BACKGROUND: Deep brain stimulation (DBS) of the subthalamic nucleus is an effective treatment for Parkinson disease. However, DBS not responsive to individual's disease state, and programming parameters, once established, do change reflect state. Local field potentials (LFPs) recorded from electrodes are being investigated as potential biomarkers no patient data exist about what happens LFPs over lifetime implant. OBJECTIVE: We whether LFP amplitude response limb movement differed between...

Falls are one of the most important problems for frail and elderly people living independently. Early detection falls is vital to provide a safe active lifestyle elderly. Sound, passive infrared (PIR) vibration sensors can be placed in supportive home environment information about daily activities an person. In this paper, signals produced by sound, PIR simultaneously analyzed detect falls. Hidden Markov Models trained regular unusual person pet each sensor signal. Decisions HMMs fused...

It is possible to detect and classify moving stationary targets using ground surveillance pulse-Doppler radars (PDRs). A two-stage support vector machine (SVM) based target classification scheme described here. The first stage tries estimate the most descriptive temporal segment of radar echo signal classified selected in second stage. Mel-frequency cepstral coefficients signals are used as feature vectors both stages. proposed system compared with covariance Gaussian mixture model (GMM)...

A fast algorithm for image classification based on a computationally efficient operator forming semigroup real numbers is developed. The new does not require any multiplications. co-difference matrix the defined and an descriptor using In proposed method, multiplication operation of well-known covariance method replaced by operator. experimentally compared with regular method. performs as well without performing Texture recognition licence plate identification examples are presented.

We tackle the problem of classifying multichannel electrocorticogram (ECoG) related to individual finger movements for a brain machine interface (BMI). For this particular aim we applied recently developed hierarchical spatial projection framework neural activity feature extraction from ECoG. The algorithm extends binary common patterns multiclass by constructing redundant set projections that are tuned paired and group-wise discrimination movements. groupings were constructed merging data...

Summary form only given. A new optimization technique based on the projections onto convex space (POCS) framework for solving and some non-convex problems are presented. The dimension of minimization problem is lifted by one sets corresponding to cost function defined. If a in R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">N</sup> set which epigraph also xmlns:xlink="http://www.w3.org/1999/xlink">N+1</sup> . iterative approach starts with an...

Deep brain stimulation of the subthalamic nucleus (STN) is a highly effective treatment for motor symptoms Parkinson's disease. However, precise intraoperative localization STN remains procedural challenge. In present study, local field potentials (LFPs) were recorded from DBS macroelectrodes during trajectory to STN, in six patients. The frequency-vs-depth map LFP activity was extracted and further analyzed within different sub-bands, investigate whether can be used border identification....

A hybrid state detection algorithm is presented for the estimation of baseline and movement states which can be used to trigger a free paced neuroprostethic. The model was constructed by fusing multiclass Support Vector Machine (SVM) with Hidden Markov Model (HMM), where internal hidden observation probabilities were represented discriminative output SVM. proposed method applied multichannel Electrocorticogram (ECoG) recordings BCI competition IV identify while subjects executing individual...

Shell-to-kernel weight ratio is a vital measurement of quality in hazelnuts as it helps to identify nuts that have underdeveloped kernels. Nuts containing kernels may contain mycotoxin-producing molds, which are linked cancer and heavily regulated international trade. A prototype system was set up detect by dropping them onto steel plate recording the acoustic signal generated when kernel hit plate. feature vector comprising line spectral frequencies time-domain maxima describes both time...

In this paper, a human face detection method in images and video is presented. After determining possible candidate regions using color information, each region filtered by high-pass filter of wavelet transform. way, edges the are highlighted, caricature-like representation obtained. Horizontal, vertical filter-like projections used as feature signals dynamic programming (DP) support vector machine (SVM) based classifiers. It turns out that classifier provides better rates compared to our...

Deep brain stimulation of the subthalamic nucleus (STN) is a highly effective treatment for motor symptoms Parkinson's disease. However, precise intraoperative localization STN remains procedural challenge. In present study, local field potentials (LFPs) were recorded from three tracks during microelectrode recording-based (MER) targeting STN, in five patients. The raw LFP data preprocessed original recording setup and then quality was compared to with common average derivation....

