Automated brain tumour segmentation from multi-modality magnetic resonance imaging data based on new particle swarm optimisation segmentation method.

BACKGROUND: Segmentation of brain tumours is a complex problem in medical image processing and analysis. It is a time-consuming and error-prone task. Therefore, computer-aided detection systems need to be developed to decrease physicians' workload and improve the accuracy of segmentation. METHODS: This paper proposes a level set method constrained by an intuitive artificial intelligence-based approach to perform brain tumour segmentation. By studying 3D brain tumour images, a new segmentation technique based on the Modified Particle Swarm Optimisation (MPSO), Darwin Particle Swarm Optimisation (DPSO), and Fractional Order Darwinian Particle Swarm Optimisation (FODPSO) algorithms were developed. RESULTS: The introduced technique was verified according to the MICCAI RASTS 2013 database for high-grade glioma patients. The three algorithms were evaluated using different performance measures: accuracy, sensitivity, specificity, and Dice similarity coefficient to prove the performance and robustness of our 3D segmentation technique. CONCLUSION: The result is that the MPSO algorithm consistently outperforms the DPSO and FO DPSO.

A sophisticated and provably grayscale image watermarking system using DWT-SVD domain.

Digital watermarking has attracted increasing attentions as it has been the current solution to copyright protection and content authentication in today's digital transformation, which has become an issue to be addressed in multimedia technology. In this paper, we propose an advanced image watermarking system based on the discrete wavelet transform (DWT) in combination with the singular value decomposition (SVD). Firstly, at the sender side, DWT is applied on a grayscale cover image and then eigendecomposition is performed on original HH (high-high) components. Similar operation is done on a grayscale watermark image. Then, two unitary and one diagonal matrices are combined to form a digital watermarked image applying inverse discrete wavelet transform (iDWT). The diagonal component of original image is transmitted through secured channel. At the receiver end, the watermark image is recovered using the watermarked image and diagonal component of the original image. Finally, we compare the original and recovered watermark image and obtained perfect normalized correlation. Simulation consequences indicate that the presented scheme can satisfy the needs of visual imperceptibility and also has high security and strong robustness against many common attacks and signal processing operations. The proposed digital image watermarking system is also compared to state-of-the-art methods to confirm the reliability and supremacy.

An Improved 3D Deep Learning-Based Segmentation of Left Ventricular Myocardial Diseases from Delayed-Enhancement MRI with Inclusion and Classification Prior Information U-Net (ICPIU-Net).

Accurate segmentation of the myocardial scar may supply relevant advancements in predicting and controlling deadly ventricular arrhythmias in subjects with cardiovascular disease. In this paper, we propose the architecture of inclusion and classification of prior information U-Net (ICPIU-Net) to efficiently segment the left ventricle (LV) myocardium, myocardial infarction (MI), and microvascular-obstructed (MVO) tissues from late gadolinium enhancement magnetic resonance (LGE-MR) images. Our approach was developed using two subnets cascaded to first segment the LV cavity and myocardium. Then, we used inclusion and classification constraint networks to improve the resulting segmentation of the diseased regions within the pre-segmented LV myocardium. This network incorporates the inclusion and classification information of the LGE-MRI to maintain topological constraints of pathological areas. In the testing stage, the outputs of each segmentation network obtained with specific estimated parameters from training were fused using the majority voting technique for the final label prediction of each voxel in the LGE-MR image. The proposed method was validated by comparing its results to manual drawings by experts from 50 LGE-MR images. Importantly, compared to various deep learning-based methods participating in the EMIDEC challenge, the results of our approach have a more significant agreement with manual contouring in segmenting myocardial diseases.

Model Predictive Control with Integrated Model Reduction for a Continuous Lactide Ring-Opening Polymerization Process.

Poly(lactic acid) production has received increasing attention, mainly due to its inherent biodegradable thermoplastic properties and to its renewable-resource-based composition. This process is affected by changes in the operating conditions and by raw material impurities which influence the reaction rate and degrade the polymer properties. As the system model is multivariable with coupled dynamics and constraints, linear model predictive control (LMPC) is employed here. A model reduction technique is proposed to obtain an approximate linear representation of the nonlinear system around the operating point to minimize the calculation cost of the controller. The proposed LMPC approach is validated by simulation and is compared to a proportional-integral controller and a nonlinear model predictive control. It is found that LMPC has a superior performance in terms of off-spec time when a disturbance occurs in the feed, and it can restore the target conditions better and faster.

