Federated Learning for the Healthcare Metaverse: Concepts, Applications, Challenges, and Future Directions

IEEE Internet of Things Journal, Journal Year: 2023, Volume and Issue: 10(24), P. 21873 - 21891

Published: Aug. 14, 2023

Recent technological advancements have considerably improved healthcare systems to provide various intelligent services, improving life quality. The Metaverse, often described as the next evolution of Internet, helps users interact with each other and environment, thus offering a seamless connection between virtual physical worlds. Additionally, by integrating emerging technologies, such artificial intelligence (AI), cloud edge computing, Internet Things (IoT), blockchain, semantic communications, can potentially transform many vertical domains in general sector (healthcare Metaverse) particular. Metaverse holds huge potential revolutionize development systems, presenting new opportunities for significant delivery, personalized experiences, medical education, collaborative research, so on. However, challenges are associated realization privacy, interoperability, data management, security. Federated learning (FL), branch AI, opens up enormous deal aforementioned exploiting computing resources available at distributed devices. This motivated us present survey on adopting FL Metaverse. Initially, we preliminaries IoT-based conventional healthcare, Furthermore, benefits discussed. Subsequently, discuss several applications FL-enabled including diagnosis, patient monitoring, infectious disease, drug discovery. Finally, highlight solutions toward realizing

Language: Английский

---

Mitigating Poor Data Quality Impact with Federated Unlearning for Human-Centric Metaverse

IEEE Journal on Selected Areas in Communications, Journal Year: 2023, Volume and Issue: 42(4), P. 832 - 849

Published: Dec. 21, 2023

Federated Learning (FL), which has been employed to train machine learning models on the data with a distributed manner, could enhance immersive user experience for human-centric metaverse. However, it's challenging accurately and promptly FL metaverse due massive communication unreliability. User be negatively affected by using low-quality metaverse, e.g., it cannot scrutinize arrive at decisions timely. To resolve this pressing issue, we propose MetaFul federated unlearning solution reduces negative influences of no transmission removing training server side. specific, includes three main components. (i) Low-throughput (LT-FL) addresses issue large model in decreasing dimension number transmitted parameters. (ii) Loss-based quality assessment (LM-QA) utilizes loss generated LT-FL estimate quality. (iii) Non-communicative (NC-FUL) revokes impact careful designed Both LM-QA NC-FUL have communications clients. Finally, extensive evaluations are conducted show improve accuracy least 2.5% decrease perception time 19.3% compared benchmarks.

Language: Английский

---

Artificial Intelligence-Enabled Metaverse for Sustainable Smart Cities: Technologies, Applications, Challenges, and Future Directions

Electronics, Journal Year: 2024, Volume and Issue: 13(24), P. 4874 - 4874

Published: Dec. 10, 2024

Rapid urbanisation has intensified the need for sustainable solutions to address challenges in urban infrastructure, climate change, and resource constraints. This study reveals that Artificial Intelligence (AI)-enabled metaverse offers transformative potential developing smart cities. AI techniques, such as machine learning, deep generative (GAI), large language models (LLMs), enhance metaverse’s capabilities data analysis, decision making, personalised user experiences. The further examines how these advanced facilitate key technologies big analytics, natural processing (NLP), computer vision, digital twins, Internet of Things (IoT), Edge AI, 5G/6G networks. Applications across various city domains—environment, mobility, energy, health, governance, economy, real-world use cases virtual cities like Singapore, Seoul, Lisbon are presented, demonstrating AI’s effectiveness However, AI-enabled presents related acquisition management, privacy, security, interoperability, scalability, ethical considerations. These challenges’ societal technological implications discussed, highlighting robust governance frameworks ethics guidelines. Future directions emphasise advancing model architectures algorithms, enhancing privacy security measures, promoting practices, addressing performance fostering stakeholder collaboration. By challenges, full can be harnessed sustainability, adaptability, livability

Language: Английский

---

Metaverse for Healthcare: Technologies, Challenges, and Vision

International Journal of Crowd Science, Journal Year: 2023, Volume and Issue: 7(4), P. 190 - 199

