Detection of Heart-Diseases Through Cloud-Enhanced ECG

Monitoring

Senthamarai N, Anoushka Sanjeev, Yashvardhan and Sai Prajwal Kacham

Department of Networking and Communication, College of Engineering and Technology,

SRM Institute of Science and Technology, Kattankulathur, Tamilnadu - 603203, India

Keywords: Data Insights, ECG Monitoring, Data Processing, Data Integration, Abnormality Detection, Deployment,

Cloud Computing, Edge Computing, Enhanced Cloud Monitoring, Internet of Things, Cloud Data, Edge

Devices, Heart Rate Data.

Abstract: Cardiovascular diseases (CVDs) is a major concern globally, with traditional ECG monitoring systems often

lacking the capability for real-time analysis and efficient data integration. This research aims to enhance ECG

monitoring by leveraging cloud computing for real-time analytics, improving the detection and management

of heart diseases. The study includes ECG data collection using advanced sensors, real-time analysis through

machine learning algorithms for immediate abnormality detection, and the development of a cloud-based

infrastructure for scalable storage, processing, and remote access of ECG data.

1 INTRODUCTION

With the rise of cardiovascular diseases (CVDs) as a

global health concern, traditional ECG monitoring

systems often fall short in delivering timely and

accurate diagnoses due to limitations in real-time data

analysis and integration. To address these challenges,

integrating cloud computing with ECG monitoring

enables real-time data processing, thereby enhancing

the detection and management of heart diseases.

Serverless computing offers a dynamic and cost-

effective approach, allowing seamless scaling and

efficient handling of ECG data streams without the

complexities of traditional infrastructure

management. This innovation significantly improves

patient outcomes by facilitating rapid, data-driven

decisions.

By harnessing the power of cloud and edge

environments, serverless data processing ensures that

ECG data is analyzed immediately as it is generated.

This reduces latency, enhances accuracy, and enables

real-time insights into heart conditions. The

flexibility of serverless platforms, such as AWS

Lambda, allows developers to focus on application

logic while the cloud platform manages resource

allocation and scaling. The unified approach

simplifies the development and deployment of ECG

monitoring applications, ensuringithat critical data is

accessible, actionable, and integrated into broader

healthcare analytics platforms. This not only

addresses the challenges of traditional ECG systems

and opens up for innovation in healthcare,

particularly in the timely detection and treatment of

cardiovascular diseases.

Serverless computing offers several advantages

over traditional models. By eliminating the need for

manual infrastructure management, serverless

platforms like AWS Lambda automatically scale

resources in response to the volume of incoming data.

This means that as ECG data is generated from

various sensors, it can be analyzed instantaneously,

ensuring that abnormalities are detected without

delay. The flexibility of serverless architectures also

allows for seamless integration with edge computing

devices, enabling a unified approach to data

processing that connects the gap between the cloud

and edge environments.

This real-time dataiprocessing capabilities are

especially crucial for healthcare, where theiability to

quickly analyze and act on data can significantly

impact patient outcomes. By deploying machine

learning algorithms within the serverless framework,

healthcare providers can automate the detection of

irregular heart patterns, facilitating faster diagnosis

and treatment. The cloud-based infrastructure further

ensures that data is securely stored, processed, and

accessed remotely, providing scalability and

N, S., Sanjeev, A., Yashvardhan, and Kacham, S. P.

Detection of Heart-Diseases Through Cloud-Enhanced ECG Monitoring.

DOI: 10.5220/0013586600004664

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 3rd International Conference on Futuristic Technology (INCOFT 2025) - Volume 2, pages 57-64

ISBN: 978-989-758-763-4

efficiency that are critical in modern healthcare

settings.

Figure 1: Data Gathering and Pre-Processing

Serverless data processing refers to the use of

cloud computing services where the management of

servers and infrastructure is completely handled by

the cloud provider, which allows developersito focus

on writing code and logic. In this model, you pay only

for the compute time you consume, without needing

to worry about provisioning, scaling, or managing

servers. This approach is highly scalable, cost-

effective.

