Search

• Review of papers on the diagnosis of brain diseases using the fusion of neuroimaging modalities is presented. • For each brain disease, the most appropriate neuroimaging modalities and best combinations are introduced. • Advantages and disadvantages of fusion levels and different modalities are introduced. • Details of various types of deep neural networks used for the task at hand are also examined. • Finally, future directions and challenges in the field are discussed. Brain diseases, including tumors and mental and neurological disorders, seriously threaten the health and well-being of millions of people worldwide. Structural and functional neuroimaging modalities are commonly used by physicians to aid the diagnosis of brain diseases. In clinical settings, specialist doctors typically fuse the magnetic resonance imaging (MRI) data with other neuroimaging modalities for brain disease detection. As these two approaches offer complementary information, fusing these neuroimaging modalities helps physicians accurately diagnose brain diseases. Typically, fusion is performed between a functional and a structural neuroimaging modality. Because the functional modality can complement the structural modality information, thus improving the performance for the diagnosis of brain diseases by specialists. However, analyzing the fusion of neuroimaging modalities is difficult for specialist doctors. Deep Learning (DL) is a branch of artificial intelligence that has shown superior performances compared to more conventional methods in tasks such as brain disease detection from neuroimaging modalities. This work presents a comprehensive review paper in the field of brain disease detection from the fusion of neuroimaging modalities using DL models like convolutional neural networks (CNNs), recurrent neural networks (RNNs), pretrained, generative adversarial networks (GANs), and Autoencoders (AEs). First, neuroimaging modalities and the need for fusion are discussed. Then, review papers published in the field of neuroimaging multimodalities using AI techniques are explored. Moreover, fusion levels based on DL methods, including input, layer, and decision, with related studies conducted on diagnosing brain diseases, are discussed. Other sections present the most important challenges for diagnosing brain diseases from the fusion of neuroimaging modalities. In the discussion section, the details of previous research on the fusion of neuroimaging modalities based on MRI and DL models are reported. In the following, the most important future directions include Datasets, DA, imbalanced data, DL models, explainable AI, and hardware resources are presented. Finally, the main findings of this study are presented in the conclusion section. [ABSTRACT FROM AUTHOR]

In this paper, the domain of attraction (DOA) of the continuous-time positive polynomial fuzzy systems subject to input saturation is estimated by using the level set of the linear copositive Lyapunov function. To relax the estimation of DOA, the restriction on the level set is removed by embedding the expression of the level set into the stability conditions and positivity conditions. Referring to the nonconvex terms caused by above novel analysis strategy, some polynomial inequality lemmas are proposed to handle them; the nonconvex terms caused by imperfect premise matching (IPM) nonlinear membership functions are dealt with by sector nonlinear methods and advanced Chebyshev membership-function-dependent (MFD) methods. In this advanced MFD method, the state space segmentation and polynomial order selection of the Chebyshev approximation method are improved based on breakpoints of the first derivative and curvature, respectively, which is helpful to reduce the conservatism and computational burden of the result. Thus, this advanced Chebyshev MFD method not only optimizes the convexification strategy, but also can further be extended to estimate the DOA when it is used to introduce the membership functions information for convex stability and positivity conditions. Finally, a numerical example and the lipoprotein metabolism and potassium ion transfer nonlinear model are presented to validate the effectiveness and feasibility of the aforementioned analysis and convexification strategies in the expansion of DOA estimation. [ABSTRACT FROM AUTHOR]

Recently, wireless communication has faced with “green” challenge. The emergence of the “green wireless communication” means environment protection and energy saving. A “green wireless communication” is aimed at energy conservation and emissions reduction on the basis of reducing communication radiation and ensuring the quality of communication in the wireless communication network. According to related statistics, the energy consumption of the base station accounts for more than 70% in wireless communication network. So, it is important to reduce the energy consumption of the base station to realize the energy saving of mobile communication system. As a result, the base station sleeping strategy in coordinated multipoint (CoMP) communication is a promising method to solve this problem. In the base station sleeping strategy, base station with the most light traffic is off, and the users in this cell are served by the surrounding base stations with CoMP communications. However, since the downlink performance is also important for users, we should save the energy as well as keeping a perfect downlink performance. This paper presents a control theory to study the base station sleeping strategy optimization issues with CoMP communications. Specifically, to make decisions for the base station sleeping strategy, we apply the threshold method and reinforcement learning (RL) algorithm. We develop the multi-step Q-learning of the RL algorithm to optimize the base station sleeping strategy. Simulation results are provided to show the process and effectiveness of the proposed scheme. [ABSTRACT FROM AUTHOR]

