A Model to Predict Heartbeat Rate Using Deep Learning Algorithms DOI Open Access
Ahmed A. Alsheikhy, Yahia Said, Tawfeeq Shawly

et al.

Healthcare, Journal Year: 2023, Volume and Issue: 11(3), P. 330 - 330

Published: Jan. 22, 2023

ECG provides critical information in a waveform about the heart's condition. This is crucial to physicians as it first thing be performed by cardiologists. When COVID-19 spread globally and became pandemic, government of Saudi Arabia placed various restrictions guidelines protect save citizens residents. One these was preventing individuals from touching any surface public private places. In addition, authorities mandatory rule all facilities sector evaluate temperature before entering. Thus, idea this study stems need have touchless technique determine heartbeat rate. article proposes viable dependable method estimate an average rate based on reflected light skin. model uses deep learning tools, including AlexNet, Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTMs), ResNet50V2. Three scenarios been conducted validate presented model. proposed approach takes its inputs video streams converts into frames images. Numerous trials volunteers assess outputs terms accuracy, mean absolute error (MAE), squared (MSE). The achieves 99.78% MAE 0.142 when combing LSTMs ResNet50V2, while MSE 1.82. Moreover, comparative measurement between algorithm some studies literature utilized methods, MAE, are performed. achieved outcomes reveal that developed surpasses other methods. findings show can applied healthcare aid physicians.

Language: Английский

Real-Time Forecasting from Wearable-Monitored Heart Rate Data Through Autoregressive Models DOI Creative Commons

Giulio De Sabbata,

Giovanni Simonini

Journal of Healthcare Informatics Research, Journal Year: 2025, Volume and Issue: unknown

Published: March 7, 2025

Language: Английский

Citations

0
Ultrasensitive Magnetic Nanomechanical Biosensors for Simultaneous Detection of Multiple Cardiovascular Disease Biomarkers in a Single Blood Drop DOI
Qiubo Chen, Lin Zhang,

Zihan Qiao

et al.

Biosensors and Bioelectronics, Journal Year: 2025, Volume and Issue: unknown, P. 117448 - 117448

Published: April 1, 2025

Language: Английский

Citations

0
A Comparison of Machine Learning and Deep Learning Techniques for Predicting Alzheimer's Disease DOI

Harsh Kumar,

Mohammed Hasan Ali,

Aijaz Ahmad Chopan

et al.

IGI Global eBooks, Journal Year: 2025, Volume and Issue: unknown, P. 115 - 150

Published: March 28, 2025

Alzheimer's disease (AD) is a degenerative condition that can cause anything from slight loss of memory to total consciousness and speech. Early detection has critical role in maintaining the patient's quality life. Despite wealth studies on AD diagnosis, early correct diagnosis most beneficial patients. Because machine learning (ML) models may identify abnormalities on, they have become indispensable diseases such as AD. ML computer-aided diagnostics (CAD) been combined, this enhanced detection—especially when integrating with MRI data. methods are preferred because produce results quickly accurately. The goal research create an automated system more sophisticated accurate by data many modalities. strategy lower rate incorrect diagnoses while offering thorough diagnostic, emphasizing accuracy, sensitivity, specificity.

Language: Английский

Citations

0
Contact Free human heart rate prediction using LWHRPnet deep learning in real time face and wrist videos DOI

S. Anusha,

R. Manjith

Biomedical Signal Processing and Control, Journal Year: 2025, Volume and Issue: 108, P. 107930 - 107930

Published: April 21, 2025

Language: Английский

Citations

0
Need of Machine Learning to Predict Happiness: A Systematic Review DOI

Naveen Naveen,

Anupam Bhatia

Edumania-An International Multidisciplinary Journal, Journal Year: 2023, Volume and Issue: 01(02), P. 306 - 335

Published: July 20, 2023

Happiness is a current important subject of study in psychology and social science because it affects people's day-to-day lives, thoughts feelings, work habits, interactions with society family. There are number challenges Computer Science Machine Learning to predict happiness index using prediction techniques. This presents systematic review PRISMA style for prediction. During the Literature survey, was found that many predictive models whether statistical or designed but major emphasis on research remains focused factors listed World Report, i.e., real Gross Domestic Product per capita, support, healthy life expectancy, freedom make choices, generosity perceptions corruption. The factor influencing varies due personal differences, age group location variation. According Gallup Poll, general annual sample each country 1,000 people approximately 0.007% population participated measurement. purpose this discover describe new related like stress emotions, location-based group. It observed there requirement develop model which works psychological mental health, depression, stress, physical well-being, safety, leisure time available, suicidal ideation addition economic used Index by targeting large size populations.

Language: Английский

Citations

9
Hyperparameter optimization for cardiovascular disease data-driven prognostic system DOI Creative Commons
Jayson Saputra,

Cindy Lawrencya,

Jecky Mitra Saini

et al.

