Ahmed Ben Hamida

3D multimodal MRI brain glioma tumor and edema segmentation: a graph cut distribution matching approach.

This study investigates a fast distribution-matching, data-driven algorithm for 3D multimodal MRI brain glioma tumor and edema segmentation in different modalities. We learn non-parametric model distributions which characterize the normal regions in the current data. Then, we state our segmentation problems as the optimization of several cost functions of the same form, each containing two terms: (i) a distribution matching prior, which evaluates a global similarity between distributions, and (ii) a smoothness prior to avoid the occurrence of small, isolated regions in the solution. Obtained following recent bound-relaxation results, the optima of the cost functions yield the complement of the tumor region or edema region in nearly real-time. Based on global rather than pixel wise information, the proposed algorithm does not require an external learning from a large, manually-segmented training set, as is the case of the existing methods. Therefore, the ensuing results are independent of the choice of a training set. Quantitative evaluations over the publicly available training and testing data set from the MICCAI multimodal brain tumor segmentation challenge (BraTS 2012) demonstrated that our algorithm yields a highly competitive performance for complete edema and tumor segmentation, among nine existing competing methods, with an interesting computing execution time (less than 0.5s per image).

A k-Means Clustering Algorithm Initialization for Unsupervised Statistical Satellite Image Segmentation

The increasing availability of satellite images acquired periodically by satellite on different area, makes it extremely interesting in many applications. In deed, the recent construction of multi and hyper spectral images will provide detailed data with information in both the spatial and spectral domain. This data shows great promise for remote sensing applications ranging from environmental and agricultural to national security interests. The exploitation of these images requires the use of different approach, and notably these founded on the unsupervised statistical segmentation principle. Indeed these methods that exploit the statistical images attributes offer some convincing and encouraging results, under the condition to have an optimal initialization step. Indeed, in order to assure a better convergence of the different images attributes, the unsupervised segmentation approaches, require a fundamental initialization step. We will present in this paper a k-means clustering algorithm and describe its importance in the initialization of the unsupervised satellite image segmentation

Iterated Robust kernel Fuzzy Principal Component Analysis and application to fault detection

Abstract In this paper, we propose an Iterated Robust kernel Fuzzy Principal Component Analysis (IRkFPCA), which is the method that attempts to combine the advantages of the state of art methods and use a more accurate multi-objective function for jointly reducing the modeling errors, optimizing the robustness to outliers and improving the time complexity since it does not require the storage and inversion of the covariance matrix to obtain a memory-efficient approximation of kernel PCA. This proposed technique computes iteratively the principal components, which are used for modeling and fault detection. The detection stage is related to the evaluation of residuals, also known as detection indices, which are signals that reveal the fault presence. Those indices are obtained from the analysis of the difference between the process measurements and their estimations using the IRkFPCA technique. The performance of the proposed method is illustrated and compared to Iterated kernel Principal Component Analysis (IkPCA) and Iterated Principal Component Analysis (IPCA) methods through two simulated examples, one using synthetic data and the other using simulated continuously stirred tank reactor (CSTR) data. The results of the comparative studies reveal that the developed IRkFPCA method provides a better performance in terms of modeling and fault detection accuracies than the Iterated Robust Fuzzy Principal Component Analysis (IRFPCA) and Iterated kernel Principal Component Analysis (IkPCA) methods; while both methods provide improved accuracy over the Iterated Principal Component Analysis (IPCA) method.

Statistical Fault Detection of Chemical Process - Comparative Studies

This paper addresses the statistical chemical process monitoring using improved principal component analysis (PCA). PCA-based fault-detection technique has been used successfully for monitoring systems with highly correlated variables. However, standard PCA-based detection charts, such as the Hotelling statistic, T2 and the sum of squared residuals, SPE, or Q statistic, are not able to detect small or moderate events since they use only data from the most recent measurements. Different fault detection (FD) charts, namely generalized likelihood ratio test (GLRT), shewhart control chart and exponentially weighted moving average chart (EWMA) control chart have been shown to be among the most effective univariate fault detection methods and more suitable for detection small faults. The objective of this work is to improve the PCA-based fault detection by using more sophisticated FD charts to achieve further improvements and widen the applicability of the process monitoring techniques in practice. The PCA presented here is investigated as modeling algorithm in the phase of fault detection. The fault detection problem is addressed so that the data are first modeled using the PCA algorithm and then the faults are detected using FD chart. The detection stage is related to the evaluation of detection charts, which are declares the presence of the fault. Those charts are computed using the PCA-based residual. The fault detection performance is illustrated through a simulated continuously stirred tank reactor (CSTR) data. The results demonstrate the effectiveness of the PCA-based FD chart methods for detecting the single and the multiple sensor faults.

