Federated Learning for Clinical Event Classification Using Vital Signs Data DOI Open Access
Kang Yoon Lee, Ruzaliev Rakhmiddin

Published: June 12, 2023

Effective healthcare relies on accurate and timely diagnosis; however, obtaining large amounts of training data while maintaining patient privacy remains challenging. This study introduces a novel approach utilizing federated learning (FL) cross-device multi-modal model for clin-ical event classification using vital signs data. Our architecture leverages FL to train machine models, including Random Forest, AdaBoost, SGD ensemble model, from diverse clientele at Boston hospital (MIMIC-IV dataset). The structure preserves by directly each client's device without transferring sensitive demonstrates the potential in privacy-preserving clinical classification, achieving an impressive accuracy 98.9%. These findings underscore significance technology applications, enabling analysis safeguarding privacy.

Language: Английский

A survey on deep learning tools dealing with data scarcity: definitions, challenges, solutions, tips, and applications DOI Creative Commons
Laith Alzubaidi, Jinshuai Bai, Aiman Al-Sabaawi

et al.

Journal Of Big Data, Journal Year: 2023, Volume and Issue: 10(1)

Published: April 14, 2023

Abstract Data scarcity is a major challenge when training deep learning (DL) models. DL demands large amount of data to achieve exceptional performance. Unfortunately, many applications have small or inadequate train frameworks. Usually, manual labeling needed provide labeled data, which typically involves human annotators with vast background knowledge. This annotation process costly, time-consuming, and error-prone. every framework fed by significant automatically learn representations. Ultimately, larger would generate better model its performance also application dependent. issue the main barrier for dismissing use DL. Having sufficient first step toward any successful trustworthy application. paper presents holistic survey on state-of-the-art techniques deal models overcome three challenges including small, imbalanced datasets, lack generalization. starts listing techniques. Next, types architectures are introduced. After that, solutions address listed, such as Transfer Learning (TL), Self-Supervised (SSL), Generative Adversarial Networks (GANs), Model Architecture (MA), Physics-Informed Neural Network (PINN), Deep Synthetic Minority Oversampling Technique (DeepSMOTE). Then, these were followed some related tips about acquisition prior purposes, well recommendations ensuring trustworthiness dataset. The ends list that suffer from scarcity, several alternatives proposed in order more each Electromagnetic Imaging (EMI), Civil Structural Health Monitoring, Medical imaging, Meteorology, Wireless Communications, Fluid Mechanics, Microelectromechanical system, Cybersecurity. To best authors’ knowledge, this review offers comprehensive overview strategies tackle

Language: Английский

Citations

366
Towards Risk-Free Trustworthy Artificial Intelligence: Significance and Requirements DOI Creative Commons
Laith Alzubaidi, Aiman Al-Sabaawi, Jinshuai Bai

et al.

International Journal of Intelligent Systems, Journal Year: 2023, Volume and Issue: 2023, P. 1 - 41

Published: Oct. 26, 2023

Given the tremendous potential and influence of artificial intelligence (AI) algorithmic decision-making (DM), these systems have found wide-ranging applications across diverse fields, including education, business, healthcare industries, government, justice sectors. While AI DM offer significant benefits, they also carry risk unfavourable outcomes for users society. As a result, ensuring safety, reliability, trustworthiness becomes crucial. This article aims to provide comprehensive review synergy between DM, focussing on importance trustworthiness. The addresses following four key questions, guiding readers towards deeper understanding this topic: (i) why do we need trustworthy AI? (ii) what are requirements In line with second question, that establish been explained, explainability, accountability, robustness, fairness, acceptance AI, privacy, accuracy, reproducibility, human agency, oversight. (iii) how can data? (iv) priorities in terms challenging applications? Regarding last six different discussed, environmental science, 5G-based IoT networks, robotics architecture, engineering construction, financial technology, healthcare. emphasises address before their deployment order achieve goal good. An example is provided demonstrates be employed eliminate bias resources management systems. insights recommendations presented paper will serve as valuable guide researchers seeking applications.

Language: Английский

Citations

40
A quantum convolutional network and ResNet (50)-based classification architecture for the MNIST medical dataset DOI

Esraa Hassan,

M. Shamim Hossain, Abeer Saber

et al.