This paper presents a neural network with a novel neuron model. In this model, the neuron has two activation functions and exhibits a node-to-node relationship in the hidden layer. This neural network provides better performance than a traditional feedforward neural network, and fewer hidden nodes are needed. The parameters of the proposed neural network are tuned by a genetic algorithm with arithmetic crossover and nonuniform mutation. Some applications are given to show the merits of the proposed neural network. [ABSTRACT FROM AUTHOR]

This paper addresses the H-infinity Takagi-Sugeno (T-S) fuzzy control for a class of T-S fuzzy discrete networked control systems with random interval communication delays and random sector nonlinearities. Firstly, the T-S fuzzy model is employed to approximate the discrete networked control system and the ℓth-order Rice fading channels model is introduced in the system model. Secondly, the T-S fuzzy dynamic output feedback controller with ℓth-order Rice fading channels output is designed for the T-S fuzzy discrete networked control system. Thirdly, the discrete delay-dependent Lyapunov-Krasovskii functional, stochastic system theory and Bernoulli probability distribution are employed to derive the stability conditions in terms of linear matrix inequalities (LMIs). Compared with previous works, the fading channels in the signal transmission are described clearly by setting the different channels coefficients of the ℓth-order Rice fading channels model. The closed-loop system is exponentially mean-square stable and prescribed H-infinity performance is guaranteed by designing the T-S fuzzy dynamic output feedback controller. The factorizations in the polynomial and the congruence transformation matrices are introduced to solve the LMIs, such that the controller gain matrices are determined. Finally, simulation examples are presented to show the effectiveness of proposed methods. [ABSTRACT FROM AUTHOR]

In this paper, an adaptive decentralized fuzzy control scheme is presented for a class of multiple-input and multiple-output (MIMO) nonlinear interconnected systems with unmodeled dynamics and dead zones. In the controller design procedure, fuzzy logic systems are used to approximate the unknown nonlinear functions and then an adaptive fuzzy controller is developed by combining adaptive technique with backstepping. The proposed control scheme can guarantee that all signals in the closed-loop system are semi-globally uniformly ultimately bounded in mean square. Simulation results are provided to illustrate the effectiveness of the proposed scheme. [ABSTRACT FROM AUTHOR]

Redundant actuated parallel robot is a multi-input and multi-output (MIMO) system which usually works in an uncertain environment. In this paper, the force/position hybrid control structure of 6PUS-UPU redundant actuation parallel robot is designed, and proportional–integral (PI) and model predictive control (MPC) cascade control strategies are used in the redundant branch of 6PUS-UPU redundant actuation parallel robot. The MPC algorithm is used in the current loop of the permanent magnet synchronous motor (PMSM) to restrain the motor parameter uncertainty and external disturbances influence on motor control. The MATLAB/ADAMS joint simulation method based on virtual 6PUS-UPU redundant actuation parallel robot prototype is used to test the performance of the proposed control strategy. The performance of proposed PI-MPC control strategy is compared with the traditional PI–PI control strategy. The simulation results show that the MPC controller can improve the tracking ability of the motor torque, guarantee the system robustness under the parameter variations and load disturbance environment. [ABSTRACT FROM AUTHOR]

The two important problems of SLAM and Path planning are often addressed independently. However, both are essential to achieve successfully autonomous navigation. In this paper, we aim to integrate the two attributes for application on a humanoid robot. The SLAM problem is solved with the EKF-SLAM algorithm whereas the path planning problem is tackled via Q -learning. The proposed algorithm is implemented on a NAO equipped with a laser head. In order to differentiate different landmarks at one observation, we applied clustering algorithm on laser sensor data. A Fractional Order PI controller (FOPI) is also designed to minimize the motion deviation inherent in during NAO’s walking behavior. The algorithm is tested in an indoor environment to assess its performance. We suggest that the new design can be reliably used for autonomous walking in an unknown environment. [ABSTRACT FROM AUTHOR]