IoT-Cloud-Based Smart Healthcare Monitoring System for Heart Disease Prediction via Deep Learning

Electronics, Journal Year: 2022, Volume and Issue: 11(15), P. 2292 - 2292

Published: July 22, 2022

The Internet of Things confers seamless connectivity between people and objects, its confluence with the Cloud improves our lives. Predictive analytics in medical domain can help turn a reactive healthcare strategy into proactive one, advanced artificial intelligence machine learning approaches permeating industry. As subfield ML, deep possesses transformative potential for accurately analysing vast data at exceptional speeds, eliciting intelligent insights, efficiently solving intricate issues. accurate timely prediction diseases is crucial ensuring preventive care alongside early intervention risk. With widespread adoption electronic clinical records, creating models enhanced accuracy key to harnessing recurrent neural network variants possessing ability manage sequential time-series data. proposed system acquires from IoT devices, stored on cloud pertaining patient history are subjected predictive analytics. smart monitoring predicting heart disease risk built around Bi-LSTM (bidirectional long short-term memory) showcases an 98.86%, precision 98.9%, sensitivity 98.8%, specificity 98.89%, F-measure which much better than existing systems.

Language: Английский

---

Advancements and Prospects of Machine Learning in Medical Diagnostics: Unveiling the Future of Diagnostic Precision

Archives of Computational Methods in Engineering, Journal Year: 2024, Volume and Issue: unknown

Published: June 26, 2024

Language: Английский

---

THE PREDICTION OF HEART DISEASE USING MACHINE LEARNING ALGORITHMS

Science Journal of University of Zakho, Journal Year: 2024, Volume and Issue: 12(3), P. 285 - 293

Published: July 14, 2024

Heart disease threatens the lives of around one individual per minute, establishing it as foremost cause mortality in contemporary era. A wide range individuals over globe has encountered intricacies associated with cardiovascular illness. Various factors, such hypertension, elevated levels cholesterol, and an irregular pulse rhythm hinder early identification a disease. In cardiology, similar to other branches Medicine, timely precise cardiac diseases is utmost importance. Anticipating onset heart failure at appropriate moment can provide challenges, particularly for cardiologists surgeons. Fortunately, categorisation forecasting models assist medical business real applications data. Regarding this, Machine Learning (ML) algorithms techniques have benefited from automated analysis several datasets complex data aid community diagnosing heart-related diseases. Predicting if patient early-stage primary goal this paper. prior study that worked on Erbil Disease dataset proved Naïve Bayes (NB) got accuracy 65%, which worst classifier, while Decision Tree (DT) obtained highest 98%. article, comparison been applied using same (i.e., dataset) between multiple ML algorithms, instance, LR (Logistic Regression), KNN (K-Nearest Neighbours), SVM (Support Vector Machine), DT (Decision Tree), MLP (Multi-Layer Perceptron), NB (Naïve Bayes) RF (Random Forest). Surprisingly, we 98% after applying LR, MLP, RF, was best outcome. Furthermore, by classifier differed incredibly received work.

Language: Английский

---

Smart Consumer Wearables as Digital Diagnostic Tools: A Review

Diagnostics, Journal Year: 2022, Volume and Issue: 12(9), P. 2110 - 2110

Published: Aug. 31, 2022

The increasing usage of smart wearable devices has made an impact not only on the lifestyle users, but also biological research and personalized healthcare services. These devices, which carry different types sensors, have emerged as digital diagnostic tools. Data from such enabled prediction detection various physiological well psychological conditions diseases. In this review, we focused applications wrist-worn wearables to detect multiple diseases cardiovascular diseases, neurological disorders, fatty liver metabolic including diabetes, sleep quality, illnesses. fruitful requires fast insightful data analysis, is feasible through machine learning. discussed machine-learning outcomes for analyses. Finally, current challenges with data, future perspectives tools domains.

Language: Английский

---

A Systematic Review of Machine Learning and IoT Applied to the Prediction and Monitoring of Cardiovascular Diseases

Healthcare, Journal Year: 2023, Volume and Issue: 11(16), P. 2240 - 2240

Published: Aug. 9, 2023

According to the Pan American Health Organization, cardiovascular disease is leading cause of death worldwide, claiming an estimated 17.9 million lives each year. This paper presents a systematic review highlight use IoT, IoMT, and machine learning detect, predict, or monitor disease. We had final sample 164 high-impact journal papers, focusing on two categories: detection using IoT/IoMT technologies techniques. For first category, we found 82 proposals, while for second, 85 proposals. The research highlights list technologies, techniques, datasets, most discussed diseases. Neural networks have been popularly used, achieving accuracy over 90%, followed by random forest, XGBoost, k-NN, SVM. Based results, conclude that can predict diseases in real time, ensemble techniques obtained one best performances metric, hypertension arrhythmia were Finally, identified lack public data as main obstacles approaches prediction.

