Privacy Preserved Brain Disorder Diagnosis Using Federated Learning

Federated learning has recently attracted significant attention as a cutting-edge technology that enables Artificial Intelligence(AI)algorithms to utilize global learning across the data of numerous individuals while safeguarding user data privacy.Recent advanced healthcare technologies have enabled the early diagnosis of various cognitive ailments like Parkinson’s.Adequate user data is frequently used to train machine learning models for healthcare systems to track the health status of patients.The healthcare industry faces two significant challenges:security and privacy issues and the personalization of cloud-trained AI models.This paper proposes a Deep Neural Network(DNN)based approach embedded in a federated learning framework to detect and diagnose brain disorders.We extracted the data from the database of Kay Elemetrics voice disordered and divided the data into two windows to create training models for two clients,each with different data.To lessen the over-fitting aspect,every client reviewed the outcomes in three rounds.The proposed model identifies brain disorders without jeopardizing privacy and security.The results reveal that the global model achieves an accuracy of 82.82%for detecting brain disorders while preserving privacy.

Understanding that Addiction Is a Brain Disorder Offers Help and Hope

We refute the controversial statement that addiction is not a brain disorder. Extensive peer-reviewed studies support the underlying neurobiological and neurogenetic basis of addiction’s “disease model”. In the 70s and 80s, a few clinical scientists suggested that it is possible to use behavioral training to teach controlled drinking. However, this controversial model failed drastically and increased labeling and stigmatization. Additionally, it was unhelpful in the search for treatment. Instead, we assert that addiction is a neuropsychiatric disorder characterized by a recurring desire to continue taking substances despite harmful physical and mental consequences. Work from our laboratory in 1995 supported the Reward Deficiency Syndrome (RDS) concept based on a common neurogenetic mechanism (hypodopaminergia) that underlies all substance and non-substance addictions. Non-substance addictions include behaviors like pathological gambling, internet addiction, and mobile phone addiction. Certain impulsive and compulsive behaviors or the acute intake of psychoactive substances result in heightened dopaminergic activity, while the opposite, hypodopaminergia, occurs following chronic abuse. Patients with Substance Use Disorder (SUD) can have a genetic predisposition compounded by stress or other epigenetic insults that can impact recovery. Relapse will occur post-short-term recovery if dopaminergic dysfunction remains untreated. Addiction, a brain disorder, requires treatment with DNA-directed pro-dopamine regulation and rehabilitation.

Bioengineering extracellular vesicles as novel nanocarriers towards brain disorders

Despite noteworthy technological progress and promising preclinical trials,brain disorders are still the leading causes of death globally.Extracellular vesicles(EVs),nano-/micro-sized membrane vesicles carrying bioactive molecules,are involved in cellular communication.Based on their unique properties,including superior biocompatibility,non-immunogenicity,and blood-brain barrier(BBB)penetration,EVs can shield their cargos from immune clearance and transport them to specific site,which have attracted increasing interests as novel nanocarriers for brain disorders.However,considering the limitations of native EVs,such as poor encapsulation efficiency,inadequate targeting capability,uncontrolled drug release,and limited production,researchers bioengineer EVs to fully exploit the clinical potential.Herein,this review initially describes the basic properties,biogenesis,and uptake process of EVs from different subtypes.Then,we highlight the application of EVs derived from different sources for personalized therapy and novel strategies to construct bioengineered EVs for enhanced diagnosis and treatment of brain disorders.Besides,it also presents a systematic comparison between EVs and other brain-targeted nanocarriers.Finally,existing challenges and future perspectives of EVs have been discussed,hoping to bolster the research from benchtop to bedside.

Non-human Primate Models for Brain Disorders – Towards Genetic Manipulations via Innovative Technology

Modeling brain disorders has always been one of the key tasks in neurobiological studies. A wide range of organisms including worms, fruit ？ies, zebra？sh, and rodents have been used for modeling brain disorders. However,whether complicated neurological and psychiatric symptoms can be faithfully mimicked in animals is still debatable.In this review, we discuss key ？ndings using non-human primates to address the neural mechanisms underlying stress and anxiety behaviors, as well as technical advances for establishing genetically-engineered non-human primate models of autism spectrum disorders and other disorders.Considering the close evolutionary connections and similarity of brain structures between non-human primates and humans, together with the rapid progress in genome-editing technology, non-human primates will be indispensable for pathophysiological studies and exploring potential therapeutic methods for treating brain disorders.