Visual Computing for Industry Biomedicine and Art, Journal Year: 2023, Volume and Issue: 6(1)

Published: Aug. 1, 2023

Abstract Prediction and diagnosis of cardiovascular diseases (CVDs) based, among other things, on medical examinations patient symptoms are the biggest challenges in medicine. About 17.9 million people die from CVDs annually, accounting for 31% all deaths worldwide. With a timely prognosis thorough consideration patient’s history lifestyle, it is possible to predict take preventive measures eliminate or control this life-threatening disease. In study, we used various datasets major hospital United States as prognostic factors CVD. The data was obtained by monitoring total 918 patients whose criteria adults were 28-77 years old. present mining modeling approach analyze performance, classification accuracy number clusters Cardiovascular Disease Prognostic unsupervised machine learning (ML) using Orange software. Various techniques then classify model parameters, such k-nearest neighbors, support vector machine, random forest, artificial neural network (ANN), naïve bayes, logistic regression, stochastic gradient descent (SGD), AdaBoost. To determine clusters, ML clustering methods used, k-means, hierarchical, density-based spatial applications with noise clustering. results showed that best performance analysis SGD ANN, both which had high score 0.900 datasets. Based most methods, k-means hierarchical clustering, can be divided into two clusters. CVD depends proposed determining diagnostic model. more accurate model, better at risk

Language: Английский

Citations

9
AI-Driven cardiac wellness: Predictive modeling for elderly heart health optimization DOI

Kamlesh Mani,

Kamlesh Kumar Singh, Ratnesh Litoriya

et al.

Multimedia Tools and Applications, Journal Year: 2024, Volume and Issue: 83(30), P. 74813 - 74830

Published: Feb. 13, 2024

Language: Английский

Citations

3
Machine Learning-Based Interpretable Modeling for Subjective Emotional Dynamics Sensing Using Facial EMG DOI Creative Commons

Naoya Kawamura,

Wataru Sato,

Koh Shimokawa

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(5), P. 1536 - 1536

Published: Feb. 27, 2024

Understanding the association between subjective emotional experiences and physiological signals is of practical theoretical significance. Previous psychophysiological studies have shown a linear relationship dynamic valence facial electromyography (EMG) activities. However, whether how dynamics relate to EMG changes nonlinearly remains unknown. To investigate this issue, we re-analyzed data two previous that measured ratings corrugator supercilii zygomatic major muscles from 50 participants who viewed film clips. We employed multilinear regression analyses nonlinear machine learning (ML) models: random forest long short-term memory. In cross-validation, these ML models outperformed in terms mean squared error correlation coefficient. Interpretation model using SHapley Additive exPlanation tool revealed interactive associations several features dynamics. These findings suggest can better fit than conventional highlight complex relationship. The encourage emotion sensing offer insight into subjective–physiological association.

Language: Английский

Citations

2
Effects of missing data imputation methods on univariate blood pressure time series data analysis and forecasting with ARIMA and LSTM DOI Creative Commons

Nicholas Niako,

Jesús D. Melgarejo, Gladys E. Maestre

et al.

BMC Medical Research Methodology, Journal Year: 2024, Volume and Issue: 24(1)

Published: Dec. 26, 2024

Missing observations within the univariate time series are common in real-life and cause analytical problems flow of analysis. Imputation missing values is an inevitable step every incomplete series. Most existing studies focus on comparing distributions imputed data. There a gap knowledge how different imputation methods for affect forecasting performance models. We evaluated prediction autoregressive integrated moving average (ARIMA) long short-term memory (LSTM) network models data using ten techniques. were generated under completely at random (MCAR) mechanism 10%, 15%, 25%, 35% rates missingness complete 24-h ambulatory diastolic blood pressure readings. The mean, Kalman filtering, linear, spline, Stineman interpolations, exponentially weighted (EWMA), simple (SMA), k-nearest neighborhood (KNN), last-observation-carried-forward (LOCF) techniques structure LSTM ARIMA compared original All either increased or decreased autocorrelation with this affected algorithms. best technique did not guarantee better predictions obtained mean imputation, LOCF, KNN, Stineman, cubic spline interpolations performed small rate missingness. Interpolation EWMA filtering yielded consistent performances across all scenarios Disregarding methods, resulted slightly predictive accuracy among performing models; otherwise, results varied. In our sample, tended to perform higher autocorrelation. recommend researchers that they consider smoothing techniques, interpolation (linear, Stineman), (SMA EWMA) imputing as well both distribution outperforms models, however, samples, simpler faster execute.

Language: Английский

Citations

2
Modeling Sleep Quality Depending on Objective Actigraphic Indicators Based on Machine Learning Methods DOI Open Access

Olga Vl. Bitkina,

Jaehyun Park, Jungyoon Kim

et al.

International Journal of Environmental Research and Public Health, Journal Year: 2022, Volume and Issue: 19(16), P. 9890 - 9890

Published: Aug. 11, 2022

According to data from the World Health Organization and medical research centers, frequency severity of various sleep disorders, including insomnia, are increasing steadily. This dynamic is associated with increased daily stress, anxiety, depressive disorders. Poor quality affects people’s productivity activity their perception life in general. Therefore, predicting classifying vital improving duration human life. study offers a model for assessing based on indications an actigraph, which was used by 22 participants experiment 24 h. Objective indicators actigraph include amount time spent bed, duration, number awakenings, awakenings. The resulting classification evaluated using several machine learning methods showed satisfactory accuracy approximately 80–86%. results this can be treat develop design new systems assess track quality, improve existing electronic devices sensors.

Language: Английский

Citations

11