Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 746 https://internationalpubls.com A Predictive Analytics-Powered Smart Healthcare Monitoring System to Improve Patient Care Mr. Niteen Babanrao Dhule1, Dr. Pankaj Agarkar2 1Student, Department of Computer Engineering, Ajeenkya D. Y. Patil School of Engineering, Lohegaon, Savitribai Phule Pune University, Pune, Maharashtra 2Assistant Professor, Department of Computer Engineering, Ajeenkya D. Y. Patil School of Engineering, Lohegaon, Savitribai Phule Pune University, Pune, Maharashtra Mail id: ndhule1988@gmail.com1, panakj.agarkar@dypic.in2 Article History: Received: 12-01-2025 Revised: 15-02-2025 Accepted: 01-03-2025 Abstract: The rapid evolution of healthcare demands innovative systems to address the increasing need for personalized, real-time care. Traditional healthcare practices face significant challenges in managing the growing prevalence of chronic diseases and aging populations. In response, smart healthcare monitoring systems powered by the Internet of Things (IoT) and advanced sensor technologies offer a promising solution. These systems enable continuous, real-time monitoring of vital health metrics such as heart rate, blood pressure, glucose levels, and oxygen saturation, providing healthcare professionals with a comprehensive view of a patient’s health status. The integration of predictive analytics further enhances these systems by leveraging historical data and machine learning algorithms to forecast potential health events, enabling proactive care and early interventions. This shift from reactive to proactive care significantly improves patient outcomes and optimizes healthcare resources. The Smart Healthcare Monitoring System (SHMS) combines IoT, big data, and artificial intelligence to provide real-time health insights, predictive analytics, and personalized care. It also helps healthcare providers monitor the effectiveness of treatments in real-time, ensuring better decision-making. By focusing on chronic disease management, the system offers early detection of potential complications, reducing hospital readmissions and improving patient quality of life. However, challenges such as data security, scalability, and interoperability remain, particularly in resource-limited settings. Future developments in IoT, predictive analytics, and data security will further enhance the system’s effectiveness and broaden its scope, making it a valuable tool in modern healthcare delivery. Keywords: Smart Healthcare, Predictive Analytics, Internet of Things (IoT), Machine Learning, Chronic Disease Management. I. INTRODUCTION There has never been a greater need for more effective and efficient systems given the rapid developments in healthcare. With an ageing population and a rise in chronic illnesses worldwide, traditional healthcare institutions typically find it difficult to meet the growing demand for individualised, real-time treatment. In order to overcome these obstacles, intelligent healthcare monitoring systems have emerged. Advanced sensor technology and the Internet of Things (IoT) are used by smart healthcare monitoring systems to allow for continuing, real-time monitoring of a patient's vital signs and health parameters. In order to give a healthcare professional a more Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 747 https://internationalpubls.com comprehensive and current understanding of a patient's health, smart healthcare monitoring systems may gather a wide range of health data, including heart rates, blood pressure, glucose levels, and oxygen saturation levels. Predictive analytics unlocks the smart healthcare monitoring system's true potential [1]. Predictive analytics forecasts a patient's future health occurrences using machine learning algorithms and historical records, enabling early and preventive treatments. Predictive models, for instance, can identify early warning indicators of possible health problems, such heart attacks, diabetic crises, or respiratory failure, before they show up clinically by examining trends in a patient's vital signs[2]. The quality of healthcare delivery has significantly improved with this move from reactive to proactive treatment. In addition to improving patient outcomes, predictive analytics eases the workload for medical staff by enabling them to concentrate on the most vulnerable patients, so maximising the use of available resources. The healthcare sector is changing as a result of the convergence of IoT, big data, and artificial intelligence, particularly with regard to patient monitoring and care delivery. An analogous game- changing project may be a predictive analytics-enabled Smart Healthcare Monitoring System (SHMS) [3] [4]. By providing the doctor with up-to-date, data-driven insights into the patient's state, a SHMS has the potential to revolutionise patient care. To give doctors access to a wealth of precise data, the information is gathered from the patient's wearable technology and other sensors, combined, examined, and interpreted. Like telemetry, SHMS can provide ongoing patient monitoring and notify the doctor when necessary to prevent delays in diagnosis and treatment. In addition to enabling the early identification of anomalous issues, this patient monitoring will provide continuous, real-time monitoring of the effects of recommended treatments. Compared to basic monitoring, the SHMS gains an additional layer of intelligence if predictive analytics is incorporated. In order to determine