Connecting neurodevelopment to neurodegeneration:a spotlight on the role of kinesin superfamily protein 2A(KIF2A)

Microtubules play a central role in cytoskeletal changes during neuronal development and maintenance.Microtubule dynamics is essential to polarity and shape transitions underlying neural cell division,differentiation,motility,and maturation.Kinesin superfamily protein 2A is a member of human kinesin 13 gene family of proteins that depolymerize and destabilize microtubules.In dividing cells,kinesin superfamily protein 2A is involved in mitotic progression,spindle assembly,and chromosome segregation.In postmitotic neurons,it is required for axon/dendrite specification and extension,neuronal migration,connectivity,and survival.Humans with kinesin superfamily protein 2A mutations suffer from a variety of malformations of cortical development,epilepsy,autism spectrum disorder,and neurodegeneration.In this review,we discuss how kinesin superfamily protein 2A regulates neuronal development and function,and how its deregulation causes neurodevelopmental and neurological disorders.

Extracting Multiple Nodes in a Brain Region of Interest for Brain Functional Network Estimation and Classification

Purpose: Brain functional networks (BFNs) has become important approach for diagnosis of some neurological or psychological disorders. Before estimating BFN, obtaining blood oxygen level dependent (BOLD) representative signals from brain regions of interest (ROIs) is important. In the past decades, the common method is generally to take a ROI as a node, averaging all the voxel time series inside it to extract a representative signal. However, one node does not represent the entire information of this ROI, and averaging method often leads to signal cancellation and information loss. Inspired by this, we propose a novel model extraction method based on an assumption that a ROI can be represented by multiple nodes. Methods: In this paper, we first extract multiple nodes (the number is user-defined) from the ROI based on two traditional methods, including principal component analysis (PCA), and K-means (Clustering according to the spatial position of voxels). Then, canonical correlation analysis (CCA) was issued to construct BFNs by maximizing the correlation between the representative signals corresponding to the nodes in any two ROIs. Finally, to further verify the effectiveness of the proposed method, the estimated BFNs are applied to identify subjects with autism spectrum disorder (ASD) and mild cognitive impairment (MCI) from health controls (HCs). Results: Experimental results on two benchmark databases demonstrate that the proposed method outperforms the baseline method in the sense of classification performance. Conclusions: We propose a novel method for obtaining nodes of ROId based on the hypothesis that a ROI can be represented by multiple nodes, that is, to extract the node signals of ROIs with K-means or PCA. Then, CCA is used to construct BFNs.

Sensor-Based Gait Analysis for Parkinson’s Disease Prediction

Parkinson’s disease is identified as one of the key neurodegenerative disorders occurring due to the damages present in the central nervous system.The cause of such brain damage seems to be fully explained in many research studies,but the understanding of its functionality remains to be impractical.Specifically,the development of a quantitative disease prediction model has evolved in recent decades.Moreover,accelerometer sensor-based gait analysis is accepted as an important tool for recognizing the walking behavior of the patients during the early prediction and diagnosis of Parkinson’s disease.This type of minimal infrastructure equipment helps in analyzing the Parkinson’s gait properties without affecting the common behavioral patterns during the clinical practices.Therefore,the Accelerometer Sensor-based Parkinson’s Disease Identi-fication System(ASPDIS)is introduced with a kernel-based support vector machine classifier model to make an early prediction of the disease.consequently,the proposed classifier can easily predict various severity levels of Parkinson’s disease from the sensor data.The performance of the proposed classifier is com-pared against the existing models such as random forest,decision tree,and k-near-est neighbor classifiers respectively.As per the experimental observation,the proposed classifier has more capability to differentiate Parkinson’s from non-Parkinson patients depending upon the severity levels.Also,it is found that the model has outperformed the existing classifiers concerning prediction time and accuracy respectively.

Multimodal Fusion of Brain Imaging Data: Methods and Applications

Neuroimaging data typically include multiple modalities,such as structural or functional magnetic resonance imaging,dif-fusion tensor imaging,and positron emission tomography,which provide multiple views for observing and analyzing the brain.To lever-age the complementary representations of different modalities,multimodal fusion is consequently needed to dig out both inter-modality and intra-modality information.With the exploited rich information,it is becoming popular to combine multiple modality data to ex-plore the structural and functional characteristics of the brain in both health and disease status.In this paper,we first review a wide spectrum of advanced machine learning methodologies for fusing multimodal brain imaging data,broadly categorized into unsupervised and supervised learning strategies.Followed by this,some representative applications are discussed,including how they help to under-stand the brain arealization,how they improve the prediction of behavioral phenotypes and brain aging,and how they accelerate the biomarker exploration of brain diseases.Finally,we discuss some exciting emerging trends and important future directions.Collectively,we intend to offer a comprehensive overview of brain imaging fusion methods and their successful applications,along with the chal-lenges imposed by multi-scale and big data,which arises an urgent demand on developing new models and platforms.

