Jul 2026· Journal of Medical and Health Studies· Vol 7, pp. 92-97· 0 citations· 11 references
TL;DR
This paper analyzes the challenges of existing methods in terms of data heterogeneity, modal absence, interpretability, interpretability, and privacy protection, explores the deficiencies in the dataset construction and evaluation system, and proposes corresponding solutions.
Abstract
Smart healthcare is moving towards precision, personalization, and intelligence. Multimodal data fusion technology, as a core means to break medical data silos and improve clinical decision-making efficiency, has received extensive attention. This paper systematically reviews the mainstream methods and their performance in typical medical tasks from four levels: data level, feature level, decision level, and hybrid fusion. Based on this, the current commonly used medical multimodal datasets and evaluation criteria are introduced, and the experimental results of various fusion methods on public datasets are summarized. Further, this paper analyzes the challenges of existing methods in terms of data heterogeneity, modal absence, interpretability, and privacy protection, explores the deficiencies in the dataset construction and evaluation system, and proposes corresponding solutions. Finally, it looks forward to future research directions, emphasizing the synergy of medical-engineering integration, privacy computing and explainable models to provide references for the in-depth research and application of multimodal fusion technologies in smart healthcare.
The aging of the global population has intensified the demand for innovative and sustainable healthcare strategies. Wearable technologies stand out as a promising alternative to enable real-time monitoring, early detection of clinical changes, and the promotion of autonomy and well-being among the older adults. This study presents a proposal for an integrated data platform aimed at monitoring elderly health through wearable devices. The proposed architecture comprises biometric sensors, secure wireless connectivity, intelligent data processing using machine learning algorithms, and interactive dashboards for decision-making. Through a comprehensive literature review and analysis of current applications, the platform was designed to ensure interoperability with healthcare systems, compliance with data protection laws, and usability for older adults. The results point to the potential of wearable technologies to support preventive, personalized, and population-based care models. Future challenges include issues of data integration, user acceptance, cost-effectiveness, and regulatory frameworks. The implementation of such platforms may contribute significantly to transforming the current healthcare model into a more inclusive, proactive, and data-driven system.
Lucas de, F. Gomes, Av et al.· JOURNAL OF BIOENGINEERING, T...· 0 citations
The data in healthcare is highly heterogeneous and is continuously increasing from Electronic Health Records (EHRs), medical imaging, wearable Internet of Things (IoT) devices, laboratory systems, and genomic sequencing, which makes it difficult to support intelligent clinical decision making. Current cloud-based healthcare systems may lack sufficient interoperability, scalability, privacy protection, and integration with multiple modalities, due to various factors. The paper proposes a novel multi-modal cloud architecture, which combines edge computing, cloud data lakes, multimodal feature fusion, explainable artificial intelligence (XAI) and federated learning technologies to achieve secure and scalable predictive clinical analytics. The architecture uses deep learning techniques, such as Convolutional Neural Networks, Vision Transformers, Long Short-Term Memory networks, Graph Neural Networks, and multimodal Transformers, to process and analyse diverse healthcare data types. The superior disease prediction accuracy (99.21%), lower latency (42 ms), greater privacy preservation and greater clinical interpretability over existing healthcare frameworks are confirmed by the experimental evaluation performed on benchmark datasets, including MIMIC-IV, eICU, CheXpert, MIMIC-CXR, PhysioNet, BRATS, NIH Chest X-ray and UK Biobank.
Deepika Dhamija, Sonali Rahul Dhave, Anshita Shukla et al.· Journal of Intelligent Decis...· 0 citations
Smart healthcare systems can be viewed as a paradigm shift in the contemporary medical practice, since modern placement of sensing technologies, electronic health records (EHRs), and predictive machine learning (ML) technologies have become a common practice that facilitates proactive, personal, and efficient healthcare delivery. Such trends as the growing rate of chronic illnesses, the aging trend, and escalating healthcare costs have led to a desperate demand of smart systems able to diagnose and predict risks early, as well as make the optimal clinical decision. Predictive machine learning predictive models are models based on historical and real-time healthcare data that reveal the latent patterns, predict the disease progress, and assist clinicians to make evidence-based decisions. In this paper, predictive machine-learning has been adopted to analyze in detail the concept of smart healthcare systems, which involve system architecture, data acquisition, feature engineering, model development, and performance evaluation. The suggested framework combines the wearable sensor data, clinical records, and demographic information into a single analytics pipeline. Some of the machine learning algorithms under analysis and supervised or unsupervised, like logistic regression, support vector machines, random forests, gradient boosting, and deep neural networks, are discussed in the context of healthcare prediction tasks, such as disease risk prediction and hospital readmission prediction and patient outcome prediction. Moreover, the paper also touches on the issues associated with data quality, privacy, security, model interpretability and ethical issues. Recent experimental findings indicate that predictive ml-based healthcare systems are much better than conventional rule-based healthcare systems in terms of prediction accuracy and decision support. The results also show the promise of predictive machine learning to turn healthcare into either a reactive treatment or a preventive, personalized healthcare, as part of the vision of smart and sustainable healthcare ecosystems.
