separate muscle activity from background/noise/artifact, followed by a random forest algorithm and CNNs for clinical category classification. Feature extraction techniques included Short-Time Fourier Transform (STFT), Discrete Wavelet Transform (DWT), Continuous Wavelet Transform (CWT), and Wavelet Packet Decomposition (WPD). EMG data from 608 participants (266 ALS, 89 myopathy, 253 non-disease controls), totaling 11,456 muscle recordings, and 15,613 segments of muscle activity were included. The muscle activity detection model achieved 85.4% accuracy. For clinical category classification, CWT with a two-layer CNN performed best on the CCFDB (62% accuracy). Deeper CNN architectures did not consistently improve performance. On the publicly available curated EMGlab dataset, the best accuracy using the same AI models was higher (91%). AI can assist in EMG analysis, but the performance gap between curated and clinically-acquired datasets underscores the need for robust models capable of handling signal variability and complexity in authentic clinical contexts. Future efforts should focus on clinically-oriented AI development to improve translational applicability.

Wavelet transforms and wavelet packets are widely imposed in the analysis and resolution of problems related to science and technical engineering. Decomposition wavelet packet allows several frequency bands according to various levels of resolutions. We apply this transform (PWT) coupled with the SPIHT coder to reduce the limitations of conventional wavelet filter bank. The results obtained using the applied algorithm, are very satisfactory and encouraging compared to many of the best coders cited in the literature and show a visual and numerical superiority over conventional methods. These the promising results are confirmed by visual evaluation parameters (PSNR, MSSIM and VIF).

This paper has the usage of energy and entropy parameters associated with Wavelet Packet Transform (WPT) as the target to the automatic signal classification as well as the detection of voltage disturbances in electric signals. One can apply Wavelet Packet Transform to remove noise presented in the signals by means of decomposition to obtain the energy and entropy characteristics. Principal component analysis (PCA) is used to reduce the dimensions of the parameters vector and to classify the kinds of signal and disturbances presented using k- nearest neighbor (kNN). Numerical simulations showed the effectiveness of the proposed method.

In this paper, an automatic face recognition system is proposed based on appearance-based features that focus on the entire face image rather than local facial features. The first step in face recognition system is face detection. Viola-Jones face detection method that capable of processing images extremely while achieving high detection rates is used. This method has the most impact in the 2000’s and known as the first object detection framework to provide relevant object detection that can run in real time. Feature extraction and dimension reduction method will be applied after face detection. Principal Component Analysis (PCA) method is widely used in pattern recognition. Linear Discriminant Analysis (LDA) method that used to overcome drawback the PCA has been successfully applied to face recognition. It is achieved by projecting the image onto the Eigenface space by PCA after that implementing pure LDA over it. Square Euclidean Distance (SED) is used. The distance between two images is a major concern in pattern recognition. The distance between the vectors of two images leads to image similarity. The proposed method is tested on three databases (MUCT, Face94, and Grimace). Different number of training and testing images are used to evaluate the system performance and it show that increasing the number of training images will increase the recognition rate.

The main barrier impeding the advancement of high-speed milling is chatter, which has a detrimental effect on the dimensional accuracy and quality of the finished workpiece. A reliable and precise chatter identification method is essential to improving the quality of machining. This paper presents a novel method for chatter identification using a comprehensive feature fusion of the Short-Time Fourier Transform (STFT) and the Fourier Synchrosqueezing Transform (FSST). The Wavelet Packet Transform (WPT) was used to pre-process the collected vibration and force signals. Wavelet packets with rich chatter information were then selected and reconstructed for further analysis. To reduce the effects of the rotating frequency and generate a hybrid spectrum with high resolution, a Gabor time-frequency filter is employed. As chatter indicators, standard deviation, skewness, and root mean square are computed. With a higher Time-Frequency Representation (TFR) resolution and a shorter computation time of 0.46 and 0.97 s across vibration and force signals, the proposed method outperforms conventional STFT and FSST. As a result, it can be used to reliably identify chatter from the onset, which is beneficial for machining monitoring.

In structural health monitoring (SHM), data-driven is a model free method that mines the internal characteristics of the raw measured responses, does not need considering the structural model. This paper proposes a new data-driven damage detection method based on wavelet packet transform (WPT) combining with principal component analysis (PCA). In this method, the responses are firstly decomposed by WPT, and each layer of wavelet packet energy of each measurement is obtained. Since the wavelet packet components of small energy magnitudes are easily contaminated by noises and should be discarded. PCA is utilised to reduce and eliminate redundant information of wavelet packet energy, and a damage index named the rate of the wavelet packet energy score change (WPESCR) is proposed to locate the damage. A simple beam is employed to demonstrate the effectiveness of this method. The numerical and experimental results show that the method combined with WPT and PCA can successfully locate the damage. As a data-driven method, the new WPESCR indicator is a better candidate for continuous online SHM.

P01-Comparison of short-time Fourier transform and continuous wavelet transform for frequency analysis of sleep EEG