Falls are one of the most important problems for frail and elderly people living independently. Early detection falls is vital to provide a safe active lifestyle elderly. In this paper, signals produced by sound passive infrared (PIR) sensors simultaneously analyzed detect suddenly falling people. A typical room in supportive home can be equipped with PIR sensors. Hidden Markov models trained regular unusual activities an person pet each sensor signal. Decisions HMMs fused together reach...

A new adaptive time-frequency (t-f) analysis and classification procedure is applied to impact acoustic signals for detecting hazelnuts with cracked shells three types of damaged wheat kernels. Kernels were dropped onto a steel plate, the resulting recorded PC-based data acquisition system. These segmented flexible local discriminant bases (F-LDB) in plane extract discriminative patterns between undamaged food The F-LDB requires no prior knowledge relevant time or frequency indices acoustics...

Considering the customer demand increase on electrical energy, it is obviously seen that new system requirements such as providing instantaneous data flow, remote monitoring and controlling, would occur. These should be adapted to existing infrastructure with minimum cost. In this paper, control communication architectures of grids are discussed. The possibility usage power line (PLC) for smart grid examined. order improve quality grid, a PLC model designed in MATLAB . Then Medium (MV) Low...

The past decade has shown the importance of adapting spatial patterns neural activity while decoding it in a Brain Machine Interface (BMI) framework. common (CSP) algorithm tackles this problem as feature extractor binary BMI setups which number projections are computed maximizing variance one class and minimizing other. Recent advances data acquisition systems sensor design now make recording brain with dense electrode grids possibility. However, high density recordings also pose new...

Two new optimization techniques based on projections onto convex space (POCS) framework for solving and some non-convex problems are presented. The dimension of the minimization problem is lifted by one sets corresponding to cost function defined. If a in R^N set R^(N+1). iterative approach starts with an arbitrary initial estimate R^(N+1) orthogonal projection performed sequential manner at each step problem. method provides globally optimal solutions total-variation, filtered variation,...

The common spatial pattern(CSP) technique linearly combines the channels to filter neural signal spatially. It operates on data that is band-pass filtered between particular cutoff frequency for all subjects and channels. On other hand spatio-spectral pattern (CSSP) method extends traditional CSP spectral filtering with original by using temporally delayed version of data. All recording versions are combined when extracting variance as input features a brain machine interface. This linear...

This study explores the use of machine learning algorithms to analyze and predict heart attacks, focusing on genetics, lifestyle, medical history, biometric factors. The data was analyzed using logistic regression, support vector machines, decision trees, random forests. Support machines were found be most effective model for predicting attack risk, with a high accuracy rate low error rate. highlights potential in assisting healthcare professionals individuals determining risk taking...

Due to low consumer acceptance and the possibility of immature kernels, closed-shell pistachio nuts should be separated from open-shell before reaching consumer. A system using impact acoustics as a means classifying has already been shown feasible have better discrimination performance than mechanical system. The accuracy an based is determined by signal processing feature extraction procedures. In this article, new time-frequency plain classification algorithm was developed discriminate...

Common Spatial Pattern algorithm (CSP) is widely used in Brain Machine Interface (BMI) technology to extract features from dense electrode recordings by using their weighted linear combination. However, the CSP algorithm, sensitive variations channel placement and can easily overfit data when number of training trials insufficient. Construction sparse spatial projections where a small subset channels feature extraction, increase stability generalization capability method. The existing ℓ <sub...

The common spatial pattern (CSP) method has large number of applications in brain machine interfaces (BMI) to extract features from the multichannel neural activity through a set linear projections. These projections minimize Rayleigh quotient (RQ) as objective function, which is variance ratio classes. CSP easily overfits data when training trials not sufficiently and it sensitive daily variation electrode placement, limits its applicability for everyday use BMI systems. To overcome these...

Kernel damage caused by insects and fungi is one of the most common reason for poor flour quality. Cracked hazelnut shells are prone to infection cancer producing mold. We propose a new adaptive time-frequency classification procedure detecting cracked damaged wheat kernels using impact acoustic emissions recorded dropping or on steel plate. The proposed algorithm based flexible local discriminant bases (F-LDB) procedure. F-LDB method combines cosine packet analysis frequency axis clustering...