Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 1205 https://internationalpubls.com Heart Disease Prediction using a Hybrid Machine Learning Model Shital Sunil Raut1, Prof. Dr. Monika Rokade2, Prof. Dr. Sunil Khatal3 1Department of Computer Engineering, Sharadchandra Pawar College of Engineering, Dumbarwadi, Otur, Pune, India shitu.rg007@gmail.com 2Department of Computer Engineering, Sharadchandra Pawar College of Engineering, Dumbarwadi, Otur, Pune, India monikarokade4@gmail.com 3HOD, Department of Computer Engineering, Sharadchandra Pawar College of Engineering, Dumbarwadi, Otur, Pune, India khatalsunil88@gmail.com Article History: Received: 12-01-2025 Revised: 15-02-2025 Accepted: 01-03-2025 Abstract: Cardiovascular disease is a prevalent issue that may be particularly severe in the elderly and in those who do not maintain a healthy lifestyle. Routine examinations and assessments, along with adherence to a nutritious diet, may partly mitigate the risk of heart disease. Hospitals provide extensive data about patients, including x-ray, pulmonary, cardiac, and thoracic pain outcomes, in addition to personal health records (PHRs). We employ a decision tree classification based on the symptoms, specifically the essential features mandatory for prediction. Utilising the decision tree technique allows us to identify the most impactful qualities that will improve our dataset predictions. Hospitals fail to effectively use the data they produce. The database uses specific techniques to identify heart disease, but it does not approve other functions. This study primarily aims to forecast cardiac illnesses via machine learning methods, informed by a synthesis of many contemporary studies. Hybrid machine learning methods will classify both real-time and synthetic samples. Keywords: Internet of Things (IoT), body sensor networks (BSN), machine learning, disease prediction systems, and monitoring systems. I. INTRODUCTION Health is an evolving process that needs continual observation. The healthcare industry is seeing several challenges related to hospital admissions owing to an increased number of patient admissions. We suggest a method for human healthcare to attain this objective. The device allows continuous monitoring of patients' metabolic parameters and illness identification using the acquired parametric values. Given the rise in unexpected fatalities attributable to chronic heart failure or hypertension, it is essential to provide continuous home health monitoring services. The primary objective was to create a dependable patient monitoring system enabling healthcare providers to observe patients in both hospital settings and during their routine everyday activities. Recently, patient monitoring systems have emerged as a significant breakthrough due to enhanced technology. Our system quantifies patients' parameters using diverse sensor data. The latest study indicates a Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 1206 https://internationalpubls.com significant rise in the death rate attributable to hearts disease. In the direction of reduce the mortality rate, an advanced cardiac disease calculation system is essential. Several factors, including lifestyle changes and increased stress, container cause heart disease. As a result, predicting heart disease is an essential component of survival. Parameters evaluated in the experiment include body temperature, blood pressure, heart rate, ECG levels, and pulse rate. Daily, we produce substantial volumes of medical data, making the extraction of significant insights from this big data a formidable challenge. The heart is essential to human existence; optimal cardiac function correlates with excellent health. II. LITERATURE SURVEY In [1] Protecting and predicting diseases in large healthcare datasets at high dimensions using machine learning. The initial part of the system is an Internet of Things (IoT)-based environment that produces patient body data. Here, you may plug in your ECG monitor, blood pressure tracker, body temperature monitor, and heart rate monitor, among other wearable gadgets. Once collected information from multiple sensors is ready, it will be uploaded to a cloud-based repository. In the second stage, we keep an eye on the information collected by several sensors. Here, an Android- based GUI was created to continuously monitor data. Where computers utilize machine learning techniques to diagnose patients. With the authentication method in place, we can restrict access to just the people who need it based on their assigned roles, and the suggested machine learning algorithms can calculate the likelihood that a certain patient has a particular ailment. According to [2] Cholesterol, triglycerides, polyunsaturated fatty acids, monounsaturated fatty acids, high-density lipoprotein, random blood sugar, QT interval, PR interval, hemoglobin status, and oxygen saturation should all be measured using IoT sensors and recorded. Sensors like blood pressure monitors, electrocardiogram monitors, and pulse rate monitors are used to collect data from patients in an Internet of Things setting. In addition, the researcher adopted a novel method of monitoring heart rate and glucose level or random sugar level using a light diode sensor due to its many benefits. It's assisting doctors in detecting heart problems early, which might save the lives of tens of thousands of individuals each year. In [3] An IoT environment built with machine learning techniques can be used to get data from a real-time body sensor network and store it on a cloud server. The BSN has an intermediate sensing system for cardiac risk assessment. The audit data utilized synthetic information for assessing the risk of cardiovascular disease. This study has demonstrated several machine learning and deep learning techniques for highly accurate disease prediction monitoring. In [4], the researchers used the RapidMiner tool and other ML algorithms to increase the previous accuracy score and predict heart disease. They analysed the heart disease database from the University of California, Irvine. The proposed method improved the previously achieved accuracy score. In [5], they present the physical equation modelling methodology that uses the partial least squares technique for data analysis. The researchers observed a correlation between coronary heart disease and hypertension and all factors except BMI. These results assisted scholars and medicinal professionals in machine learning who are endeavouring to identify relationships among these parameters. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 1207 https://internationalpubls.com According to [6], they used deep learning to create an automated method for predicting heart attacks. They assessed the accuracy of machine learning systems using various datasets. The suggested method used an automated data preparation approach in the direction of eradicating anomalies from the system. As noted in [7], we developed the Talos hyperparameter optimization framework for predicting cardiac & heart diseases. Cardiovascular disease is a significant domain in which deep neural networks may enhance the consistency of heart classification. A group-learning architecture using one-way and two-way BiLSTM or BiGRU models in conjunction with a CNN was proposed in [8]. It accurately diagnosed 91% of various forms of cardiac disease. In [9], researchers used a learning algorithm to improve their outcome goals by using deep neural networks (RNNs), specifically PPRNN, to look at patient diagnostic narratives. The suggested PP- RNN uses several RNNs to look at different sets of diagnostic data for patients in order to predict the onset of diseases with a high risk. The proposed technique enhanced precision. According to [10] a cutting-edge IoT and cloud-based medical application seeks to track and identify serious diseases. Data from the validation set is used to train the SVM classifier during the training phase. We used actual patient information during system testing to identify illnesses and the presence of diseases. III. PROPOSED SYSTEM This project involves designing a system beneficial for both healthcare centers and hospitals. We should use the dataset, which includes temperature, blood pressure, and pulse rate sensors, to ascertain patient physiological characteristics. We will use a range of effective hybrid machine learning methodologies to develop the graphical user interface. We have categorized the proposed system into two separate phases: training and testing. This study presents an efficient illness prediction model using deep learning methodologies. The dataset is crucial for achieving classification accuracy throughout the whole process. Figure 1: Proposed System Design Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 1208 https://internationalpubls.com Implement Module 1) Training: ➢ Gather dataset ➢ Utilize data mining techniques. ➢ The database stores data, also known as background knowledge, and uses it throughout the test. 2) Testing: ➢ The system works with synthetic & real-time patient dataset obtained from the Internet by predicting disease probabilities based on trained models. ➢ A centralized database stores all aggregated data using a link-oriented design. ➢ Completely testing and training dataset are read concurrently throughout the testing phase. ➢ Implement machine classification & future implementation of decision-making processes. ➢ Ultimately, assess the learning's alignment with the system's standard & essential parameters. RESULT AND ANALYSIS Integrating quantitative assessments with clinical ratings is logical and may effectively resolve several evaluation issues. We used hybrid machine learning models on the training dataset that was made to find common, suspicious, and dangerous patterns of behavior. To test and evaluate the machine learning methods, we used cross-validation models with the behavioral categorization training dataset. Figure 1.2 depicts the classification system in overall factors and illustrates the uniformity of classifiers. Figure 2: The accuracy evaluation of various ML classifiers CONCLUSION The Internet of Things architecture is an effective technology since it offers the typical individual an accessible and affordable network across several industrial sectors. Healthcare discipline is a necessary and inevitable aspect of our daily existence. The sensor dataset suggests a more Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 10s (2025) 1209 https://internationalpubls.com comprehensive framework for aggregating sensory data within the medical field and incorporating it into smart devices. Web applications, sometimes referred to as smart devices; provide the most basic degree of cognitive processing. The conventional method mostly employs intrusive testing, leading to patient worry, dissatisfaction, or negligence over their health. They find it challenging to navigate such settings. In this situation, patients can contact the doctor continuously, seven days a week, and get information using Internet technology. REFERENCES [1] Sunil S.Khatal ,Dr.Yogesh kumar Sharma, “Health Care Patient Monitoring using IoT and Machine Learning.”, IOSR Journal of Engineering (IOSR JEN), ISSN (e): 2250-3021, ISSN (p): 2278-8719 [2] Khatal, Sunil S., and Monika D. Rokade. ”An Detection of Cardiovascular Disease using Deep Learning.” 2023 International Conference on Emerging Smart Computing and Informatics (ESCI). IEEE, 2023. [3] Khatal, Sunil S., and Yogesh Kumar Sharma. ”Analyzing the role of heart disease prediction system using IoT and machine learning.” International Journal of Advanced Science and Technology 29.9s (2020): 2340-2346. [4] Fahd Saleh Alotaibi,” Implementation of Machine Learning Model to Predict Heart Failure Disease”, (IJACSA) International Journal of Advanced Computer Science and Applications,Vol. 10, No. 6, 2019. [5] Lewlyn L. R. Rodrigues, Dasharathraj K Shetty, Nithesh Naik , Chetana Balakrishna Maddodi and et al., “Machine learning in coronary heart disease prediction: Structural equation modelling approach ”, Rodrigues et al., Cogent Engineering (2020), 7: 1723198. [6] Mohd Ashraf, M. A. Rizvi and Himanshu Sharma, “Improved Heart Disease Prediction Using Deep Neural Network”, Asian Journal of Computer Science and Technology , ISSN: 2249-0701 Vol.8 No.2, 2019, pp. 49-54 © The Research Publication. [7] Sumit Sharma, Mahesh Parmar , “Heart Diseases Prediction using Deep Learning Neural Network Model ”, International Journal of Innovative Technology and Exploring Engineering (IJITEE) ISSN: 2278-3075, Volume-9 Issue-3, January 2020. [8] Asma Baccouche, Begonya Garcia-Zapirain , Cristian Castillo Olea and Adel Elmaghraby, “Ensemble Deep Learning Models for Heart Disease Classification: A Case Study from Mexico”, Information 2020, 11, 207; [9] N. Sowri Raja Pillai , K.Kamurunnissa Bee, J.Kiruthika , “Prediction of heart disease using RNN algorithm”, International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395- 0056 Volume: 06 Issue: 03 — Mar 2019 www.irjet.net p-ISSN: 2395-0072. [10] M.Ganesan and Dr.N.Sivakumar, “IoT based heart disease prediction and diagnosis model for healthcare using machine learning models “, Proceedings of International Conference on Systems computation automation and networking 2019, IEEE- 978-17281-1524