Biomedical Signal Processing and Control, Journal Year: 2023, Volume and Issue: 87, P. 105560 - 105560

Published: Oct. 7, 2023

Language: Английский

Citations

37
Deep study on autonomous learning techniques for complex pattern recognition in interconnected information systems DOI
Zahra Mohtasham‐Amiri, Arash Heidari,

Nima Jafari

et al.

Computer Science Review, Journal Year: 2024, Volume and Issue: 54, P. 100666 - 100666

Published: Sept. 20, 2024

Language: Английский

Citations

12
MEFF – A model ensemble feature fusion approach for tackling adversarial attacks in medical imaging DOI Creative Commons
Laith Alzubaidi, Khamael Al-Dulaimi,

Huda Abdul-Hussain Obeed

et al.

Intelligent Systems with Applications, Journal Year: 2024, Volume and Issue: 22, P. 200355 - 200355

Published: March 16, 2024

Adversarial attacks pose a significant threat to deep learning models, specifically medical images, as they can mislead models into making inaccurate predictions by introducing subtle distortions the input data that are often imperceptible humans. Although adversarial training is common technique used mitigate these on it lacks flexibility address new attack methods and effectively improve feature representation. This paper introduces novel Model Ensemble Feature Fusion (MEFF) designed combat in image applications. The proposed model employs fusion combining features extracted from different DL then trains Machine Learning classifiers using fused features. It uses concatenation method merge features, forming more comprehensive representation enhancing model's ability classify classes accurately. Our experimental study has performed evaluation of MEFF, considering several challenging scenarios, including 2D 3D greyscale colour binary classification, multi-label classification. reported results demonstrate robustness MEFF against types across six distinct A key advantage its capability incorporate wide range without need train scratch. Therefore, contributes developing diverse robust defense strategy. More importantly, leveraging ensemble modeling, enhances resilience face attacks, paving way for improved reliability analysis.

Language: Английский

Citations

11
Synchronizing Object Detection: Applications, Advancements and Existing Challenges DOI Creative Commons
Md. Tanzib Hosain,

Asif Zaman,

Mushfiqur Rahman Abir

et al.

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 54129 - 54167

Published: Jan. 1, 2024

From pivotal roles in autonomous vehicles, healthcare diagnostics, and surveillance systems to seamlessly integrating with augmented reality, object detection algorithms stand as the cornerstone unraveling complexities of visual world. Tracing trajectory from conventional region-based methods latest neural network architectures reveals a technological renaissance where metamorphose into digital artisans. However, this journey is not without hurdles, prompting researchers grapple real-time detection, robustness varied environments, interpretability amidst intricacies deep learning. The allure addressing issues such occlusions, scale variations, fine-grained categorization propels exploration uncharted territories, beckoning scholarly community contribute an ongoing saga innovation discovery. This research offers comprehensive panorama, encapsulating applications reshaping our advancements pushing boundaries perception, open extending invitation next generation visionaries explore frontiers within detection.

Language: Английский

Citations

10
Application of big data technology in enterprise information security management DOI Creative Commons
Ping Li, Limin Zhang

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 6, 2025

This study aims to explore the application value of big data technology (BDT) in enterprise information security (EIS). Its goal is develop a risk prediction model based on analysis enhance protection capability enterprises. A system that can monitor and intelligently identify potential risks real-time constructed by designing complex network algorithms machine learning models. For different types threats, uses feature engineering training processes extract key indicators optimize performance. The experimental results show has excellent performance test set, its Area Under Curve reaches 0.95, indicating good differentiation ability high accuracy. In addition, multi-class identification task, achieves an average precision 0.87. Compared with traditional method, it remarkably improved early warning accuracy response speed enterprises various incidents. Therefore, this confirms effectiveness feasibility applying BDT EIS management, successfully provides strong technical support for protection.

Language: Английский

Citations

1
Fuzzy Decision‐Making Framework for Evaluating Hybrid Detection Models of Trauma Patients DOI Open Access
Rula A. Hamid, Idrees A. Zahid, A. S. Albahri

et al.

