Prediction and Detection of Terminal Diseases Using Internet of Medical Things: A Review

TL;DR

The reviewed work addresses the challenge of predicting and detecting chronic and terminal diseases by integrating AI with the Internet of Medical Things (IoMT). It surveys a wide range of ML and DL techniques applied to heterogeneous data from IoMT devices, medical imaging, and EMRs, highlighting high performance in conditions like heart disease, CKD, Alzheimer’s, and liver/pancreatic diseases. The paper emphasizes data availability, privacy-preserving strategies, and the need for data standardization to improve generalizability across clinical settings, while also outlining gaps in multi-morbidity modeling and deployment in real-world environments. It proposes future directions such as transfer learning, ensemble methods, and standardized open tools to integrate federated learning, blockchain, and differential privacy into IoMT systems, aiming to enhance robustness, privacy, and scalability in predictive healthcare. Overall, the review underscores the substantial potential of AI-IoMT to advance early prediction and personalized care, while calling for standardized data practices and privacy safeguards to enable broader clinical impact.

Abstract

The integration of Artificial Intelligence (AI) and the Internet of Medical Things (IoMT) in healthcare, through Machine Learning (ML) and Deep Learning (DL) techniques, has advanced the prediction and diagnosis of chronic diseases. AI-driven models such as XGBoost, Random Forest, CNNs, and LSTM RNNs have achieved over 98\% accuracy in predicting heart disease, chronic kidney disease (CKD), Alzheimer's disease, and lung cancer, using datasets from platforms like Kaggle, UCI, private institutions, and real-time IoMT sources. However, challenges persist due to variations in data quality, patient demographics, and formats from different hospitals and research sources. The incorporation of IoMT data, which is vast and heterogeneous, adds complexities in ensuring interoperability and security to protect patient privacy. AI models often struggle with overfitting, performing well in controlled environments but less effectively in real-world clinical settings. Moreover, multi-morbidity scenarios especially for rare diseases like dementia, stroke, and cancers remain insufficiently addressed. Future research should focus on data standardization and advanced preprocessing techniques to improve data quality and interoperability. Transfer learning and ensemble methods are crucial for improving model generalizability across clinical settings. Additionally, the exploration of disease interactions and the development of predictive models for chronic illness intersections is needed. Creating standardized frameworks and open-source tools for integrating federated learning, blockchain, and differential privacy into IoMT systems will also ensure robust data privacy and security.

Figures (5)

• Figure 1: Flowchart adapted from pagePRISMA2020Statement2021 illustrating the PRISMA-based selection process for the 67 research articles that met our inclusion criteria, in alignment with the PRISMA guidelines.
• Figure 2: Showing year of publication and name of publisher of each report
• Figure 3: The number of papers reporting on each area of the disease under consideration, as well as the distribution of papers based on ML and DL models.
• Figure 4: horizontal bar chart summarising IoMT-data privacy strategies reported in the review and pie chart summarizing the number of reports for various aspects covered in the review
• Figure 5: This illustration demonstrates the various data sources and the deployment of privacy protocols during data transmission for medical prognosis.