the probability of particular health occurrences in the future—such as acute exacerbations of chronic diseases or crisis signs and symptoms predictive analytics would use machine learning. In the field of managing chronic diseases, where early detection and treatment can enhance long-term prognoses and general quality of life, this kind of foresight is crucial [4][5]. Predictive models, for instance, can provide early warnings about potential changes in vital signs as trends emerge in patients with diabetes or heart disease. This can alert clinicians to a changing condition trend towards a potential complication before it poses a threat to health. All things considered, combining predictive analytics with continuous monitoring promotes a proactive, individualised approach to care that may lead to better patient outcomes, lower healthcare costs, and fewer hospital stays [6]. Fig 1: Smart Healthcare System Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 748 https://internationalpubls.com The image depicts the workflow of a Smart Healthcare System that involves several essential elements and processes. It commences with the user's health being tracked by wearables or sensors that record the necessary vital readings. These readings are then sent to the Internet of Things (IoT) to establish data transfer [7]. The data is sent to a data center which analyzes all of the data minutes or hours later to derive a "snapshot" of the user's health status. The user or health worker may retrieve health information from the data center in order to make decisions or take action. It is a Smart Healthcare System, as it fuses real-time monitoring, IoT technology and data analytics research for health care delivery [8]. II. RELATED WORK A system for managing chronic illnesses through ongoing patient monitoring and physician collaboration is put out by Nora Mahmoud et al. (2020). Cloud computing, fog computing, wireless body area networks (WBAN), semantic ontology, and a clinical decision support system (CDSS) are all integrated into the system. This method improves patient quality of life [9] while reducing healthcare expenditures by enabling real-time monitoring, remote supervision, and shorter hospital stays. The integration of WBAN data into electronic health records (EHRs) is another way that the framework handles semantic interoperability [10]. During the COVID-19 pandemic, when hospital visits are discouraged, Olutosin Taiwo et al. (2020) suggest a remote smart home healthcare support system (ShHeS) to monitor patients' health and get prescriptions from physicians while they remain at home [11]. The solution reduces the need for in-person visits while guaranteeing ongoing health monitoring and treatment by leveraging developments in IoT, smart home automation, and healthcare technology to enable clinicians to diagnose conditions remotely by gathering and evaluating patient data. The increasing use of artificial intelligence (AI) in healthcare, namely in remote patient monitoring (RPM) systems, is reviewed by Thanveer Shaik et al. (2023). RPM helps with the monitoring of hospitalised patients, elderly people receiving home care, and patients with acute or chronic diseases. AI-enabled RPM systems offer continuous, non-invasive health monitoring via IoT wearables and sensors, in contrast to traditional monitoring, which depends on invasive techniques and human staff supervision. The influence of AI in RPM is examined in the paper, with particular attention paid to behaviour pattern identification through reinforcement learning, personalised monitoring using federated learning, and health deterioration detection. It focusses on the advantages, difficulties, and potential developments of AI in RPM, highlighting how it might revolutionise patient care and solve implementation issues[9]. The necessity for cutting-edge healthcare treatments in response to chronic illnesses like COVID-19 is emphasised by Fazlullah Khan et al. (2021). The function of IoT-enabled smart wearables, which gather enormous volumes of data about behavioural, psychological, and physical health, is examined in this research. It stresses the significance of big data analytics and machine learning (ML) in enhancing healthcare systems and talks about the difficulties in managing the large amounts of data produced by these devices. The paper examines current machine learning approaches for healthcare big data analysis, highlighting their advantages, disadvantages, and areas for further research. It provides insightful information for policymakers and healthcare professionals. Using the Internet of Things (IoT) for real-time monitoring and predictive analytics, John Owen et al. (2024) investigate the revolutionary effects of smart healthcare [11]. Vital signs and health status are continuously tracked Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 749 https://internationalpubls.com by IoT-enabled devices, including wearables, sensors, and smart medical equipment. Clinicians may take preventative measures and identify any problems early thanks to this real-time data, which gives them an accurate and current picture of patient situations. By predicting results using both historical and current data, predictive analytics further improves decision-making and enables prompt modifications to treatment regimens. Smart healthcare solutions seek to enhance patient outcomes, maximise resources, and lower healthcare costs as IoT and predictive analytics develop, bringing about a