Brain asymmetry: a novel perspective on hemispheric network

Brain asymmetry,involving structural and functional differencesbetween the two hemispheres,is a major organizational principle ofthe human brain.The structural and functional connectivity withineach hemisphere defines the hemispheric network or connectome.Elucidating left-right differences of the hemispheric network providesopportunities for brain asymmetry exploration.This review examinesthe asymmetry in the hemispheric white matter and functionalnetwork to assess health and brain disorders.In this article,the brain asymmetry in structural and functional connectivity includingnetwork topologies of healthy individuals,involving brain cognitivesystems and the development trend,is highlighted.Moreover,theabnormal asymmetry of the hemispheric network related to cognition changes in brain disorders,such as Alzheimer’s disease,schizophrenia,autism spectrum disorder,attention deficit hyperactivity disorder,and bipolar disorder,is presented.This review suggests that thehemispheric network is highly conserved for measuring human brain asymmetries and has potential in the study of the cognitivesystem and brain disorders.

Predictive power of abnormal electroencephalogram for post-cerebral infarction depression

Electroencephalography is a sensitive indicator for measuring brain condition, and can reflect early changes in brain function and severity of cerebral ischemia. However, it is not yet known whether electroencephalography can predict development of post-cerebral infarc- tion depression. A total of 321 patients with ischemic stroke underwent electroencephalography and Hamilton Depression Rating Scale assessment to analyze the relationship between electroencephalography and post-cerebral infarction depression. Our results show that electroencephalograms of ischemic stroke patients with depression exhibit low-amplitude alpha activity and slow theta activity. In con- trast, electroencephalograms of ischemic stroke patients without depression show fast beta activity and slow delta activity. ＂Ihese findings confirm that low-amplitude alpha activity and slow theta activity can be considered as independent predictors for post-cerebral infarction depression.

Inattentiveness in attention-deficit/hyperactivity disorder

Attention-deficit/hyperactivity disorder （ADHD） is a neurodevelopmental disorder with a long-term impact on functioning, productivity and quality of life of patients. This impact is largely due to the symptoms of inattentiveness. However, despite its impairing role in the lives of ADHD patients, inattentiveness has been studied relatively less frequently than have symptoms of impulsivity/hyperactivity and problems with executive function. This review therefore seeks to integrate the neuropsychological theories and current findings in the research fields of neuropsychology, neurophysiology, and neuroimaging, in an attempt to gain a more complete understanding of the role that inattentiveness plays in ADHD, as well as to suggest directions for future studies. The need for a more comprehensive understanding of inattentiveness and ADHD, which integrates findings from each of the three disciplines mentioned above, is emphasized.

Developing DNA methylation-based diagnostic biomarkers

An emerging paradigm shift for disease diagnosis is to rely on molecular characterization beyond traditional clinical and symptom-based examinations. Although genetic alterations and transcription signature were first introduced as potential biomarkers, clinical implementations of these markers are limited due to low reproducibility and accuracy. Instead, epigenetic changes are considered as an alternative approach to disease diagnosis. Complex epigenetic regulation is required for normal biological functions and it has been shown that distinctive epigenetic disruptions could contribute to disease pathogenesis. Disease-specific epigenetic changes, especially DNA methylation, have been observed,suggesting its potential as disease biomarkers for diagnosis. In addition to specificity, the feasibility of detecting disease-associated methylation marks in the biological specimens collcted noninvasively,such as blood samples, has driven the clinical studies to validate disease-specific DNA methylation changes as a diagnostic biomarker. Here, we highlight the advantages of DNA methylation signature for diagnosis in different diseases and discuss the statistical and technical challenges to be overcome before clinical implementation.

The therapeutic prospects and challenges of human neural stem cells for the treatment of Alzheimer’s Disease

Alzheimer’s disease(AD)is a multifactorial neurodegenerative disorder associated with aging.Due to its insidious onset,protracted progression,and unclear pathogenesis,it is considered one of the most obscure and intractable brain disorders,and currently,there are no effective therapies for it.Convincing evidence indicates that the irreversible decline of cognitive abilities in patients coincides with the deterioration and degeneration of neurons and synapses in the AD brain.Human neural stem cells(NSCs)hold the potential to functionally replace lost neurons,reinforce impaired synaptic networks,and repair the damaged AD brain.They have therefore received extensive attention as a possible source of donor cells for cellular replacement therapies for AD.Here,we review the progress in NSC-based transplantation studies in animal models of AD and assess the therapeutic advantages and challenges of human NSCs as donor cells.We then formulate a promising transplantation approach for the treatment of human AD,which would help to explore the disease-modifying cellular therapeutic strategy for the treatment of human AD.