Language: Английский

---

An Embedded System Using Convolutional Neural Network Model for Online and Real-Time ECG Signal Classification and Prediction

Diagnostics, Journal Year: 2022, Volume and Issue: 12(4), P. 795 - 795

Published: March 24, 2022

This paper presents an automatic ECG signal classification system that applied the Deep Learning (DL) model to classify four types of signals. In first part our work, we present development. Four different classes signals from PhysioNet open-source database were selected and used. preliminary study used a technique namely Convolutional Neural Network (CNN) predict classes: normal, sudden death, arrhythmia, supraventricular arrhythmia. The prediction process includes pulse extraction, image reshaping, training dataset, testing process. general, accuracy achieved up 95% after 100 epochs. However, each single type shows differentiation. Among classes, results show predictions for death waveforms are highest, i.e., 80 out samples correct (100% accuracy). contrast, lowest is normal sinus waveforms, 74 (92.5% due features being almost similar arrhythmia waveforms. has been tuned achieve optimal prediction. second part, presented hardware implementation with predictive embedded in NVIDIA Jetson Nanoprocessor online real-time

Language: Английский

---

Heart rate prediction with contactless active assisted living technology: a smart home approach for older adults

Frontiers in Artificial Intelligence, Journal Year: 2024, Volume and Issue: 6

Published: Jan. 12, 2024

Background As global demographics shift toward an aging population, monitoring their heart rate becomes essential, a key physiological metric for cardiovascular health. Traditional methods of are often invasive, while recent advancements in Active Assisted Living provide non-invasive alternatives. This study aims to evaluate novel prediction method that utilizes contactless smart home technology coupled with machine learning techniques older adults. Methods The was conducted residential environment equipped various sensors. We recruited 40 participants, each whom instructed perform 23 types predefined daily living activities across five phases. Concurrently, data were collected through Empatica E4 wristband as the benchmark. Analysis involved prominent models: Support Vector Regression, K-nearest neighbor, Random Forest, Decision Tree, and Multilayer Perceptron. Results All models achieved commendable performance, average Mean Absolute Error 7.329. Particularly, Forest model outperformed other models, achieving 6.023 Scatter Index value 9.72%. also showed robust capabilities capturing relationship between individuals' corresponding responses, highest R 2 0.782 observed during morning exercise activities. Environmental factors contribute most performance. Conclusions utilization proposed non-intrusive approach enabled innovative observe fluctuations different findings this research have significant implications public By predicting based on technologies activities, healthcare providers health agencies can gain comprehensive understanding individual's profile. valuable information inform implementation personalized interventions, preventive measures, lifestyle modifications mitigate risk diseases improve overall outcomes.

Language: Английский

---

A robust neural network for privacy-preserving heart rate estimation in remote healthcare systems

Healthcare Analytics, Journal Year: 2024, Volume and Issue: 5, P. 100329 - 100329

Published: April 4, 2024

In this study, we propose a computationally-light and robust neural network for estimating heart rate in remote healthcare systems. We develop model that can be trained on consumer-grade graphics processing units (GPUs), deployed edge devices swift inference. hybrid based convolutional (CNN) bidirectional long short-term memory (BiLSTM) architectures from Electrocardiogram (ECG) Photoplethysmography (PPG) signals. Considering the sensitive nature of ECG signals, ensure formal privacy guarantee, differential privacy, training. perform tight accounting overall budget our training algorithm using Rényi Differential Privacy technique. demonstrate outperforms state-of-the-art networks benchmark dataset both PPG signals despite having much smaller number trainable parameters and, consequently, inference times. Our CNN-BiLSTM architecture also provide excellent estimation performance even under strict constraints. prototype Arduino-based data collection system is low-cost, efficient, useful providing access to modern services people living areas.

Language: Английский

---

Search and Rescue–Based Sparse Auto‐Encoder for Detecting Heart Disease in IoT Healthcare Environment

Published: Feb. 20, 2025

The Internet of Things (IoT) facilitates effortless communication between humans and inanimate objects. With the widespread adoption cutting-edge learning techniques, predictive analytics in medical domain has latent ability to transform healthcare industry from a responsive practical one. Yet, cardiovascular disease is leading killer worldwide. Forecasting cardiac difficult since it takes both specialists heart relatively new application IoT technology systems. Although several studies have focused on diagnosing disease, outcomes been unreliable. In order better assess illness, suggested system employs an enhanced Sparse Auto-Encoder (ISAE) model. Additionally, classification accuracy through use Artificial Fish Swarm Optimisation (AFO) pick features dataset. smartwatch/heart monitor device worn by patient keeps track their BP ECG readings. ISAE employed categorise incoming sensor data as normal or abnormal, with SAE's hyper-parameter tuning optimally set SRO. associated algorithms scheme's efficiency. show that projected ISAE-based forecast scheme outperforms alternatives. method demonstrates preexisting classifiers reaching 98% largest possible records.

Language: Английский

---

Determining the Optimal Recovery Time for Fatigued Construction Workers: Machine Learning Approach Based on Physiological and Environmental Measurements

Building and Environment, Journal Year: 2025, Volume and Issue: unknown, P. 112808 - 112808

Published: Feb. 1, 2025

Language: Английский

---