more adaptable, data-driven healthcare system [10] [11]. Research gap Research on smart healthcare systems, specifically those designed with predictive analytics and IoT, is advancing rapidly, but gaps remain. While frameworks, like those presented by Nora Mahmoud et al. (2020) and Olutosin Taiwo et al. (2020), deal with remote patient monitoring and clinician collaboration, there continue to be challenges with seamless integration, real-time decision-making support, and interoperability of systems. In addition, while using artificial intelligence (AI) in remote patient monitoring (RPM) systems is promising, as stated by Thanveer Shaik et al. (2023), further exploration is needed regarding privacy, data-security, and how to scale AI systems. The importance of big data analytics was presented by Fazlullah Khan et al. (2021), while there is little focused research on how machine learning techniques have evolved for IoT healthcare applications. Lastly, as highlighted by John Owen et al. (2024), predictive analytics in smart healthcare still has challenges in terms of integrating technologies into various healthcare contexts, such as rural and resource-poor healthcare settings. Closing some of these gaps will improve the efficiency, security, and accessibility of smart healthcare solutions [12]. III. RESEARCH METHODOLOGY In order to support ongoing patient monitoring and proactive healthcare management, the approach for creating a Smart Healthcare Monitoring System with Predictive Analytics for Enhanced Patient Care is centred on the integration of cutting-edge technologies, such as the Internet of Things (IoT), machine learning (ML), predictive analytics, and wireless body area networks (WBAN). Design of the System: IoT-enabled wearable sensors are part of the proposed system's design, which allows for real-time monitoring of vital indicators including blood pressure, temperature, heart rate, and oxygen saturation. Using communication protocols like Wi-Fi or Bluetooth for local connectivity and LoRa (Long Range) for reliable long-distance transmission, these sensors provide data to a cloud platform. To deliver useful health insights, the system is connected with a Clinical Decision Support System (CDSS). Data collection: Wearable technology is used to continually collect patient data, guaranteeing that vital health indicators are constantly tracked. After that, data is sent to local and cloud storage platforms. Semantic interoperability is emphasised in the system's architecture to efficiently integrate data from many sources. Furthermore, strong security measures are put in place to guarantee adherence to healthcare laws, protecting patient confidentiality and data integrity. Predictive Analytics & Model Development: Both historical and real-time data are analysed using machine learning techniques, which include supervised learning (like regression and classification) Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 750 https://internationalpubls.com and unsupervised learning (like clustering). To anticipate possible health problems, identify irregularities, and offer early warnings, predictive models are created. Individual patient health monitoring is customised through the use of techniques like Reinforcement Learning and Federated Learning. Metrics like accuracy, precision, and recall are used to evaluate performance. System Evaluation: To assess the system's scalability, effectiveness, and influence on healthcare outcomes, it is evaluated in both clinical and non-clinical contexts, including rural and metropolitan areas. Its capacity to lower hospital readmissions, improve patient involvement, and maximise resource use is evaluated through comparisons with conventional healthcare monitoring techniques. Issues like data interoperability and AI model scalability are also covered, with future research attempting to resolve these problems and enhance system flexibility. This all-encompassing strategy guarantees that the suggested system may greatly improve healthcare delivery by increasing clinical results, cutting expenses, and improving patient monitoring. IV. RESULT AND DISCUSSION This section represents the findings derived from the proposed smart healthcare monitoring system with predictive analytics. This section critically analyzes the outcomes, compares them with existing solutions, and discusses the implications for patient care and healthcare system improvements. Fig 2: Smart Health Monitor Login Screen The image shows the login screen of the "Smart Health Monitor" system, designed to offer personalized health management. The interface is simple and user-friendly, providing two main login options: one for patients and another for healthcare providers. This structure allows the system to cater to both individuals seeking health monitoring and healthcare professionals managing multiple patients. Users can also access options for password recovery and sign-up, making the platform accessible and easy to use. The background, featuring a stethoscope, reinforces the medical nature of the platform, emphasizing its healthcare focus. This login screen is the entry point for users to access various functionalities such as health tracking, data analysis, and communication with healthcare providers, thus enabling eamless integration of health management technologies in everyday life. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 751 https://internationalpubls.com Fig 3: Smart Health Monitor Dashboard and Health Risk Prediction The picture shows the "Smart Health Monitor" system's dashboard, which provides a thorough summary of a user's health. It has vital indicators including blood pressure, heart rate, and oxygen saturation that are shown graphically to monitor changes over time. Based on current data, the "Health Risk Prediction" section warns the user of possible hazards, suggesting a moderate risk of cardiovascular problems and recommending a medical examination. Recent warnings that highlight issues including elevated blood pressure, irregular heartbeats, and dehydration are displayed beneath the health metrics. Additionally, the system offers tailored suggestions for enhancing well-being, like cutting back on salt consumption, increasing water intake, and keeping an eye on heart rate trends. The system's integration of predictive analytics enables users to take proactive measures for improved health and supports physicians in making defensible judgements based on up-to-date information.. Fig 4: Smart Health Monitor Predictive Alerts and Recommendations Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 752 https://internationalpubls.com The "Smart Health Monitor" system's "Predictive Alerts" part is seen in the picture. Users are empowered to take prompt action by this section's real-time notifications on problematic health data. Three alarms are highlighted by the system: Low Hydration Level, Irregular Heart Rate Pattern, and High Blood Pressure Alert. Every warning provides information about the health issue's present state, suggested courses of action, and ways to make an appointment with a medical professional. For example, users are recommended to keep hydrated and continue to check their blood pressure in reaction to high blood pressure. Likewise, consumers are advised to speak with a healthcare professional if their heart rate is erratic, as this is associated with possible arrhythmias. By identifying and offering advice on critical medical concerns, the system provides a thorough approach to proactive health management. Fig 5: Smart Health Monitor: Predictive Alerts and User Recommendations The picture displays the "Smart Health Monitor" system's Predictive Alerts section, which focusses on three major health issues: low hydration, irregular heart rate, and high blood pressure. With every alert, the user receives personalised suggestions. It recommends getting fresh readings and drinking plenty of water if you have high blood pressure. The irregular heart rate signal suggests getting medical help and keeping an eye on the heart rate for the following hour. In order to assist the user maintain their water consumption, the low hydration indicator has a reminder option and highlights the significance of quick hydration. The technology is intended to promote proactive health management and offer real-time insights. The system assists in preventing major health crises and fosters general wellbeing by identifying possible problems and directing users towards healthier behaviours. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 753 https://internationalpubls.com Fig 6: Smart Health Monitor: Personalized Recommendations and Health Goals Tracker The "Smart Health Monitor" system's Personalised Recommendations and Health Goals area is seen in the picture. It offers personalised health guidance with an emphasis on topics like hydration and cardiovascular health. Reducing salt intake, beginning a walking regimen, and drinking more water are among the suggestions. A Mediterranean diet is also recommended by the system to enhance heart health. Users may also monitor their health goals, which include controlling resting heart rate, boosting daily step count, and reducing blood pressure. Users may alter their objectives using the UI, which makes it simpler to track their health progress and modify their lifestyle as necessary. Users may take charge of their health and make wise decisions with the aid of this system's actionable information. Fig 7: Smart Health Monitor: Account Settings and Profile Management Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 754 https://internationalpubls.com The picture shows the "Smart Health Monitor" system's Account Settings area, where users may edit and manage their personal profiles and health information. The user's name, email address, and phone number are among the essential information elements in this area. It also offers choices for entering medical information, such as blood type, height, weight, and date of birth. The emergency contact information is an essential component of this area. It allows users to add significant contacts, such a spouse or family member, along with their relationship and phone number, guaranteeing that the appropriate individuals can be contacted in the event of an emergency. With the ability to save modifications after making any changes, the interface is easy to use. This component is essential for maintaining the health monitoring system's personalisation and readiness to deliver precise, emergency-aware services, as well as for guaranteeing effective administration of the user's profile and medical data. Fig 8: Health Monitoring Data for Respiratory Imbalances and Vital Parameters A dataset pertaining to health monitoring metrics, including a range of measures for identifying respiratory imbalances and other medical conditions, is shown in the picture. Numerous health factors, including dehydration, medicine overdose, coughing, cold, heart rate, temperature, oxygen saturation, and others, are included in the data, which is divided into several categories, including chronic, mild, normal, and severe problems. Every medical problem is examined in light of how it affects these crucial metrics. For instance, it provides useful information on the physiological effects of various disorders by including metrics such as pulse, BPYSYS, BPDIA, and respiratory rate across a range of severity levels. Predictive analytics and health evaluations are probably supported by this information, allowing medical practitioners to efficiently monitor and intervene. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 755 https://internationalpubls.com Fig 9: Machine Learning Model Evaluation: Confusion Matrix and Classification Metrics A portion of Python code pertaining to machine learning model assessment is seen in the picture. To evaluate a classification model's performance, a confusion matrix is created. The classification report is then used to calculate a number of metrics, including precision, recall, F1-score, and support. Deeper insights into model performance, especially for unbalanced datasets, are provided by the code's computation of the Matthews Correlation Coefficient (MCC) and balanced accuracy. Cross-validation is used to evaluate generalisation performance as part of the model assessment process. This stage is essential for confirming the prediction power of the model and making sure it works well with different datasets. DISCUSSION: By using predictive analytics, the Smart Health Monitor system offers a thorough method of managing and monitoring healthcare. The system enables proactive actions by empowering users and healthcare professionals to monitor and control critical health data, as shown in Figures 2 through 7. With a focus on security and accessibility, the login screen (Fig 2) provides a user-friendly interface that serves both patients and healthcare professionals. After logging in, users are shown a dashboard (Fig. 3) that shows health parameters such as blood pressure, oxygen saturation, and heart rate. Based on real-time data, the dashboard also predicts health risks. The section on predictive alarms (Figs. 4 and 5) identifies possible health hazards including hypertension and irregular heartbeats and provides personalised advice to reduce them. Furthermore, users are guided towards better lifestyle choices by personalised suggestions and health objectives (Fig. 6), which strengthens the system's role in preventive care. Additionally, the Account Settings (Fig 7) function guarantees that users can effectively maintain their emergency contact details and profiles, enabling individualised treatment. The system is a useful tool Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 756 https://internationalpubls.com for medical practitioners as the dataset (Fig. 8) provides predictive analytics for the diagnosis and treatment of ailments including respiratory imbalances. Finally, the system's capacity to provide precise and trustworthy health insights is further improved by the incorporation of machine learning models for predictive analytics, which are assessed using confusion matrices and other metrics (Fig 9). All things considered, the Smart Health Monitor offers a comprehensive, data-driven method of managing healthcare, encouraging improved patient outcomes and resource efficiency. CONCLUSION: The Smart Health Monitor system, which combines cutting-edge machine learning methods with continuous health monitoring and predictive analytics, provides a revolutionary approach to patient care. The technology enables proactive, data-driven treatments by empowering people and healthcare professionals to track important health variables. With separate login choices for patients and healthcare professionals, the system's user-friendly interface, as shown in Fig. 2, guarantees accessibility and security for users. Together with real-time health risk forecasts, the dashboard (Fig. 3) offers a thorough summary of important health parameters, facilitating the early identification of any health problems. By providing prompt warnings of aberrant health data, the predictive alerts (Figs. 4 and 5) assist users in taking preventative actions like changing their lifestyle or seeing a doctor. Additionally, the section on health objectives and personalised suggestions (Fig. 6) offers practical advice to promote healthy habits, which has a direct effect on long-term wellbeing. By enabling users to effectively update their emergency and medical contact information, the Account Settings (Fig 7) provide personalised care management. A further layer of intelligence is added to healthcare management by the system's capacity to use machine learning models for predictive analytics (Fig. 9) and monitor situations like respiratory imbalances via datasets (Fig. 8). The system's use of sophisticated algorithms improves patient outcomes overall, optimises resource allocation, lowers hospital readmissions, and strengthens the decision-making skills of medical personnel. In the end, the Smart Health Monitor system represents a comprehensive, proactive, and customised approach to contemporary healthcare, achieving notable advancements in both patient care and system effectiveness. LIMITATION: The Smart Health Monitor system has several interesting features, but there are also some drawbacks to take into account. The system's primary reliance is on wearable technology and sensors enabled by the Internet