Combining formant frequency based on variable order LPC coding with acoustic features for TIMIT phone recognition

Combination of multiple acoustic features has great potential to improve Automatic Speech Recognition (ASR) accuracy. Our contribution in this research was to investigate one novel method when using voiced formants’ features in combination with standard MFCC features in order to enhance TIMIT phone recognition. These voiced features provide accurate formants frequencies using a Variable Order LPC Coding (VO-LPC) algorithm that was combined with continuity constraints. The overall estimating formants were concatenated with MFCC features when a voiced frame could be detected. For feature-level combination, Heteroscedastic Linear Discriminant Analysis (HLDA) based approach had been used successfully to find an optimal linear combination of successive vectors of a single feature stream.A series of experiments on phone recognition speaker-independent continuous-speech had been carried out using a subset of the large read-speech TIMIT phone corpus. Hidden Markov Model Toolkit (HTK) was also used throughout all carried experiments. Using such feature level combination, optimized mixture splitting and a bigram language model, a detailed analysis on this combination performance was discussed for Context-Independent (CI) and Context-Dependent (CD) Hidden Markov Models (HMM). Experimental results from our proposed procedure showed that phone error rate was successfully decreased by about 3%. At phonetic level group, an increase of 8% and of 10% was observed respectively for vowel and liquid group. These results proved clear phone enhancement regarding existing state of the art.

Fully integrated CMOS data and clock recovery for wireless biomedical implants

In order to minimize the size and improve the efficiency of power consumption, most of wireless implantable Microsystems use Amplitude Shift Keying (ASK) modulator to transmit, through a RF link, data and energy to the internal implants. In this paper, we propose a new structure of wireless data and clock recovery dedicated for biomedical implants. It consists in a fully integrated ASK demodulator unit and a digital Manchester decoder. The demodulator is based on active components instead of the passive elements, as resistance or capacitance, to extract a robust data with a low modulation index. The whole digital Manchester decoder intends to recover data and clock from the demodulated signal. The following design has been validated by CADENCE environment using Specter simulation in a standard AMS 0.35μm CMOS process. The data rate is simulated by 1Mbps speed for 10MHz carrier frequency and 20% modulation index.

Efficient MLP constructive training algorithm using a neuron recruiting approach for isolated word recognition system

This paper describes an efficient constructive training algorithm using a Multi Layer Perceptron (MLP) neural network dedicated for Isolated Word Recognition (IWR) systems. Incremental training procedure was employed and this approach was based on novel hidden neurons recruiting for a single hidden-layer. During Neural Network (NN) training phase, the number of pronunciation samples extracted from the Training Data (TD) was sequentially increased. Optimal structure of the NN classifier with optimized TD size was obtained using this proposed MLP constructive training algorithm.Isolated word recognition system based on MLP neural network was then constructed and tested for recognizing ten words extracted from TIMIT database. Mel Frequency Cepstral Coefficient (MFCC) feature extraction method was employed including energy, first and second derivative coefficients.A proposed Frame-by-Frame Neural Network (FFNN) classification method was explored and compared with the Conventional Neural Network (CNN) classification approach. Principal Component Analysis (PCA) technique was also investigated in order to reduce both TD size as well as recognition system complexity.Experimental results showed superior performance of the proposed FFNN classifier compared to the CNN counter part which was illustrated by the significant improvement obtained in terms of recognition rate.

CDHMM parameters selection for speaker-independent phone recognition in continuous speech system

Pattern recognition has long been a topic of fundamental importance in a wide range of science and technology. Over the years there have been a range of several tasks developed for speech recognition. While in recent years speech recognizer evaluation has focused on LVCSR research, we believe that evaluating recognition at the phone level is important since the words are always represented by the concatenation of phones units. These phones are acoustically modeled by the predominant static model in automatic speech recognition remains the Hidden Markov Model ‘HMM’. In this paper, we investigate the behavior of speaker-independent phone recognition in continuous speech based on the technique of HMM. This study focus on the selection of an optimal model topology in order to achieve a robust phone recognition system which accomplishes the tradeoff between model size and data training. To evaluate and compare the performance of our conceived system to other previous works, we choose the standard TIMIT Database and the platform HTK. We obtain a phone recognition correct rate 69.33 percent and accuracy rate of 63.05 percent which are comparable with others works.

A distribution-matching approach to MRI brain tumor segmentation

This study investigates a fast distribution-matching algorithm for brain tumor segmentation. From a very simple user input, we learn a non-parametric model distribution which contains all the statistical information about the normal regions in the current brain image. We state the problem as the optimization of a cost function containing (1) an intensity distribution matching prior which measures a global similarity between non-parametric distributions, and (2) a smoothness prior to avoid the occurrence of small, isolated regions in the solution. Obtained following recent bound-relaxation results, the optimum of the cost function yields the complement of the tumor region in nearly real-time. Based on global rather than pixelwise information, the proposed algorithm does not require a complex learning from a large training set, as is the case in existing methods. Therefore, the ensuing results are independent of the choice of a training set. Quantitative evaluations and comparisons with several existing methods over publicly available data demonstrate that the proposed algorithm can yield a competitive performance.

A full digital low power dpsk demodulator and clock recovery circuit for high data rate neural implants

A novel full digital and non-coherent DPSK demodulator is presented for inductively powered biomedical systems. The transmitter uses differential phase encoding technique that requires in the demodulation, a precise symbol clock recovery. This was achieved by the detection of the rising and falling edges of the digitized received carrier. Very low power consumption and high data transmission rate are obtained with an excellent data-rate-to-carrier-frequency ratio of 100% without increasing the carrier frequency. The proposed demodulator is especially appropriate for high data rate biomedical applications such as visual prostheses and brain-machine interface. The circuit is designed in the 0.35-µm CMOS technology of Austria Micro Systems and it consumes 136.3µW @ 3.3 V at a data transmission rate of 10 Mbps.