A Deep Auto-Optimized Collaborative Learning (DACL) model for disease prognosis using AI-IoMT systems

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: May 4, 2024

Abstract In modern healthcare, integrating Artificial Intelligence (AI) and Internet of Medical Things (IoMT) is highly beneficial has made it possible to effectively control disease using networks interconnected sensors worn by individuals. The purpose this work develop an AI-IoMT framework for identifying several chronic diseases form the patients’ medical record. For that, Deep Auto-Optimized Collaborative Learning (DACL) Model, a brand-new framework, been developed rapid diagnosis like heart disease, diabetes, stroke. Then, Auto-Encoder Model (DAEM) used in proposed formulate imputed preprocessed data determining fields characteristics or information that are lacking. To speed up classification training testing, Golden Flower Search (GFS) approach then utilized choose best features from data. addition, cutting-edge Bias Integrated GAN (ColBGaN) model created precisely recognizing classifying types records patients. loss function optimally estimated during Water Drop Optimization (WDO) technique, reducing classifier’s error rate. Using some well-known benchmarking datasets performance measures, DACL’s effectiveness efficiency evaluated compared.

Language: Английский

---

Ensemble-based Heart Disease Diagnosis (EHDD) Using Feature Selection and PCA Extraction Methods

The Open Bioinformatics Journal, Journal Year: 2025, Volume and Issue: 18(1)

Published: Feb. 6, 2025

Introduction Heart disease is a growing health crisis in India, with mortality rates on the rise alongside population. Numerous studies have been undertaken to understand, predict, and prevent this critical illness. The dimensionality of dataset, other hand, reduces prediction's accuracy. Methods We propose an Ensemble-based Disease Diagnosis (EHDD) model which dimension lowered through filter-based feature selection. experimental conducted using UCI Cleveland dataset cardiac disease. precision achieved three key steps. scatter matrix utilized divide distinct class points first phase, highest eigenvalue eigenvectors are picked for new decreased dataset. extraction carried out second stage utilizing statistical approach based mean, covariance, standard deviation. Results classification component uses training test datasets smaller sample space. last samples into two groups: healthy subjects diseased subjects. Since basic binary classifier will not yield best results, ensemble strategy SVM. Conclusion Random Forest chosen create accurate predictions. When compared existing models, suggested EHDD outperforms them by 98%.

Language: Английский

---

Navigating Challenges and Harnessing Opportunities: Deep Learning Applications in Internet of Medical Things

Future Internet, Journal Year: 2025, Volume and Issue: 17(3), P. 107 - 107

Published: March 1, 2025

Integrating deep learning (DL) with the Internet of Medical Things (IoMT) is a paradigm shift in modern healthcare, offering enormous opportunities for patient care, diagnostics, and treatment. Implementing DL IoMT has potential to deliver better diagnosis, treatment, management. However, practical implementation challenges, including data quality, privacy, interoperability, limited computational resources. This survey article provides conceptual framework synthesizes identifies state-of-the-art solutions that tackle challenges current applications DL, analyzes existing limitations future developments. Through an analysis case studies real-world implementations, this work insights into best practices lessons learned, importance robust preprocessing, integration legacy systems, human-centric design. Finally, we outline research directions, emphasizing development transparent, scalable, privacy-preserving models realize full healthcare. aims serve as foundational reference researchers practitioners seeking navigate harness rapidly evolving field.

Language: Английский

---

A comprehensive survey of honey badger optimization algorithm and meta-analysis of its variants and applications

Franklin Open, Journal Year: 2024, Volume and Issue: 8, P. 100141 - 100141

Published: Aug. 10, 2024

Metaheuristic algorithms are commonly used in solving complex and NP-hard optimization problems various fields. These have become popular because of their ability to explore exploit solutions problem domains. Honey Badger Algorithm (HBA) is a population-based metaheuristic algorithm inspired by the dynamic hunting strategy honey badgers, utilizing digging-seeking techniques. Since its introduction 2020, HBA has garnered widespread attention been applied across This review aims comprehensively survey improvement application problems. Additionally, conducts meta-analysis HBA's improvements, hybridization since introduction. According result survey, 52 studies presented improved using chaotic maps, levy flight mechanism, adaptive mechanisms, transfer functions, multi-objective mechanism opposition based learning techniques, 20 hybrid with other metaheuristics 101 uses original for wide acceptance within research community stems from straightforwardness, ease use, efficient computational time, accelerated convergence speed, high efficacy, capability address different kind issues, distinguishing it well-known approches presented.