of Things (IoT), which may have technical drawbacks including poor battery life, poor connection, or inaccurate sensors. These restrictions may make it more difficult for the system to offer accurate, ongoing monitoring, especially in rural or isolated locations with inadequate infrastructure. Furthermore, since cloud platforms are used to transport and retain sensitive patient data, there are serious data storage and security issues. There is still a chance of data breaches or illegal access to sensitive health information even with strong security measures in place. Predictive analytics presents another drawback. The quality and volume of accessible historical data determine how accurate prediction models are, even if they might provide insightful information about possible health problems. Incomplete or inaccurate data might result in inaccurate forecasts, which could cause missed diagnosis or false alarms. Furthermore, the system's machine learning models may not always provide clear or useful justifications for their predictions, which might restrict the uptake of these models by Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 757 https://internationalpubls.com healthcare professionals. This is known as model interpretability. Lastly, there can be issues with the system's scalability and compatibility with the current healthcare infrastructure. To maintain effective and efficient performance across various contexts, integration with multiple health systems may need considerable modification and continuous maintenance, especially in areas with limited resources or antiquated technology. Future scope: Despite being successful in its present configuration, the Smart Health Monitor system has room for improvement and will only develop further. Enhancing wearable technology with Internet of Things capabilities is one area of future development. To solve the issues of battery economy, connection, and sensor accuracy, technological advancements may lead to the development of more robust, accurate, and energy-efficient sensors. Better ongoing monitoring may result from this, particularly in rural regions and nations that export resources. Furthermore, improvements in predictive analytics may enhance the system's capacity to provide more accurate forecasts. Predictive dependability may increase with the use of better machine learning models, more and more varied datasets, and enhanced health risk forecasting. Furthermore, the possible addition of deep learning algorithms may enhance interpretability and make it simpler for medical experts to use the system. Lastly, interoperability and scalability are important issues that must be resolved for the system to be accepted. Although health care infrastructures vary globally, enabling enhanced functionality and integrating the Smart Health Monitor with current systems is essential to its ultimate uptake. Furthermore, improving data security using blockchain technology and encryption would allay worries about illegal access and privacy invasion. Finally, adding real-time medication interactions and mental health monitoring to the system's functionality would provide a more comprehensive healthcare solution that takes into account patients' psychological and physical demands. References: [1] Mahmoud, N., et al. (2020). A framework for managing chronic diseases through continuous patient monitoring and physician collaboration. Journal of Healthcare Engineering, 2020, 1-12. [2] Taiwo, O., et al. (2020). Remote smart home healthcare support system (ShHeS) for health monitoring during the COVID-19 pandemic. International Journal of Medical Informatics, 2020, 112-124. [3] Shaik, T., et al. (2023). Adoption of artificial intelligence in remote patient monitoring systems: Benefits, challenges, and future trends. Journal of Healthcare Technology, 2023, 89-102. [4] Khan, F., et al. (2021). The role of big data analytics and machine learning in improving healthcare systems through IoT-enabled wearables. IEEE Transactions on Industrial Informatics, 2021, 32(3), 445-456. [5] Owen, J., et al. (2024). The transformative impact of IoT and predictive analytics in smart healthcare systems. Journal of Medical Systems, 2024, 48(5), 15-30. [6] Zhang, Y., & Li, X. (2021). Predictive analytics in healthcare: An overview of trends and applications in health monitoring. Healthcare Informatics Research, 27(4), 259-267. [7] Kumar, S., & Singh, R. (2020). Machine learning for predictive healthcare: A survey of current and future trends. Journal of Medical Imaging and Health Informatics, 10(8), 1974-1984. Communications on Applied Nonlinear Analysis ISSN: 1074-133X Vol 32 No. 3s (2025) 758 https://internationalpubls.com [8] Patel, V., & Patel, S. (2021). Smart wearable devices and their impact on chronic disease management. Journal of Health Technology and Innovation, 5(3), 202-210. [9] Wang, Y., et al. (2020). IoT-based health monitoring systems for personalized medicine. Journal of Personalized Medicine, 10(4), 56-64. [10] Kim, J., & Choi, Y. (2022). AI-enabled systems in healthcare: The role of IoT and predictive analytics. Artificial Intelligence in Medicine, 117, 58-69. [11] Clark, T., & Davis, M. (2020). Improving healthcare delivery using real-time health data from IoT devices. Journal of Healthcare Management, 45(6), 510-520. [12] Gupta, N., & Mehta, A. (2021). Big data in healthcare: Managing challenges and opportunities for improved care delivery. Health Information Science and Systems, 9(1), 11-20.