Smart Technologies used as Smart Tools in the Management of Cardiovascular Disease and their Future Perspective

Current Problems in Cardiology, Journal Year: 2023, Volume and Issue: 48(11), P. 101922 - 101922

Published: July 10, 2023

Language: Английский

---

An overview of methods and techniques in multimodal data fusion with application to healthcare

International Journal of Data Science and Analytics, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 10, 2025

Language: Английский

---

A survey of multimodal hybrid deep learning for computer vision: Architectures, applications, trends, and challenges

Information Fusion, Journal Year: 2023, Volume and Issue: 105, P. 102217 - 102217

Published: Dec. 30, 2023

Language: Английский

---

Artificial intelligence-assisted dermatology diagnosis: From unimodal to multimodal

Computers in Biology and Medicine, Journal Year: 2023, Volume and Issue: 165, P. 107413 - 107413

Published: Sept. 1, 2023

Artificial Intelligence (AI) is progressively permeating medicine, notably in the realm of assisted diagnosis. However, traditional unimodal AI models, reliant on large volumes accurately labeled data and single type usage, prove insufficient to assist dermatological Augmenting these models with text from patient narratives, laboratory reports, image skin lesions, dermoscopy, pathologies could significantly enhance their diagnostic capacity. Large-scale pre-training multimodal offer a promising solution, exploiting burgeoning reservoir clinical amalgamating various types. This paper delves into models' methodologies, applications, shortcomings while exploring how can accuracy reliability. Furthermore, integrating cutting-edge technologies like federated learning multi-party privacy computing substantially mitigate concerns datasets further fosters move towards high-precision self-diagnosis. Diagnostic systems underpinned by large-scale facilitate dermatology physicians formulating effective treatment strategies herald transformative era healthcare.

Language: Английский

---

Predictive modelling and identification of key risk factors for stroke using machine learning

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

Published: May 20, 2024

Strokes are a leading global cause of mortality, underscoring the need for early detection and prevention strategies. However, addressing hidden risk factors achieving accurate prediction become particularly challenging in presence imbalanced missing data. This study encompasses three imputation techniques to deal with To tackle data imbalance, it employs synthetic minority oversampling technique (SMOTE). The initiates baseline model subsequently an extensive range advanced models. thoroughly evaluates performance these models by employing k-fold cross-validation on various balanced datasets. findings reveal that age, body mass index (BMI), average glucose level, heart disease, hypertension, marital status most influential features predicting strokes. Furthermore, Dense Stacking Ensemble (DSE) is built upon previous after fine-tuning, best-performing as meta-classifier. DSE demonstrated over 96% accuracy across diverse datasets, AUC score 83.94% imputed dataset 98.92% one. research underscores remarkable model, compared same dataset. It highlights model's potential stroke improve patient outcomes.

Language: Английский

---

Diagnostic Strategies Using AI and ML in Cardiovascular Diseases: Challenges and Future Perspectives

Algorithms for intelligent systems, Journal Year: 2025, Volume and Issue: unknown, P. 135 - 165

Published: Jan. 1, 2025

Language: Английский

---

Radiology's Ionising Radiation Paradox: Weighing the Indispensable Against the Detrimental in Medical Imaging

Cureus, Journal Year: 2023, Volume and Issue: unknown

Published: July 10, 2023

Ionising radiation stands as an indispensable protagonist in the narrative of medical imaging, underpinning diagnostic evaluations and therapeutic interventions across array conditions. However, this poses a paradox - its inestimable service to medicine coexists with undercurrent potential health risks, primarily DNA damage subsequent oncogenesis. The comprehensive review unfurls around intricate enigma, delicately balancing utility against non-negotiable commitment patient safety. In critical discourse, intricacies ionising are dissected, illuminating not only sources but also associated biological hazards. exploration delves into labyrinth strategies currently deployed minimise exposure safeguard patients. By casting light on scientific nuances X-rays, computed tomography (CT), nuclear medicine, it traverses complex terrain use radiology, promote safer imaging practices, facilitate ongoing dialogue about necessity risk. Through rigorous examination, pivotal relationship between dose response is elucidated, unravelling mechanisms injury distinguishing deterministic stochastic effects. Moreover, protection illuminated, demystifying concepts such justification, optimisation, As Low Reasonably Achievable (ALARA) principle, reference levels, along administrative regulatory approaches. With eye horizon, promising avenues future research discussed. These encompass low-radiation techniques, long-term risk assessment large cohorts, transformative artificial intelligence optimisation. This nuanced complexities radiology aims foster collaborative impetus towards practices. It underscores need for risk, thereby advocating continual reassessment imaging.

