A Comprehensive Analysis of Psychiatric Disorders Using Deep Learning DOI
Nandini Manickam, Vijayakumar Ponnusamy

Опубликована: Дек. 29, 2023

In recent days, advancements in science, technology, and the social environment have played a vital role changing emotional well-being of person's life. These exposures not only create awareness about technologies but also lead to problems handling emotions, relationships, anxiety, depression. Mental health is essential at every stage life which helps determine how handle stress relationships. Psychiatric evaluation professionals identify severity mental through interview sessions based on questionnaires. Deep learning techniques can help classify diagnose psychological disorders brain signals received from MRI electrode EEG. This supports accurate prediction diagnosis. A comprehensive review various psychiatric analysis methods classifications presented.

Язык: Английский

The history, current state and future possibilities of the non-invasive brain computer interfaces DOI Creative Commons

Frederico Caiado,

А. И. Уколов

Medicine in Novel Technology and Devices, Год журнала: 2025, Номер 25, С. 100353 - 100353

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

1
Artifact removal and motor imagery classification in EEG using advanced algorithms and modified DNN DOI Creative Commons
Srinath Akuthota,

K. Rajkumar,

Janapati Ravichander

и другие.

Heliyon, Год журнала: 2024, Номер 10(7), С. e27198 - e27198

Опубликована: Март 19, 2024

This paper presents an advanced approach for EEG artifact removal and motor imagery classification using a combination of Four Class Iterative Filtering Filter Bank Common Spatial Pattern Algorithm with Modified Deep Neural Network (DNN) classifier. The research aims to enhance the accuracy reliability BCI systems by addressing challenges posed artifacts complex tasks.The methodology begins introducing FCIF, novel technique ocular removal, utilizing iterative filtering filter banks. FCIF's mathematical formulation allows effective mitigation, thereby improving quality data. In tandem, FC-FBCSP algorithm is introduced, extending handle four-class classification. DNN classifier enhances discriminatory power features, optimizing process.The showcases comprehensive experimental setup, featuring utilization Competition IV Dataset 2a & 2b. Detailed preprocessing steps, including feature extraction, are presented rigor. Results demonstrate remarkable capabilities FCIF prowess combined Comparative analysis highlights superiority proposed over baseline methods achieves mean 98.575%

Язык: Английский

Процитировано

3
Evaluating Deep Learning with different feature scaling techniques for EEG-based Music Entrainment Brain Computer Interface DOI Creative Commons

C. Rashmi,

Shantala C P

e-Prime - Advances in Electrical Engineering Electronics and Energy, Год журнала: 2024, Номер 7, С. 100448 - 100448

Опубликована: Янв. 28, 2024

Music Entrainment Brain-Computer Interface (BCI) systems influence music as a modulatory tool, synchronizing neural activities to the rhythm and structure of auditory stimuli. This innovative interface uses electroencephalogram (EEG) signals decode cognitive states influenced by music, enabling novel pathways for enhancement, stress reduction, therapeutic interventions. paper investigates different feature scaling techniques on performance deep learning model within EEG-based systems. Deep network (DNN) is implemented classify EEG into three classes i) during listening ii) singing bowl sounds iii) relax states. The comparison effect both therapy brain lobes, such frontal, temporal, central occipital lobes are analyzed. DNN evaluated employing various methods like StandardScaler(), MinMaxScaler(), Normalizer(), RobustScaler(). loss computed using four functions mean squared error(MSE), absolute error (MAE), logcosh categorical cross entropy. StandardScaler with function showed test accuracy 87.26%. research offers valuable insights BCI's potential in management integration tool.

Язык: Английский

Процитировано

2
Effects of Different Culture Densities on the Acoustic Characteristics of Micropterus salmoide Feeding DOI Creative Commons

Renyu Qi,

Huang Liu, Shijing Liu

и другие.

Fishes, Год журнала: 2023, Номер 8(3), С. 126 - 126

Опубликована: Фев. 23, 2023

The intensity and frequency of the acoustic signals generated by different behaviors largemouth bass (Micropterus salmoides) have characteristics. during feeding can be used to analyze characteristic patterns their behavior, which provide a theoretical basis for applications such as automatic based on signals. We passive acoustics combined with video study in recirculating water culture system (4, 8, 12, 16 fish/m3). result time–frequency power spectrum analysis sound showed that short-time average amplitude signal was well distinguished from background noise, both swallowing chewing sounds were positively correlated density, correlation between number fish stronger; at densities, zero-crossing phase suddenly dropped about 500 rose 1000 process. Therefore, parameters automatically identify process signal. entropy maintained 4–6 densities. In spectrum, main sounding frequencies farming densities distinguishable spectral range noised ranged 1 20 kHz, peak within 1.2 3.0 value density.

Язык: Английский

Процитировано

6
An Optimization-Linked Intelligent Security Algorithm for Smart Healthcare Organizations DOI Open Access

Reyazur Rashid Irshad,

Ahmed Abdu Alattab, Omar Ali Saleh Alsaiari

и другие.