2 RELATED WORKS

B. Ramesh and Kuruva Lakshmanna (Ramesh, and

Lakshmanna, 2024) presents an advanced deep

learning approach to predict and prevent coronary

heart disease (CHD) iniindividuals with Type 2

Diabetes Mellitus (T2DM). The study introduces a

hybrid model, O-SBGC-LSTM, which combines

Enhanced Optimization Algorithm (EOA) with

Stochastic Gradient Boosting Classifier LongiShort-

Term Memory (SBGC-LSTM) for accurate CHD

prediction. The proposed system integrates multiple

layers for data preprocessing, feature extraction, and

classification, achieving high accuracy with a Kaggle

dataset. In addition to predicting CHD risk, the

framework employs fuzzy inference for disease

prevention, focusing on lifestyle management and

treatment optimization, using minimal data from IoT

trackers to help diabetic patients manage and control

CHD complications effectively.

Md. Razu Ahmed et al. (Ahmed, Mahmud, et al 2018)

proposes a cloud-based architecture designed to

enhance the early detection of chronic heart disease

(CHD) by leveraging machine learning techniques.

CHD, characterized by plaque buildup in the

coronary arteries, is the leading cause of death

worldwide, particularly in low-income countries. The

authors aim to reduce the complexity and cost of heart

disease diagnosis by applying machine learning

algorithms, such as Artificial Neural

NetworksI(ANN), Support Vector Machines (SVM),

Decision Treesi(DT), Random Forest (RF), and

Naïve Bayes (NB), within a cloud-based system.

They evaluate the performance of these algorithms

using metrics like confusion matrices and ROC

curves, offering a solution that optimizes accuracy

while minimizing clinical errors and costs.

Neha Gaigawali (Gaigawali and Chaskar, 2018)

emphasizes the global burden of cardiovascular

diseases (CVDs), particularly in India, where CVDs

account for 18.8% of deaths and are projected to

become the leading cause of disability and death by

2020. The paper attributes this high mortality rate to

lifestyle and dietary habits, stressing that better access

to healthcare could improve outcomes. With the rapid

growth of cloud computing, specifically mobile cloud

computing (MCC), the paper highlights the potential

for transforming healthcare by providing affordable,

real-time medical services via smartphones. It also

addresses the limitations of traditional ECG

monitoring systems in handling large physiological

data, proposing an advanced cloud-based system for

monitoringiECG and detecting atrial fibrillation, a

leading cause of strokes.

Teofil Ilie Ursache et al. (Ursache, Pogoreanu et

al 2022) focuses on the significance of continuous

cardiovascular monitoring, particularly for patients

with cardiovascular diseases, myocardial infarction,

or other heart conditions. It emphasizes the need for

long-term observation ofiphysiological parameters to

guide treatments and monitor the success of

interventions. Traditional monitoring methods were

cumbersome and required hospitalization, often

missing critical episodes. However, advancements in

telemedicine and portable heart rate monitoring

devices now allow for remote, real-time tracking of

patient health. This enables timely medical

interventions and reduces healthcare costs. The study

highlights the use of wireless devices for continuous

monitoring, which can automatically alert healthcare

providers when abnormal readings occur, offering

significant benefits for both patient care and resource

management.

Pedro Sa et al. (Pedro Sa et al. 2019) presents a

novel architecture for real-time electrocardiogram

(ECG) processing in off-the-person setups,

specifically using a single-lead configuration from

the wrists (Lead I). This system is designed for

integration with wearable devices and daily-use

objects, such as steering wheels, to simplify heart

INCOFT 2025 - International Conference on Futuristic Technology

condition monitoring in non-intrusive ways. It offers

a scalable, programmable architecture that allows

easy tuning for different acquisition setups, filtering,

and classification parameters. The system provides

end-to-end ECG analysis, from rawisignal

preprocessing to classification, achieving an accuracy

of 96.5% in a four-class setup when tested on a Zynq-

7 ZC702 board.

3 PROPOSED METHODOLOGY

Figure 2: System Architecture describing the workflow

3.1 ECG Data Collection and Sensor

Integration

Our proposed system focuses on continuous ECG data

collection using advanced biosensors worn by

patients. These sensors are capable of collecting high-

precision ECG signals, even in real-time, and

transmitting them wirelessly to a cloud-based

infrastructure. The integration of sensors ensures a

consistent and reliable data stream for continuous

monitoring.

Sensor Technology: High-fidelity biosensors

capture ECG signals, which are then transmitted

using secure, low-latency communication

protocolsilike MQTT (Message Queuing Telemetry

Transport) oriCoAP (Constrained Application

Protocol). Edge Device Integration: Edge computing

devices are deployed to act as intermediaries between

the sensors and the cloud infrastructure. Preliminary

data filtering, signal preprocessing, and noise

reduction techniques are executed at the edge level to

make sure that the transmitted data to the cloud is

ready for analysis.