Language: Английский

---

AI Driven False Data Injection Attack Recognition Approach for Cyber-Physical Systems in Smart Cities

Journal of Smart Internet of Things, Journal Year: 2023, Volume and Issue: 2023(2), P. 13 - 32

Published: Dec. 1, 2023

Abstract Cyber-physical systems (CPS) combine the typical power grid with recent communication and control technologies, generating new features for attacks. False data injection attacks (FDIA) contain maliciously injecting fabricated as to system measurements, capable of due improper decisions disruptions in distribution. Identifying these is vital preserving reliability integrity grid. Researchers this domain utilize modern approaches namely machine learning (ML) deep (DL) detecting anomalous forms that signify existence such By emerging accurate effective detection approaches, research purposes improve resilience CPS make sure a secure continuous supply consumers. This article presents an Improved Equilibrium Optimizer Deep Learning Enabled Data Injection Attack Recognition (IEODL-FDIAR) technique platform. The main purpose IEODL-FDIAR enable FDIA attack accomplishes security CPSS environment. In presented technique, IEO algorithm used feature subset selection process. Moreover, applies stacked autoencoder (SAE) model detection. Furthermore, pelican optimization (POA) can be utilized optimum hyperparameter chosen SAE which turn boosts outcomes model. To portray better outcome system, wide range simulation analyses are executed. A comparison analysis described improved results existing DL models.

Language: Английский

---

Cardio vascular disease prediction by deep learning based on IOMT: review

Smart Science, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 11

Published: Aug. 19, 2024

The global burden of disease caused by cardiovascular diseases (CVDs) is increasing despite technical advancements in healthcare because a dramatic rise the developing nations that are experiencing rapid health transitions. World Health Organization (WHO) estimates 17.9 million deaths worldwide 2021 and connected to CVDs, or 32% all deaths. Since ancient times, people have experimented with methods extend their lives. proposed technology still long way for attaining aim lessening mortality rates. Early detection proactive management CVD risk factors crucial reducing these diseases. In recent years, researchers been exploring potential deep learning predicting depending upon data collected from IoMT devices. Deep (DL) used prediction popular this domain. Several DL techniques implemented accomplish efficient prediction-based CVD. There several steps employing model. IoT sensors process large amounts patient-related biomedical data, enabling doctors closely monitor patients make choices real-time. An outline IoT, sensors, provided after discussion cardiac its existing treatments. A complete analysis current pertinent deep-learning heart reviewed. result shows performance metrics comparison different approaches. This review undertaken pulling 44 papers published between years 2020 2023, provides thorough statistical analysis. Finally, survey will be beneficial researchers.

Language: Английский

---

The Combinations of Fuzzy Membership Functions on Discretization in the Decision Tree-ID3 to Predict Degenerative Disease Status

Symmetry, Journal Year: 2024, Volume and Issue: 16(12), P. 1560 - 1560

Published: Nov. 21, 2024

Degenerative diseases are one of the leading causes chronic disability on a global scale, significantly affecting quality life sufferers. These also burden health care system and individuals financially. The implementation preventive strategies can be postponed until an accurate prediction disease status achieved. that cause death in many countries coronary heart (CHD), while diabetes mellitus (DMD) increases risk CHD. Most predictor variables from dataset to predict both continuous. However, not all methods, including Decision Tree Iterative Dichotomiser3 (DTID3) method, process continuous data. This work aims degenerative diseases, CHD DM, using DTID3 method with type transformed discretization concept set membership. Seven models proposed each disease. One model uses crisp membership, six use fuzzy membership (FDTID3). Each FDTID3 represents combination functions discretizing variables, consists three functions. performance depends used. hypothesis seven differs at least metric is higher than discretized sets has been proven.

Language: Английский

---