Healthcare, Год журнала: 2023, Номер 11(4), С. 580 - 580

Опубликована: Фев. 15, 2023

IoT-enabled healthcare apps are providing significant value to society by offering cost-effective patient monitoring solutions in buildings. However, with a large number of users and sensitive personal information readily available today's fast-paced, internet, cloud-based environment, the security these systems must be top priority. The idea safely storing patient's health data an electronic format raises issues regarding privacy security. Furthermore, traditional classifiers, processing amounts is difficult challenge. Several computational intelligence approaches useful for effectively categorizing massive quantities this goal. For many reasons, novel system that tracks disease processes forecasts diseases based on obtained from patients distant communities proposed study. framework consists three major stages, namely collection, secured storage, detection. collected using IoT sensor devices. After that, homomorphic encryption (HE) model used storage. Finally, detection designed help Centered Convolutional Restricted Boltzmann Machines-based whale optimization (CCRBM-WO) algorithm. experiment conducted Python-based cloud tool. outperforms current e-healthcare solutions, according findings experiments. accuracy, precision, F1-measure, recall our suggested technique 96.87%, 97.45%, 97.78%, 98.57%, respectively, method.

Язык: Английский

Процитировано

4
Identifying the Risk in Lie Detection for Assessing Guilty and Innocent Subjects for Healthcare Applications DOI
Tanmayi Nagale, Anand Khandare

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

0
Membaca Sinyal Electroencephalogram (EEG) Dalam Menangkap Tingkat Emosi (Berdasarkan Ontologi) DOI Creative Commons
Yudo Devianto, Eko Sediyono, Sri Yulianto Joko Prasetyo

и другие.

Faktor Exacta, Год журнала: 2024, Номер 17(2), С. 152 - 152

Опубликована: Июль 17, 2024

Philosophically based electroencephalography (EEG) signal data processing is an interdisciplinary approach that opens up new perspectives in understanding brain function.In this context, it necessary to examine from a technical or biological perspective and consider its metaphysical, epistemological, ontological aspects.Ontology branch of metaphysics deals with objects the types exist according metaphysical (or even physical) theory, their properties, relationship.This article attempts provide philosophical view science on ontology for EEG data, source which waves.With results trials using Artificial Neural Network (ANN) classification, accuracy value 46.73 was obtained.The Convolutional (CNN) algorithm can also be used process determine person's emotional level; has been proven previous research.Although overall emotion recognition increased significantly, several problems have caused low DEAP DREAMER datasets.Other experiments conducted CNN, experimental show weight channels related emotions greater than different channels.Continuous Capsule (CCN) Deep (DNN) algorithms level emotion.

Язык: Английский

Процитировано

0
A hybrid network using transformer with modified locally linear embedding and sliding window convolution for EEG decoding DOI

Ketong Li,

Peng Chen, Chen Qian

и другие.

Journal of Neural Engineering, Год журнала: 2024, Номер 21(6), С. 066049 - 066049

Опубликована: Дек. 1, 2024

. Brain-computer interface(BCI) is leveraged by artificial intelligence in EEG signal decoding, which makes it possible to become a new means of human-machine interaction. However, the performance current decoding methods still insufficient for clinical applications because inadequate information extraction and limited computational resources hospitals. This paper introduces hybrid network that employs transformer with modified locally linear embedding sliding window convolution decoding.

Язык: Английский

Процитировано

0
A deep learning approach to detect the electroencephalogram-based cognitive task states DOI Open Access
Hitesh Yadav, Surita Maini

International Journal of Advanced Technology and Engineering Exploration, Год журнала: 2023, Номер 10(100)

Опубликована: Март 31, 2023

Brain-computer interface (BCI) is an important topic for researchers and the scientific community, as indicated by abundance of research study materials in field.The purpose BCI to allow interaction with any device or computer via brain signals.According this definition, strives collect signals using sensors, analyze process these received signals, then extract features operate device.Simply, it a link between device.The user can control brain's neural activities.*Author correspondence was first developed biomedical applications enable physically impaired persons move around substituting lost motor functions [1].Nowadays, includes non-medical well [2, 3].Newer areas include lie detection, drowsiness cognitive studies, imagery, virtual reality, video games, driver fatigue stress many more.From applications, ability understanding functioning.Cognitive depends from person essential controlling various mental activities [4].BCI has been accelerated technological advances enabling processing observing [5].Any task reveals how thinks, utilizes,

Язык: Английский

Процитировано

1
Inter Subject Emotion Recognition Using Spatio-Temporal Features From EEG Signal DOI
Mohammad Asif,

Diya Srivastava,

Aditya Gupta

и другие.

Опубликована: Сен. 14, 2023

Inter-subject or subject-independent emotion recognition has been a challenging task in affective computing. This work is about an easy-to-implement model that classifies emotions from EEG signals subject independently. It based on the famous EEGNet architecture, which used EEG-related BCIs. We 'Dataset Emotion using Naturalistic Stimuli' (DENS) dataset. The dataset contains 'Emotional Events'- precise information of timings participants felt. combination regular, depthwise and separable convolution layers CNN to classify emotions. capacity learn spatial features channels temporal variability with time. evaluated for valence space ratings. achieved accuracy 73.04%.

Язык: Английский

Процитировано

0