Figure 3: Data Pre-Processing

3.2 Real-Time Data Processing Using

Serverless Architecture

For scalable and real-time processing of ECG data, a

serverless computing architecture is implemented.

Platforms such as AWS Lambda,iGoogle Cloud

Functions, or Microsoft Azure Functions are

employed for eliminating the complexities of manual

infrastructure management. This architecture allows

automatic scaling based on demand while

maintaining cost-effectiveness.

Serverless Real-Time Analytics: ECG data is

analyzed in real time using serverless functions that

automatically scale based on incoming data loads.

These functions are event-driven, being triggered

upon the arrival of new ECG data streams. This

method ensures that the computational resources are

efficiently utilized without over-provisioning.

Machine Learning Algorithms: Machine learning

models deployed in the serverless environment

handle real-time ECG analysis. These models are

trained on large datasets to detect various cardiac

conditions such as arrhythmias and ischemia. Key

ECG features like heart rate variability (HRV), P-

wave, QRS complex, and T-wave are extracted for

anomaly detection.

Anomaly Detection Algorithms: Time-series

models such as LongiShort-Term Memory (LSTM)

networks and autoencoders are utilized to detect

irregularities in heart rhythms. These models are

persistently upgraded utilizing unused information to

make strides location accuracy.

Edge-Cloud Synergy: Edge computing is

incorporated for low-latency preliminary analysis,

where immediate detection of critical heart conditions

is required. The hybrid edge-cloud approach ensures

that the bulk of intensive analytics is done in the cloud

while time-sensitive tasks are handled locally at the

edge.

Detection of Heart-Diseases Through Cloud-Enhanced ECG Monitoring

3.3 Scalable Cloud-Based Storage and

Data Management

For handling the high volumes of data generated

by continuous ECG monitoring, a scalable cloud

storage system is employed. This system leverages

distributed storage solutions such as AWS S3 to

securely manage ECG data at scale.

Data Partitioning: To optimize retrieval speed

and improve query efficiency, data is partitioned

using strategies like horizontal partitioning and time-

based partitioning. This permits quicker access to

patient-specific data.

Data Redundancy: Redundant storage

mechanisms are applied to ensure high availability

and fault tolerance. Multi-region replication is used to

safeguard against data loss and ensure continuous

access.

Security and Compliance: Data encryption

techniques such as AESi(Advanced Encryption

Standard) and TLS (Transport Layer Security) are

applied to secure patient data both in transit and at

rest. Also, compliance with healthcare controls like

HIPAA (Health InsuranceiPortability and

Accountability Act) and GDPR (General Data

Protection Regulation) is implemented to keep up

patient confidentiality.

3.4 Real-Time Visualization and

Healthcare System Integration

The processed ECG data is integrated with

healthcare systems, offering healthcare providers

immediate access to patient data for monitoring,

diagnosis, and treatment adjustments.

Real-Time Dashboards: Clinicians can monitor

patient health through real-time dashboards that

display vital heart metrics such as heart rate, heart

rhythm, and abnormal patterns. These dashboards

generate alerts when critical conditions such as

arrhythmias are detected.

API Integration with EHR Systems: Secure

APIs, such as FHIR (Fast Healthcare Interoperability

Resources), are used to integrate ECG data with

existing electronic health record (EHR) systems. This

ensures thatihealthcare providers have easy access to

old and real-time ECG data.

3.5 Machine Learning Model Optimization

and Continuous Learning

The system employs machine learning models for

continuous analysis of ECG data. To ensure high

accuracy in detecting cardiovascular abnormalities,

models are continuously retrained using newly

acquired data. This process ensures that the models

remain effective over time and adapt to new types of

cardiac abnormalities.

Model Training and Optimization: Cloud

computing platforms such as AWS Sagemaker and

Google AI are used for training deep learning models

on ECG data. Hyperparameter tuning techniques are

employed to maximize the accuracy and performance

of the models.

Continuous Learning: The models are

periodically updated and retrained using new patient

data to ensure that they adapt to emerging cardiac

conditions and improve in detecting a wider range of

anomalies.

3.6 Hybrid Edge-Cloud Architecture

for Latency Reduction

A hybrid edge-cloud architecture is utilized to

upgrade the real-time capabilities of the ECG

monitoring system. This framework ensures that

critical ECG data is processed immediately at the

edge, while more complex analyses and data storage

are handled by the cloud.

