Transformers in medical image analysis

Intelligent Medicine, Год журнала: 2022, Номер 3(1), С. 59 - 78

Опубликована: Авг. 24, 2022

Transformers have dominated the field of natural language processing and recently made an impact in area computer vision. In medical image analysis, transformers also been successfully used to full-stack clinical applications, including synthesis/reconstruction, registration, segmentation, detection, diagnosis. This paper aimed promote awareness applications analysis. Specifically, we first provided overview core concepts attention mechanism built into other basic components. Second, reviewed various transformer architectures tailored for discuss their limitations. Within this review, investigated key challenges use different learning paradigms, improving model efficiency, coupling with techniques. We hope review would provide a comprehensive picture readers interest

Язык: Английский

---

Classification of autism spectrum disorder by combining brain connectivity and deep neural network classifier

Neurocomputing, Год журнала: 2018, Номер 324, С. 63 - 68

Опубликована: Май 24, 2018

Язык: Английский

---

Networks behind the morphology and structural design of living systems

Physics of Life Reviews, Год журнала: 2022, Номер 41, С. 1 - 21

Опубликована: Март 10, 2022

Technological advances in imaging techniques and biometric data acquisition have enabled us to apply methods of network science study the morphology structural design organelles, organs, tissues, as well coordinated interactions among them that yield a healthy physiology at level whole organisms. We here review research dedicated these advances, particular focusing on networks between cells, topology multicellular structures, neural interactions, fluid transportation networks, anatomical networks. The percolation blood vessels, connectivity within brain, porous structure bones, relations different parts human body are just some examples we explore detail. argue show models, methods, algorithms developed realm ushering new era network-based inquiry into living systems broadest possible terms. also emphasize need applicability this is likely increase significantly years come due rapid progress made development bioartificial substitutes tissue engineering.

Язык: Английский

---

A Graph Theory-Based Modeling of Functional Brain Connectivity Based on EEG: A Systematic Review in the Context of Neuroergonomics

IEEE Access, Год журнала: 2020, Номер 8, С. 155103 - 155135

Опубликована: Янв. 1, 2020

Graph theory analysis, a mathematical approach, has been applied in brain connectivity studies to explore the organization of network patterns. The computation graph metrics enables characterization stationary behavior electroencephalogram (EEG) signals that cannot be explained by simple linear methods. main purpose this study was systematically review applications for mapping functional EEG data neuroergonomics. Moreover, article proposes pipeline constructing an unweighted from using both source and sensor Out 57 articles, our results show used characterize have attracted increasing attention since 2006, with highest frequency publications 2018. Most focused on cognitive tasks comparison motor tasks. mean phase coherence method, based “phase-locking value,” most frequently estimation technique reviewed studies. Furthermore, received substantially more literature than weighted network. global clustering coefficient characteristic path length were prevalent differentiating between integration local segregation, small-worldness property emerged as compelling metric information processing. This provides insight into use model context neuroergonomics research.

Язык: Английский

---

Diagnosis of Alzheimer Disease Using 2D MRI Slices by Convolutional Neural Network

Applied Bionics and Biomechanics, Год журнала: 2021, Номер 2021, С. 1 - 9

Опубликована: Фев. 2, 2021

There are many kinds of brain abnormalities that cause changes in different parts the brain. Alzheimer's disease is a chronic condition degenerates cells leading to memory asthenia. Cognitive mental troubles such as forgetfulness and confusion one most important features patients. In literature, several image processing techniques, well machine learning strategies, were introduced for diagnosis disease. This study aimed at recognizing presence based on magnetic resonance imaging We adopted deep methodology discrimination between patients healthy from 2D anatomical slices collected using imaging. Most previous researches implementation 3D convolutional neural network, whereas we incorporated usage input network. The data set this research was obtained OASIS website. trained network structure exhibit weightings named Alzheimer Network (AlzNet). accuracy our enhanced 99.30%. work investigated effects parameters AlzNet, number layers, filters, dropout rate. results interesting after performance metrics evaluating proposed AlzNet.

Язык: Английский

---

Diagnosis of Alzheimer’s Disease Severity with fMRI Images Using Robust Multitask Feature Extraction Method and Convolutional Neural Network (CNN)

Computational and Mathematical Methods in Medicine, Год журнала: 2021, Номер 2021, С. 1 - 15

Опубликована: Апрель 27, 2021

The automatic diagnosis of Alzheimer’s disease plays an important role in human health, especially its early stage. Because it is a neurodegenerative condition, seems to have long incubation period. Therefore, essential analyze symptoms at different stages. In this paper, the classification done with several methods machine learning consisting $K$ -nearest neighbor (KNN), support vector (SVM), decision tree (DT), linear discrimination analysis (LDA), and random forest (RF). Moreover, novel convolutional neural network (CNN) architecture presented diagnose severity. relationship between patients’ functional magnetic resonance imaging (fMRI) images their scores on MMSE investigated achieve aim. feature extraction performed based robust multitask algorithm. severity also calculated Mini-Mental State Examination score, including low, mild, moderate, severe categories. Results show that accuracy KNN, SVM, DT, LDA, RF, CNN method 77.5%, 85.8%, 91.7%, 79.5%, 85.1%, 96.7%, respectively. for architecture, sensitivity status Alzheimer patients 98.1%, 95.2%,89.0%, 87.5%, Based findings, classifier outperforms other can stages maximum accuracy.

Язык: Английский

---

A biodegradable and self-deployable electronic tent electrode for brain cortex interfacing

Nature Electronics, Год журнала: 2024, Номер 7(9), С. 815 - 828

Опубликована: Авг. 5, 2024

Язык: Английский

---

A novel joint HCPMMP method for automatically classifying Alzheimer’s and different stage MCI patients

Behavioural Brain Research, Год журнала: 2019, Номер 365, С. 210 - 221

Опубликована: Март 2, 2019

Язык: Английский

---

Effects of HD‐tDCS on Resting‐State Functional Connectivity in the Prefrontal Cortex: An fNIRS Study

Complexity, Год журнала: 2018, Номер 2018(1)

Опубликована: Янв. 1, 2018

Functional connectivity is linked to several degenerative brain diseases prevalent in our aging society. Electrical stimulation used for the clinical treatment and rehabilitation of patients with many cognitive disorders. In this study, effects high‐definition transcranial direct current (HD‐tDCS) on resting‐state networks human prefrontal cortex were investigated by using functional near‐infrared spectroscopy (fNIRS). The intrahemispheric as well interhemispheric changes induced 1 mA HD‐tDCS examined 15 healthy subjects. Pearson correlation coefficient‐based matrices generated from filtered time series oxyhemoglobin (ΔHbO) signals converted into binary matrices. Common graph theory metrics computed evaluate network changes. Systematic interhemispheric, intrahemispheric, intraregional analyses demonstrated that positively affected cortex. poststimulation was increased throughout region, while focal an rate stimulated hemisphere. clearly distinguished prestimulation a range thresholds. results study suggest can be increase explored clinically neurorehabilitation diseases.

Язык: Английский

---

Enhancing the feature representation of multi-modal MRI data by combining multi-view information for MCI classification

Neurocomputing, Год журнала: 2020, Номер 400, С. 322 - 332

Опубликована: Март 14, 2020

Язык: Английский

---