Multi-objective generalized predictive control for switching systems under unknown switching sequences.

This paper describes a new generalized predictive control to track and stabilize a class of discrete time switching system. The focus is specifically centered on classes of switching systems characterized by unstable modes, undetermined switching signal, anon-minimumphase, and variable dead-times. To overcome this type of issue, an effective predictive control law is established by solving a dynamic multi-objective optimization function. Control problems are formulated in order to stabilize and regulate the system response around the targeted reference. The theoretical background of the proposed method is inspired by the standard generalized predictive control (GPC), in such a way that all desirable features are retained as possible. As a result, the obtained controller is more efficient in terms of stability and tracking. In fact within the framework of this research, the control problem has been formulated by taking into account behaviors of subsystems as well as the switching phase. The optimization of the problem is established in such a way that the obtained control law will be adapted to system dynamics, regardless of the mode. A number of simulation tests are established to evaluate the performance of the developed method. Four benchmark examples were considered for the simulation tests. Simulation results have shown the potential of the developed strategy to control and stabilize switching systems under unknown switching sequences. For further evaluation, the closed-loop performance of the developed strategy has been compared to that obtained with the Multi-Criteria Predictive Control (MOMPC) method. Comparison results have highlighted the effectiveness of the proposed method in terms of stability and tracking than MOMPC method.

Pole assignment stabilization for a class of switched nonlinear time-varying delay systems.

This paper investigates the pole assignment stabilization with memory state feedback controllers for switched nonlinear time-varying delay systems. More general time-varying delays which depend on the subsystem number are considered. Based on novel common Lyapunov functions (CLFs), the aggregation techniques, the proprieties and the Borne-Gentina criterion, new algebraic stabilization criteria are established. The proposed design can guarantee the stability of the closed-loop systems under arbitrary switching. It allows to avoid researching a common Lyapunov function (CLF), considered a difficult matter. Four illustrate examples were finally utilized to exhibit the effectiveness of the obtained results.

A New Mathematical Force Model that Predicts the Force-pulse Amplitude Relationship of Human Skeletal Muscle.

Current functional electrical stimulation (FES) systems vary the stimulation intensity to control the muscle force in order to produce precise functional movements. However, mathematical model that predicts the intensity effect on the muscle force is required for model-based controller design. The most previous force model designed by Ding et al was validated only for a standardized stimulation pulse amplitude (intensity). Thus, the purpose of this study was to adapt the Ding et al model to be able to predict the force-pulse amplitude relationship. The experimental results tested on quadriceps femoris muscles of healthy subjects (N=5) show that our adapted model accurately predicts the force response for trains of a wide range of stimulation intensities (30-100 mA). The accurate predictions indicate that our adapted model could be used for designing model-based control strategies to control the muscle force through FES.

An approach to a muscle force model with force-pulse amplitude relationship of human quadriceps muscles.

BACKGROUND: Recent advanced applications of the functional electrical stimulation (FES) mostly used closed-loop control strategies based on mathematical models to improve the performance of the FES systems. In most of them, the pulse amplitude was used as an input control. However, in controlling the muscle force, the most popular force model developed by Ding et al. does not take into account the pulse amplitude effect. The purpose of this study was to include the pulse amplitude in the existing Ding et al. model based on the recruitment curve function. METHODS: Quadriceps femoris muscles of eight healthy subjects were tested. Forces responses to stimulation trains with different pulse amplitudes (30-100 mA) and frequencies (20-80 Hz) were recorded and analyzed. Then, specific model parameter values were identified by fitting the measured forces for one train (50 Hz, 100 mA). The obtained model parameters were then used to identify the recruitment curve parameter values by fitting the force responses for different pulse amplitudes at the same frequency train. Finally, the extended model was used to predict force responses for a range of stimulation pulse amplitudes and frequencies. RESULTS: The experimental results indicated that our adapted model accurately predicts the force-pulse amplitude relationship with an excellent agreement between measured and predicted forces (R2=0.998, RMSE = 6.6 N). CONCLUSIONS: This model could be used to predict the pulse amplitude effect and to design control strategies for controlling the muscle force in order to obtain precise movements during FES sessions using intensity modulation.