Edge Computing for Latency Reduction:

Immediate preliminary analysis, such as heart rate

detection and noise reduction, is performed on the

edge to provide fast feedback in cases of potential

cardiac emergencies. This reduces latency and

ensures that healthcare providers receive real-time

alerts.

Edge-Cloud Collaboration: The system

dynamically distributes tasks between the edge and

the cloud, ensuring efficient resource utilization.

Edge devices handle short-term data storage and

immediate analysis, while long-term data storage and

deeper analytics are managed by the cloud.

INCOFT 2025 - International Conference on Futuristic Technology

Figure 4: Graphs Plotted for data processing

3.7 Serverless Computing for Cost

Optimization and Scalability

The serverless architecture used in the system ensures

a cost-effective solution by only charging for the

actual computation used. The system leverages an

event-driven approach, where functions are triggered

only when ECG data is received, minimizing idle

resource consumption.

Event-Driven Architecture: The serverless

functions are event-triggered, which means they are

activated only when new ECG data is uploaded. This

model optimizes resource usage and reduces costs

associated with idle server time.

Figure 5: Serverless Architetcure of the amazon cluster for

cloud implementation

4 EXPERIMENT RESULT

The experiment for detecting heart diseases using

cloud-enhanced ECG monitoring incorporates

advanced cloud-based techniques and imachine

ilearning algorithms, primarily Convolutional Neural

Networks (CNN) and iSupport Vector Machines

(SVM). The focus is on developing an efficient

system for ireal-time analysis of ECG data, which is

crucial for early diagnosis and management of

cardiovascular diseases (CVDs). CVDs are a ileading

cause of mortality worldwide, and traditional

methods of monitoring often struggle with issues like

slow data processing, limited data integration, and

poor real-time analysis, which hinder timely medical

interventions. By leveraging cloud computing, the

proposed system aims to enhance the detection of

abnormalities and improve overall health outcomes

through faster processing and robust analytics.

4.1 Training Process

The training phase of this system relies on

sophisticated architectures, including CNN and SVM

models, to process ECG signals collected from

patients. The goal is to identify patterns in these

signals that could indicate the presence of heart

disease. A model like CNN can learn intricate

features from ECG data, making it highly suitable for

tasks such as arrhythmia detection. CNN models,

with 16 layers in this case, are trained on a dataset of

ECG recordings, using a total of 250 epochs, 64

batches, and a learning rate of 0.01 to ofine-tune the

performance. Information expansion methods are

connected to increment the dataset’s differences,

which makes a difference the show generalize

superior. This step is crucial in improving ithe

accuracy and irobustness of the model as it exposes

the system to a wide range of ECG signal variations,

including those from healthy individuals and patients

with heart disease.

The learning process is continuously monitored,

with loss and accuracy metrics visualized at various

stages, much like the standard approach used in

neural network training. After extensive training, the

model's accuracy approaches approximately 93%,

which is a significant improvement over traditional

ECG monitoring systems.

Detection of Heart-Diseases Through Cloud-Enhanced ECG Monitoring

Figure 6: Training Data

4.2 Testing Process

After training the models, the next step is to test their

performance on a new set of ECG data that was not

included in the training phase. This new data is

critical for evaluating the model's generalization

ability—whether it can accurately classify unseen

signals. During testing, the CNN model, CNN with

data augmentation, and the SVM model are used.

Each of these models offers a unique perspective on

classification: CNN provides in-depth feature

extraction, while SVM is a powerful classifier.

In this phase, loss values for both CNN and SVM

architectures are calculated. It is observed that, after

the 250th epoch, both models show minimal loss and

achieve an accuracy of 91%. However, fluctuations

were noticed in the CNN model’s performance during

initial epochs, but this stabilized over time, showing

the system's ability to adapt and improve. This

minimal loss and high accuracy indicate that the

models are not overfitting the training data, but rather

learning generalizable features that apply to new

cases as well.

Figure 7: Testing Data

4.3 Performance Evaluation and

Validation

A comprehensive evaluation of the system’s

performance iis carried out by calculating key metrics

such as True Positive (TP), iTrue Negative (TN),

False Positive (FP), and False Negative (FN) values.

