China
Africa
Explore chapters and articles related to this topic
Introduction
Published in Narayan Panigrahi, Saraju P. Mohanty, Brain Computer Interface, 2022
Narayan Panigrahi, Saraju P. Mohanty
Brain imaging techniques or neuroimaging techniques allow doctors and researchers to view activity or problems within the human brain, without invasive neurosurgery. There are a number of accepted, safe imaging techniques in use today in research facilities and hospitals throughout the world. Prominent brain imaging techniques that are available to cognitive neuroscientists, including positron emission tomography (PET), near infrared spectroscopy (NIRS), magnetoencephalogram (MEG), electroencephalography (EEG), and functional magnetic resonance imaging (fMRI). We discuss most of the available neuroimaging techniques in this section but focus on EEG and fMRI because they are the most widely used techniques.
Deep Learning in Brain Segmentation
Published in Saravanan Krishnan, Ramesh Kesavan, B. Surendiran, G. S. Mahalakshmi, Handbook of Artificial Intelligence in Biomedical Engineering, 2021
Brain imaging (also known as neuroimaging), involves using noninvasive techniques to obtain the structural or functional image of the brain. Structural imaging of the brain refers to obtaining the structure of the central nervous system. Structural imaging is useful for detecting large-scale intracranial diseases such as a tumor or brain trauma. Two of the most common structural brain imaging techniques used in clinical settings are MRI and CT.
Neuroimaging techniques for brain analysis
Published in Munsif Ali Jatoi, Nidal Kamel, Brain Source Localization Using EEG Signal Analysis, 2017
This chapter deals with the discussion concerning neuroimaging techniques that are used to understand the brain's physiological and cognitive dynamics. Thus, the chapter first provides an introduction to overall functional brain imaging and its related applications. The chapter then moves on to describe briefly famous neuroimaging techniques such as functional magnetic resonance imaging (fMRI), magnetoencephalography (MEG), and positron emission tomography (PET). However, because we are here to discuss EEG-based source localization, EEG as a neuroimaging technique is discussed in detail with its particular application in brain source localization. After this, both forward and inverse problems for EEG source localization are described. Hence, the potential applications of EEG source localization are provided to learn about solving this ill-posed problem. Finally, a list of optimization algorithms that are used to localize the active brain sources is given.
Early Prediction of Progression to Alzheimer’s Disease using Multi-Modality Neuroimages by a Novel Ordinal Learning Model ADPacer
Published in IISE Transactions on Healthcare Systems Engineering, 2023
Lujia Wang, Zhiyang Zheng, Yi Su, Kewei Chen, David Weidman, Teresa Wu, ShihChung Lo, Fleming Lure, Jing Li
Neuroimaging has shown great promise to predict MCI progression to AD. Especially, images of different types/modalities measure different aspects of the brain affected by the disease. Combining data from different neuroimaging modalities has demonstrated improved prediction power than using a single modality alone. Two commonly used neuroimaging modalities are volumetric magnetic resonance imaging (MRI) and positron emission tomography (PET), which measure brain structure and function, respectively. There is an abundance of research for predicting MCI progression to AD by integrating MRI and FDG-PET—a type of PET image that measures cerebral glucose metabolism, together with some non-imaging data. The existing research typically formulates the progression prediction into a classification problem, i.e., to classify an MCI subject as a converter if the subject progresses/converts to AD within a pre-defined timeframe, and a non-converter otherwise. Next, we provide a brief review of this area, with a focus on methods developed in the recent few years.
Brain activation associated with eccentric movement: A narrative review of the literature
Published in European Journal of Sport Science, 2018
Lower EMG activity during eccentric muscle actions is modulated at both the spinal and cortical levels (Duclay, Pasquet, Martin, & Duchateau, 2011; Gruber, Linnamo, Strojnik, Rantalainen, & Avela, 2009). Based on changes in transcranial magnetic stimulation (TMS)-induced motor evoked potentials, the excitability of the corticospinal output neurons is reduced during the execution of eccentric muscle actions compared to concentric muscle actions (Sekiguchi, Kimura, Yamanaka, & Nakazawa, 2001). This depressed corticospinal excitability has been proposed to be controlled by higher-order cortical centres that are associated with the execution of eccentric muscle actions (Fang, Siemionow, Sahgal, Xiong, & Yue, 2004). Brain activation associated with eccentric muscle actions has not been investigated to a great extent. However, how the cortical regions in the motor network are activated during eccentric muscle actions may be critical for a better understanding in the cortical control of eccentric movement as well as for improvements in both performance and rehabilitation. This is an emerging topic that has only recently begun to be investigated through the advancements in brain-imaging methods. Two neuroimaging methods used in this context are electroencephalography (EEG) and functional magnetic resonance imaging (fMRI).
Improving Alzheimer’s disease classification by performing data fusion with vascular dementia and stroke data
Published in Journal of Experimental & Theoretical Artificial Intelligence, 2021
Zoran Bosnić, Brankica Bratić, Mirjana Ivanović, Marija Semnic, Iztok Oder, Vladimir Kurbalija, Tijana Vujanić Stankov, Vojislava Bugarski Ignjatović
Suk et al.’s (2014) used MRI and PET images to extract groups of voxels that are relevant for predicting the Alzheimer’s disease. They applied the multimodal Deep Boltzmann Machine to extract the attributes from the input data and feed them into the neural network. The features were then fed into the ensemble with SVM as a weak classifier (Liu et al., 2012). Again, models built on multimodal data outperformed the models built on only one modality. As expected, most of the recent research mainly focus on novel and advanced technologies like deep learning (Kim & Lee, 2018; Ning et al., 2018; Suk & Shen, 2013) where neuroimaging data are fused with other types of data.