A Review on Mental Stress Detection Using Wearable Sensors and Machine Learning Techniques

IEEE Access, Journal Year: 2021, Volume and Issue: 9, P. 84045 - 84066

Published: Jan. 1, 2021

Stress is an escalated psycho-physiological state of the human body emerging in response to a challenging event or demanding condition. Environmental factors that trigger stress are called stressors. In case prolonged exposure multiple stressors impacting simultaneously, person's mental and physical health can be adversely affected which further lead chronic issues. To prevent stress-related issues, it necessary detect them nascent stages possible only by continuous monitoring stress. Wearable devices promise real-time data collection, helps personal monitoring. this paper, comprehensive review has been presented, focuses on detection using wearable sensors applied machine learning techniques. This paper investigates approaches adopted accordance with sensory such as sensors, Electrocardiogram (ECG), Electroencephalography (EEG), Photoplethysmography (PPG), also depending various environments like during driving, studying, working. The stressors, techniques, results, advantages, limitations, issues for each study highlighted expected provide path future research studies. Also, multimodal system sensor-based deep technique proposed at end.

Language: Английский

---

A Review on Mental Stress Assessment Methods Using EEG Signals

Sensors, Journal Year: 2021, Volume and Issue: 21(15), P. 5043 - 5043

Published: July 26, 2021

Mental stress is one of the serious factors that lead to many health problems. Scientists and physicians have developed various tools assess level mental in its early stages. Several neuroimaging been proposed literature workplace. Electroencephalogram (EEG) signal important candidate because it contains rich information about states condition. In this paper, we review existing EEG analysis methods on assessment stress. The highlights critical differences between research findings argues variations data contribute several contradictory results. results could be due including lack standardized protocol, brain region interest, stressor type, experiment duration, proper processing, feature extraction mechanism, type classifier. Therefore, significant part related recognition choosing most appropriate features. particular, a complex diverse range features, time-varying, functional, dynamic connections, requires integration understand their associations with Accordingly, suggests fusing cortical activations connectivity network measures deep learning approaches improve accuracy assessment.

Language: Английский

---

Tomato Leaf Disease Classification via Compact Convolutional Neural Networks with Transfer Learning and Feature Selection

Horticulturae, Journal Year: 2023, Volume and Issue: 9(2), P. 149 - 149

Published: Jan. 22, 2023

Tomatoes are one of the world’s greatest valuable vegetables and regarded as economic pillar numerous countries. Nevertheless, these harvests remain susceptible to a variety illnesses which can reduce destroy generation healthy crops, making early precise identification diseases critical. Therefore, in recent years, studies have utilized deep learning (DL) models for automatic tomato leaf illness identification. However, many methods based on single DL architecture that needs high computational ability update hyperparameters leading rise classification complexity. In addition, they extracted large dimensions from networks added complication. this study proposes pipeline utilizing three compact convolutional neural (CNNs). It employs transfer retrieve features out final fully connected layer CNNs more condensed high-level representation. Next, it merges benefit every CNN structure. Subsequently, applies hybrid feature selection approach select generate comprehensive set lower dimensions. Six classifiers procedure. The results indicate K-nearest neighbor support vector machine attained highest accuracy 99.92% 99.90% using 22 24 only. experimental proposed also compared with previous research verified its competing capacity.

Language: Английский

---

CerCan·Net: Cervical cancer classification model via multi-layer feature ensembles of lightweight CNNs and transfer learning

Expert Systems with Applications, Journal Year: 2023, Volume and Issue: 229, P. 120624 - 120624

Published: June 2, 2023

Language: Английский

---

A Framework for Lung and Colon Cancer Diagnosis via Lightweight Deep Learning Models and Transformation Methods

Diagnostics, Journal Year: 2022, Volume and Issue: 12(12), P. 2926 - 2926

Published: Nov. 23, 2022

Among the leading causes of mortality and morbidity in people are lung colon cancers. They may develop concurrently organs negatively impact human life. If cancer is not diagnosed its early stages, there a great likelihood that it will spread to two organs. The histopathological detection such malignancies one most crucial components effective treatment. Although process lengthy complex, deep learning (DL) techniques have made feasible complete more quickly accurately, enabling researchers study lot patients short time period for less cost. Earlier studies relied on DL models require computational ability resources. Most them depended individual extract features high dimension or perform diagnoses. However, this study, framework based multiple lightweight proposed utilizes several transformation methods feature reduction provide better representation data. In context, histopathology scans fed into ShuffleNet, MobileNet, SqueezeNet models. number acquired from these subsequently reduced using principal component analysis (PCA) fast Walsh-Hadamard transform (FHWT) techniques. Following that, discrete wavelet (DWT) used fuse FWHT's obtained three Additionally, models' PCA concatenated. Finally, diminished as result FHWT-DWT fusion processes four distinct machine algorithms, reaching highest accuracy 99.6%. results show can distinguish variants with lower complexity compared existing methods. also prove utilizing reduce offer superior interpretation data, thus improving diagnosis procedure.