These values help compute several evaluation metrics

like accuracy, precision, recall, and F1 score, all of

which are critical for determining how effective the

model is at correctly identifying the presence or

absence of heart disease.

For instance, precision measures how many of the

predicted positive cases are actual positive cases (i.e.,

the system accurately detected heart disease), while

recall measures the system’s ability to identify

positive cases from the dataset. In this study, both

CNN and SVM models achieve high precision and

recall values, indicating that the system is highly

reliable in distinguishing between healthy and

diseased states in the ECG data.

However, despite the high accuracy and precision

scores, it is essential to acknowledge that some details

of the study are lacking. For example, specifics

regarding the data sources and simulation

environments were not provided, which could affect

the study’s reproducibility. A comprehensive

understanding of these aspects is crucial for future

researchers to replicate and validate the results.

Figure 8: Performance and Validation

4.4 Key Contributions

The study contributes to existing literature by

demonstrating how cloud-based systems can be

leveraged for real-time ECG monitoring, enabling

quicker responses and better management of

cardiovascular diseases. The use of data

augmentation in particular stands out as an innovative

INCOFT 2025 - International Conference on Futuristic Technology

way to enhance the training dataset, leading to

improved model accuracy. Furthermore, combining

CNN and SVM provides a balanced approach

between feature extraction and classification,

allowing the system to perform well in diverse

scenarios.

Additionally, the application of cloud-based

infrastructure ensures that the system is scalable,

making it feasible for ilarge-scale deployments in

healthcare settings. As healthcare data grows

increasingly complex, systems like the one proposed

in this study can help manage the data overload by

offering efficient, on-demand processing capabilities.

The cloud's inherent flexibility also allows for easy

updates and integration with other medical systems,

making this approach highly adaptable.

5 CONCLUSION AND FUTURE

WORK

The proposed cloud-based ECG monitoring system

represents a significant advancement in the

management of cardiovascular diseases by

integrating modern cloud and serverless computing

technologies. Traditional ECG systems often struggle

with real-time data analysis due to complex

infrastructure needs and latency issues.

Our system addresses these challenges by

leveraging serverless computing to offer a scalable,

cost-effective solution for continuous heart health

monitoring.

The system employs high-fidelity biosensors for

real-time ECG data collection, which are seamlessly

integrated with edge devices. These edge devices

perform preliminary tasks such as data filtering and

noise reduction before transmitting refined data to the

cloud. This setup minimizes latency and enhances

data accuracy, which is crucial for timely detection of

cardiac anomalies.

Serverlessicomputing platforms, including AWS

Lambda, Google Cloud iFunctions, and Microsoft

Azure Functions, provide the backbone for real-time

data processing.

These platforms dynamically allocate

computational resources based on data load,

optimizing both cost and performance. The event-

driven model triggers functions only when new ECG

data is received, reducing idle resource consumption

and mitigating cold start delays with provisioned

concurrency techniques.

The integration of machine learning algorithms,

such as Long Short-Term Memory (LSTM)

networksiand autoencoders, enables sophisticated

analysis of ECG data, enhancing the detection of

various cardiac conditions. Continuous retraining of

these models ensures their accuracy and adaptability

to emerging cardiac conditions.

Our system also incorporates robust cloud-based

storage solutions, iincluding AWS S3, Google Cloud

Storage, and Azure Blob Storage, to manage the large

volumes of ECG data efficiently. Data partitioning

and redundancy mechanisms improve retrieval

speeds and ensure high availability, while encryption

and compliance with regulations like HIPAA and

GDPR safeguard patient privacy.

Real-time visualization is achieved through

interactive dashboards that display critical heart

metrics and generate alerts for detected

abnormalities. Integration with electronic health

record (EHR) systems via secure APIs ensures

healthcare providers have immediate access to both

historical and real-time ECG data, facilitating timely

and informed decision-making.

The hybrid edge-cloud architecture optimizes

performance by performing preliminary analysis at

the edge, reducing latency, while handling complex

analytics and long-term data storage in the cloud. This

synergy between edge and cloud computing enhances

the system's efficiency and scalability.

Overall, the proposed system demonstrates

significant improvements in ECG monitoring

accuracy. By addressing the limitations of traditional

ECG systems and incorporating real-time data

analysis, it enhances the detection and management

of cardiovascular diseases. This innovative approach

has the potential to transform cardiac health

monitoring, offering timely, data-driven insights that

can lead to better patient outcomes and more effective

management of heart conditions.