Language: Английский

---

CardioRiskNet: A Hybrid AI-Based Model for Explainable Risk Prediction and Prognosis in Cardiovascular Disease

Bioengineering, Journal Year: 2024, Volume and Issue: 11(8), P. 822 - 822

Published: Aug. 12, 2024

The global prevalence of cardiovascular diseases (CVDs) as a leading cause death highlights the imperative need for refined risk assessment and prognostication methods. traditional approaches, including Framingham Risk Score, blood tests, imaging techniques, clinical assessments, although widely utilized, are hindered by limitations such lack precision, reliance on static variables, inability to adapt new patient data, thereby necessitating exploration alternative strategies. In response, this study introduces CardioRiskNet, hybrid AI-based model designed transcend these limitations. proposed CardioRiskNet consists seven parts: data preprocessing, feature selection encoding, eXplainable AI (XAI) integration, active learning, attention mechanisms, prediction prognosis, evaluation validation, deployment integration. At first, preprocessed cleaning handling missing values, applying normalization process, extracting features. Next, most informative features selected categorical variables converted into numerical form. Distinctively, employs learning iteratively select samples, enhancing its efficacy, while mechanism dynamically focuses relevant precise prediction. Additionally, integration XAI facilitates interpretability transparency in decision-making processes. According experimental results, demonstrates superior performance terms accuracy, sensitivity, specificity, F1-Score, with values 98.7%, 99%, respectively. These findings show that can accurately assess prognosticate CVD risk, demonstrating power surpass conventional Thus, CardioRiskNet's novel approach high advance management CVDs provide healthcare professionals powerful tool care.

Language: Английский

---

A comprehensive patient-specific prediction model for temporomandibular joint osteoarthritis progression

Proceedings of the National Academy of Sciences, Journal Year: 2024, Volume and Issue: 121(8)

Published: Feb. 12, 2024

Temporomandibular joint osteoarthritis (TMJ OA) is a prevalent degenerative disease characterized by chronic pain and impaired jaw function. The complexity of TMJ OA has hindered the development prognostic tools, posing significant challenge in timely, patient-specific management. Addressing this gap, our research employs comprehensive, multidimensional approach to advance prognostication. We conducted prospective study with 106 subjects, 74 whom were followed up after 2 3 y conservative treatment. Central methodology an innovative, open-source predictive modeling framework, Ensemble via Hierarchical Predictions through Nested cross-validation tool (EHPN). This framework synergistically integrates 18 feature selection, statistical, machine learning methods yield accuracy 0.87, area under ROC curve 0.72 F1 score 0.82. Our study, beyond technical advancements, emphasizes global impact OA, recognizing its unique demographic occurrence. highlight key factors influencing progression. Using SHAP analysis, we identified personalized predictors: lower values headache, back pain, restless sleep, condyle high gray level-GL-run emphasis, articular fossa GL nonuniformity, long-run low emphasis; higher superior space, mouth opening, saliva Vascular-endothelium-growth-factor, Matrix-metalloproteinase-7, serum Epithelial-neutrophil-activating-peptide, age indicate recovery likelihood. multimodal EHPN enhances clinicians' decision-making, offering transformative translational infrastructure. model stands as contribution precision medicine, paradigm shift management temporomandibular disorders potentially broader applications healthcare.

Language: Английский

---

Advancing healthcare through multimodal data fusion: a comprehensive review of techniques and applications

PeerJ Computer Science, Journal Year: 2024, Volume and Issue: 10, P. e2298 - e2298

Published: Oct. 30, 2024

With the increasing availability of diverse healthcare data sources, such as medical images and electronic health records, there is a growing need to effectively integrate fuse this multimodal for comprehensive analysis decision-making. However, despite its potential, fusion in remains limited. This review paper provides an overview existing literature on healthcare, covering 69 relevant works published between 2018 2024. It focuses methodologies that different types enhance analysis, including techniques integrating with structured unstructured data, combining multiple image modalities, other features. Additionally, reviews various approaches fusion, early, intermediate, late methods, examines challenges limitations associated these techniques. The potential benefits applications diseases are highlighted, illustrating specific strategies employed artificial intelligence (AI) model development. research synthesizes information facilitate progress using improved diagnosis treatment planning.

Language: Английский

---