Expert Systems, Journal Year: 2025, Volume and Issue: 42(3)

Published: Feb. 13, 2025

ABSTRACT This study introduces a new multi‐criteria decision‐making (MCDM) framework to evaluate trauma injury detection models in intensive care units (ICUs). research addresses the challenges associated with diverse machine learning (ML) models, inconsistencies, conflicting priorities, and importance of metrics. The developed methodology consists three phases: dataset identification pre‐processing, hybrid model development, an evaluation/benchmarking framework. Through meticulous is tailored focus on adult patients. Forty were by combining eight ML algorithms four filter‐based feature‐selection methods principal component analysis (PCA) as dimensionality reduction method, these evaluated using seven weight coefficients for metrics are determined 2‐tuple Linguistic Fermatean Fuzzy‐Weighted Zero‐Inconsistency (2TLF‐FWZIC) method. Vlsekriterijumska Optimizcija I Kompromisno Resenje (VIKOR) approach applied rank models. According 2TLF‐FWZIC, classification accuracy (CA) precision obtained highest weights 0.2439 0.1805, respectively, while F1, training time, test time lowest 0.1055, 0.0886, 0.1111, respectively. benchmarking results revealed following top‐performing models: Gini index logistic regression (GI‐LR), decision tree (GI_DT), information gain (IG_DT), VIKOR Q score values 0.016435, 0.023804, 0.042077, proposed MCDM assessed examined systematic ranking, sensitivity analysis, validation best‐selected two unseen datasets, mode explainability SHapley Additive exPlanations (SHAP) We benchmarked against other benchmark studies achieved 100% across six key areas. provides several insights into empirical synthesis this study. It contributes advancing medical informatics enhancing understanding selection ICUs.

Language: Английский

Citations

1
Machine Learning for Evaluating Hospital Mobility: An Italian Case Study DOI Creative Commons
Vito Santamato, Caterina Tricase, Nicola Faccilongo

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(14), P. 6016 - 6016

Published: July 10, 2024

This study delves into hospital mobility within the Italian regions of Apulia and Emilia-Romagna, interpreting it as an indicator perceived service quality. Utilizing logistic regression alongside other machine learning techniques, we analyze impact structural, operational, clinical variables on patient perceptions quality, thus influencing their healthcare choices. The analysis trends has uncovered significant regional differences, emphasizing how context shapes To further enhance analysis, SHAP (SHapley Additive exPlanations) values have been integrated model. These quantify specific contributions each variable to quality service, significantly improving interpretability fairness evaluations. A methodological innovation this is use these scores weights in data envelopment (DEA), facilitating a comparative efficiency facilities that both weighted normative. combination SHAP-weighted DEA provides deeper understanding dynamics offers essential insights for optimizing distribution resources. approach underscores importance data-driven strategies develop more equitable, efficient, patient-centered systems. research contributes promotes investigations accessibility leveraging tool increase services across diverse settings. findings are pivotal policymakers system managers aiming reduce disparities promote responsive personalized service.

Language: Английский

Citations

4
USSC-YOLO: Enhanced Multi-Scale Road Crack Object Detection Algorithm for UAV Image DOI Creative Commons

Yanxiang Zhang,

Yao Lu,

Zijian Huo

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(17), P. 5586 - 5586

Published: Aug. 28, 2024

Road crack detection is of paramount importance for ensuring vehicular traffic safety, and implementing traditional methods cracks inevitably impedes the optimal functioning traffic. In light above, we propose a USSC-YOLO-based target algorithm unmanned aerial vehicle (UAV) road based on machine vision. The aims to achieve high-precision at all scale levels. Compared with original YOLOv5s, main improvements USSC-YOLO are ShuffleNet V2 block, coordinate attention (CA) mechanism, Swin Transformer. First, address problem large network computational spending, replace backbone YOLOv5s blocks, reducing overhead significantly. Next, reduce problems caused by complex background interference, introduce CA mechanism into network, which reduces missed false rate. Finally, integrate Transformer block end neck enhance accuracy small cracks. Experimental results our self-constructed UAV near-far scene i(UNFSRCI) dataset demonstrate that model giga floating-point operations per second (GFLOPs) compared while achieving 6.3% increase in mAP@50 12% improvement mAP@ [50:95]. This indicates remains lightweight meanwhile providing excellent performance. future work, will assess safety conditions these prioritize maintenance sequences targets facilitate further intelligent management.

Language: Английский

Citations

4