Chinedu Okafor· International Journal of Art...· 0 citations
The recent fast progress of artificial intelligence (AI) has changed the general situation in the sphere of healthcare dramatically as the creation of smart systems that can provide individual medical advice is possible. The conventional health care models are largely based on standardized treatment regimens and hence they seldom take into account individual differences that could be in genetic, physiological and behavioral aspects. This drawback has resulted in the development of AI-based personalized healthcare recommendation systems that are intended to give specified interventions, foretelling revelations, and adaptive treatment plans to individual patients. The current paper is a detailed discussion on the AI-based personalized healthcare recommendation systems, their designs, procedures, and uses before 2018. The paper will look at how machine learning algorithms like supervised learning, unsupervised learning and hybrid models have been used to process patient data in the form of electronic health records (EHRs), wearable sensor data, and genomic data. These systems have also been improved in terms of scale and efficiency with the integration of big data analytics and cloud computing. Other critical topics that are being discussed in the paper include data heterogeneity, privacy issues, model interpretability and clinical validation. It particularly focuses on such methods of recommendation as collaborative, content-based, and customized approaches to recommendations. Mathematical expression of prediction model, and measure of similarity are discussed to give a theoretical basis of system design. In addition, the paper measures the performance of the system through measures like accuracy, precision, recall and patient satisfaction indices. A comparative study helps to point out how well AI-based systems can be effective in terms of bettering health results, decreasing readmission rates, and increasing the effectiveness of the decisions made by clinicians. According to the results, AI-powered personalized healthcare can transform the field of patient care and make it proactive, preventive, and precision medicine. Nonetheless, challenges of ethics, regulations and technical issues must be overcome to achieve success in implementation. The conclusion of this paper presents the future directions of research to enhance the robustness, ease-of-interoperability, and clinical adoption of systems.
Tendai Chikore· International Journal of Mod...· 0 citations
This paper presents a comprehensive review of IoT-based smart health risk prediction systems that integrate Artificial Intelligence (AI), biomedical sensors, and Internet of Medical Things (IoMT) technologies for advanced healthcare monitoring and chronic disease management. The study discusses the architecture of IoMT systems, which utilize wearable and implantable sensors to continuously collect physiological and biomedical data from patients in real time. These data are analyzed using Machine Learning (ML) and Deep Learning (DL) algorithms for disease prediction, classification, and early diagnosis of conditions such as cardiovascular diseases, diabetes, and respiratory disorders. The review highlights the major components of IoMT systems, including data acquisition, wireless data transmission, intelligent data processing, and healthcare user interfaces for clinical decision-making. Communication technologies such as Bluetooth, Wi-Fi, and edge computing are discussed for enabling efficient and low-latency healthcare monitoring. The paper also examines interoperability standards including HL7 and FHIR for secure and scalable healthcare data exchange. Furthermore, major challenges associated with IoMT healthcare systems, such as data privacy, cybersecurity, interoperability, scalability, and AI bias, are critically analyzed. Emerging technologies including Federated Learning, blockchain, Explainable AI (XAI), and energy-efficient wearable sensors are also reviewed as future research directions for intelligent healthcare systems. The review demonstrates that AI-integrated IoMT systems can importantly improve remote patient monitoring, predictive healthcare analytics, and personalized medical services.
Sushilkumar S. Salve, Nagesh B. Mapari, H. Sarode et al.· Journal of integrated scienc...· 0 citations
The research offers a performance assessment and structural issues affecting reliability, including sensor diversity, dominance, noise, and intermittent data loss due to motion artifacts and dropouts, and explores multimodal fusion failure modes, demonstrating how asynchronous failure and partial observability can cause instability in multimodal representations.
N. O. Adelakun, M. Olajide, S. Omolola· Smart Wearable Technology· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