Cuckoo Search Approach for Parameter Identification of an Activated Sludge Process.

A parameter identification problem for a hybrid model is presented. The latter describes the operation of an activated sludge process used for waste water treatment. Parameter identification problem can be considered as an optimization one by minimizing the error between simulation and experimental data. One of the new and promising metaheuristic methods for solving similar mathematical problem is Cuckoo Search Algorithm. It is inspired by the parasitic brood behavior of cuckoo species. To confirm the effectiveness and the efficiency of the proposed algorithm, simulation results will be compared with other algorithms, firstly, with a classical method which is the Nelder-Mead algorithm and, secondly, with intelligent methods such as Genetic Algorithm and Particle Swarm Optimization approaches.

TLBO-Based Adaptive Neurofuzzy Controller for Mobile Robot Navigation in a Strange Environment.

This work investigates the possibility of using a novel evolutionary based technique as a solution for the navigation problem of a mobile robot in a strange environment which is based on Teaching-Learning-Based Optimization. TLBO is employed to train the parameters of ANFIS structure for optimal trajectory and minimum travelling time to reach the goal. The obtained results using the suggested algorithm are validated by comparison with different results from other intelligent algorithms such as particle swarm optimization (PSO), invasive weed optimization (IWO), and biogeography-based optimization (BBO). At the end, the quality of the obtained results extracted from simulations affirms TLBO-based ANFIS as an efficient alternative method for solving the navigation problem of the mobile robot.

Extracting T-S Fuzzy Models Using the Cuckoo Search Algorithm.

A new method called cuckoo search (CS) is used to extract and learn the Takagi-Sugeno (T-S) fuzzy model. In the proposed method, the particle or cuckoo of CS is formed by the structure of rules in terms of number and selected rules, the antecedent, and consequent parameters of the T-S fuzzy model. These parameters are learned simultaneously. The optimized T-S fuzzy model is validated by using three examples: the first a nonlinear plant modelling problem, the second a Box-Jenkins nonlinear system identification problem, and the third identification of nonlinear system, comparing the obtained results with other existing results of other methods. The proposed CS method gives an optimal T-S fuzzy model with fewer numbers of rules.

M-matrix based robust stability and stabilization for uncertain discrete-time switched TS fuzzy systems with time-varying delays.

This paper provides novel sufficient conditions on robust asymptotic stability and stabilization for a class of uncertain discrete-time switched fuzzy with time-varying delays. The attention is focused on developing new algebraic criteria to break with classical criteria in terms of Linear Matrix Inequalities (LMIs). Firstly, based on the M-matrix proprieties and through l1,∞ induced norms notion, new delay-dependent sufficient conditions are derived to ensure the asymptotic stability and stabilization for a class of uncertain discrete-time switched fuzzy systems with time-varying delay. Secondly, these results are extended for a class of uncertain discrete-time switched fuzzy systems with time delays, modeled by difference equations. Finally, two numerical examples and practical example (a robot arm) are provided to demonstrate the advantage and the effectiveness of our results.

Stability and stabilization studies for a class of switched nonlinear systems via vector norms approach.

This paper is concerned with the problems of stability analysis and stabilization with a state feedback controller through pole placement for a class of both continuous and discrete-time switched nonlinear systems. These systems are modeled by differential or difference equations. Then, a transformation under the arrow form is employed. Note that, the main contribution in this work is twofold: firstly, based on the construction of an appropriated common Lyapunov function, as well the use of the vector norms notion, the recourse to the Kotelyanski lemma, the M-matrix proprieties, the aggregation techniques and the application of the Borne-Gentina criterion, new sufficient stability conditions under arbitrary switching for the autonomous system are deduced. Secondly, this result is extended for designing a state feedback controller by using pole assignment control, which guarantee that the corresponding closed-loop system is globally asymptotically stable under arbitrary switching. The main novelties features of these obtained results are the explicitness and the simplicity in their application. Moreover, they allow us to avoid the search of a common Lyapunov function which is a difficult matter. Finally, as validation to stabilize a shunt DC motor under variable mechanical loads is performed to demonstrate the effectiveness of the proposed results.