Language: Английский

---

An Intelligent ECG-Based Tool for Diagnosing COVID-19 via Ensemble Deep Learning Techniques

Biosensors, Journal Year: 2022, Volume and Issue: 12(5), P. 299 - 299

Published: May 5, 2022

Diagnosing COVID-19 accurately and rapidly is vital to control its quick spread, lessen lockdown restrictions, decrease the workload on healthcare structures. The present tools detect experience numerous shortcomings. Therefore, novel diagnostic are be examined enhance accuracy avoid limitations of these tools. Earlier studies indicated multiple structures cardiovascular alterations in cases which motivated realization using ECG data as a tool for diagnosing coronavirus. This study introduced automated based diagnose COVID-19. utilizes ten deep learning (DL) models various architectures. It obtains significant features from last fully connected layer each DL model then combines them. Afterward, presents hybrid feature selection chi-square test sequential search select features. Finally, it employs several machine classifiers perform two classification levels. A binary level differentiate between normal cases, multiclass discriminate other cardiac complications. proposed reached an 98.2% 91.6% levels, respectively. performance indicates that could used alternative means diagnosis

Language: Английский

---

Validation of a Light EEG-Based Measure for Real-Time Stress Monitoring during Realistic Driving

Brain Sciences, Journal Year: 2022, Volume and Issue: 12(3), P. 304 - 304

Published: Feb. 24, 2022

Driver's stress affects decision-making and the probability of risk occurrence, it is therefore a key factor in road safety. This suggests need for continuous monitoring. work aims at validating neurophysiological measure-a Neurometric-for out-of-the-lab use obtained from lightweight EEG relying on two wet sensors, real-time, without calibration. The Neurometric was tested during multitasking experiment validated with realistic driving simulator. Twenty subjects participated experiment, resulting compared Random Forest (RF) model, calibrated by using features both intra-subject cross-task approaches. also measure based skin conductance level (SCL), representing one physiological parameters investigated literature mostly correlated variations. We found that experiments, able to discriminate between low high levels an average Area Under Curve (AUC) value higher than 0.9. Furthermore, showed AUC stability SCL RF approach. In conclusion, proposed this proved be suitable monitoring levels.

Language: Английский

---

Applications of artificial intelligence−machine learning for detection of stress: a critical overview

Molecular Psychiatry, Journal Year: 2023, Volume and Issue: 29(6), P. 1882 - 1894

Published: April 5, 2023

Language: Английский

---

GabROP: Gabor Wavelets-Based CAD for Retinopathy of Prematurity Diagnosis via Convolutional Neural Networks

Diagnostics, Journal Year: 2023, Volume and Issue: 13(2), P. 171 - 171

Published: Jan. 4, 2023

One of the most serious and dangerous ocular problems in premature infants is retinopathy prematurity (ROP), a proliferative vascular disease. Ophthalmologists can use automatic computer-assisted diagnostic (CAD) tools to help them make safe, accurate, low-cost diagnosis ROP. All previous CAD for ROP original fundus images. Unfortunately, learning discriminative representation from ROP-related images difficult. Textural analysis techniques, such as Gabor wavelets (GW), demonstrate significant texture information that artificial intelligence (AI) based models improve accuracy. In this paper, an effective automated tool, namely GabROP, on GW multiple deep (DL) proposed. Initially, GabROP analyzes using generates several sets Next, these are used train three convolutional neural networks (CNNs) independently. Additionally, actual pictures build networks. Using discrete wavelet transform (DWT), features retrieved every CNN trained with various combined create textural-spectral-temporal demonstration. Afterward, each CNN, concatenated spatial obtained Finally, all incorporated cosine (DCT) lessen size caused by fusion process. The outcomes show it accurate efficient ophthalmologists. effectiveness compared recently developed techniques. Due GabROP's superior performance competing tools, ophthalmologists may be able identify more reliably precisely, which could result reduction effort examination time.

Language: Английский

---

FUSI-CAD: Coronavirus (COVID-19) diagnosis based on the fusion of CNNs and handcrafted features

PeerJ Computer Science, Journal Year: 2020, Volume and Issue: 6, P. e306 - e306

Published: Oct. 12, 2020

The precise and rapid diagnosis of coronavirus (COVID-19) at the very primary stage helps doctors to manage patients in high workload conditions. In addition, it prevents spread this pandemic virus. Computer-aided (CAD) based on artificial intelligence (AI) techniques can be used distinguish between COVID-19 non-COVID-19 from computed tomography (CT) imaging. Furthermore, CAD systems are capable delivering an accurate faster diagnosis, which consequently saves time for disease control provides efficient compared laboratory tests. study, a novel system called FUSI-CAD AI is proposed. Almost all methods literature individual convolutional neural networks (CNN). Consequently, fusion multiple different CNN architectures with three handcrafted features including statistical textural analysis such as discrete wavelet transform (DWT), grey level co-occurrence matrix (GLCM) were not previously utilized diagnosis. SARS-CoV-2 CT-scan dataset test performance proposed FUSI-CAD. results show that could accurately differentiate images, accuracy achieved 99%. Additionally, proved reliable well. This because sensitivity, specificity, precision attained diagnostics odds ratio (DOR) ≥ 100. recent related studies same dataset. comparison verifies competence over other systems. Thus, employed real diagnostic scenarios achieving testing avoiding human misdiagnosis might exist due fatigue. It also reduce exertion made by radiologists during examination process.

Language: Английский

---