REFERENCES

Nastic, S., Rausch, T., Scekic, & Prodan, R.i(2017). A

serverless real-time data analytics platform for edge

computing. IEEE Internet Computing, 21(4), 64-71.

Pendke, V., Khadatkar, G., Gawate & Keote, M. (2024).

ECG VisionSense: A cloud-deployed, Flask-based

intelligent cardiovascular disease diagnosis

system.2024(InCACCT), 1-6.

Ramesh, B., & Lakshmanna, K. (2024). A novel early

detection and prevention of coronary heart disease

framework using hybrid deep learning model and

neural fuzzy inference system. IEEE Access, 12,

3366537.

Lim, B. D., Lim, W. H., Chong, K. S., Mokayef, M., Tiang,

S. S., & Tan, B. S. (2023). Development of a iwireless

Detection of Heart-Diseases Through Cloud-Enhanced ECG Monitoring

real-time computation system for fast and accurate

heart rate monitoring using photoplethysmography

(PPG) signals. 2023 (MICC), 1-6.

Sá, P., Aidos, H., Roma, N., & Tomás, P. (2019). Heart

disease detection architecture for Lead I ioff-the-person

ECG monitoring devices. 2019 27th European Signal

Processing Conference (EUSIPCO), 1-5.

Ahmed, M. R., Mahmud, S. M. H., Hossin, M. A., Jahan,

H., & Noori, S. R. H. (2018).iA cloud based four-tier

architecture for early detection of heart idisease with

machine learning algorithms. 2018 (ICCC), 1-5.

Ursache, T. I., Pogoreanu, R., Bozomitu, R. G., & Rotariu,

C. (2022). Web-based medical systemiifor remote heart

rate monitoring with cloud integration. 2022 10th IEEE

(EHB), 1-4.

Gaigawali, N., & Chaskar, U.iM. (2018). Cloud based ECG

monitoring and fibrillation detection forihealthcare

system. Proceedings of the Second International

Conference on (ICICCS 2018), 1-5.

Varsha, S., Rajesh, A., Sabarisha, K., & Priya, B. (2023).

Detection of heart disease usingisensor-enabled device

and machine learning.(ICCEBS), 1-6.

Kaiyu Zhang, Lixin Song and Di Lu,i"Design of remote

ECG monitoring system based on GPRS.

S. Satheeskumaran, K. Sasikala, K. Neeraj, A.

SenthilKumar and N. S. Babu, "IoT based ECG Signal

Feature Extraction and Analysis for Heart Disease Risk

Assessment,"

S. Krishnakumar., M. Yasodha., J. Veronica

Priyadharshini., J. Bethanney Janney., S. Divakaran

and V. Lumen Christy., "Detection of Arrhythmia and

Congestive Heart Failure Through Classification of

ECG Signals Using Deep Learning Neural Network,"

T. M. Rahman, S. Siddiqua, S. E. Rabby, N. Hasan and M.

H. Imam, "Early Detection of Kidney Disease Using

ECG Signals Through Machine Learning Based

Modelling," 2019

M. A. Rahman, M. J. A. Samin and Z. A. Hasan, "Remote

ECG Monitoring and Syncope Detection System Using

Deep Learning,"

K. C. Wee and M. S. M. Zahid, "Cloud Computing for ECG

Analysis Using MapReduce,"

S. Satheeskumaran, K. Sasikala, K. Neeraj, A.

SenthilKumar and N. S. Babu, "IoT based ECG Signal

Feature Extraction and Analysis for Heart Disease Risk

Assessment,"

M. T. Almalchy, A. S. Al-Khaleefa, M. A. Alazzawi, A.

AlShammari, H. M. Albehadili and H. A.AL-wzwazy,

"Cloud Computing Approach for ECG Diagnose

Module,"

A. Ristovski and M. Gusev, "SaaS solution for ECG

monitoring expert system,"

O. Kocabas and T. Soyata, "Utilizing Homomorphic

Encryption to Implement Secure and Private Medical

Cloud Computing,"

J. Zhou, J. Riekki, W. Zhang and T. Qiu, "HRV: Cloud-

Based Mobile Heart Rate Variability Monitoring

System,"

E. A. -A. Karajah and I. Ishaq, "Online Monitoring Health

Station Using Arduino Mobile Connected to Cloud

service: “Heart Monitor” System,"

INCOFT 2025 - International Conference on Futuristic Technology