Published: Dec. 1, 2023

The continuous enhancement of living conditions imposes higher requirements for medical and healthcare services. Although improved to a certain extent, there still exist critical challenges in current pattern, such as the shortage resources, inefficient treatment, limited technology level. metaverse can offer novel mechanism address these problems traditional domain, thus, enhance quality Generally, is dynamic feedback system that facilitates collaboration coexistence between virtual physical worlds. By fostering evolution intelligent agents world, knowledge this interdependence be reconstructed digital realm. This allows existed real world abstracted represented space, where models established computational experiments conducted. outcomes obtained dynamically guide or control execution strategies with real-world results serving data inputs continually update world's model. In addition, paper summarizes research status different application scenarios metaverse, highlights vision, aims inspire further field.

Language: Английский

---

Toward Diagnosis of Diseases Using Emerging Technologies: A Comprehensive Survey of the State of the Art in Metaverse

International Journal of Intelligent Systems, Journal Year: 2025, Volume and Issue: 2025(1)

Published: Jan. 1, 2025

Introduction: The Metaverse, a rapidly growing technology in healthcare, is proving to be game‐changer early disease detection and diagnosis. This study aimed identify the latest scientific achievements such as its effects, associated technologies, obstacles for diagnosing diseases. Methods: In this review study, databases, including PubMed Web of Science, were searched using related keywords. Related studies about Metaverse diagnosis included according inclusion exclusion criteria. Data extraction was done data form. findings summarized reported tables figures objectives. Results: From 1706 retrieved articles, 28 Most conducted 2023 (13 out 28). 13 groups specialists used diagnose diseases; oncologists neurologists it more than others. most important technological aspects six main categories, computer vision, artificial intelligence, virtual reality, blockchain, digital twin, cloud computing. Metaverse’s effects diagnostic interventions 22 subcategories five improving diagnosis, facilitating interactions, education, better future, uncertainty. role particularly significant. challenges seven subcategories: studies, financial limitations, issues, structural legal ethical acceptance, nature Metaverse. Conclusion: Given pivotal accurate patients treatment plans, potential complex challenging diagnoses However, note that can only fully realized through further research on utilizing specifically call additional not just suggestion but necessity future healthcare.

Language: Английский

---

A progressive growing generative adversarial network composed of enhanced style-consistent modulation for fetal ultrasound four-chamber view editing synthesis

Engineering Applications of Artificial Intelligence, Journal Year: 2024, Volume and Issue: 133, P. 108438 - 108438

Published: April 13, 2024

Language: Английский

---

HCMMNet: Hierarchical Conv-MLP-Mixed Network for Medical Image Segmentation in Metaverse for Consumer Health

IEEE Transactions on Consumer Electronics, Journal Year: 2023, Volume and Issue: 70(1), P. 2078 - 2089

Published: Nov. 28, 2023

In the burgeoning metaverse for consumer health (MCH), medical image segmentation methods with high accuracy and generalization capability are essential to drive personalized healthcare solutions enhance patient experience. To address inherent challenges of capturing complex structures features in segmentation, we propose a convolutional neural network (CNN) multi-layer-perceptron (MLP) mixed module named HCMM, which hierarchically incorporates local priors CNN into fully-connected (FC) layers, ingeniously specific details broader range contextual information focused object from diverse perspectives. Then, an MLP-based fusion (MIF) designed dynamically merge feature maps varying levels different pathways, enhancing expression discriminative power. Based on above-proposed modules, design novel model, HCMMNet, can adeptly capture input images at scales Through comparative experiments, demonstrate outstanding performance HCMMNet three publicly available datasets one self-organized dataset. Notably, our showcases remarkable efficacy while maintaining extraordinarily lightweight profile, weighing mere 3M, rendering it ideal MCH application.

Language: Английский

---

SKGC: A General Semantic-level Knowledge Guided Classification Framework for Fetal Congenital Heart Disease

IEEE Journal of Biomedical and Health Informatics, Journal Year: 2024, Volume and Issue: 28(10), P. 6105 - 6116

Published: July 10, 2024

Congenital heart disease (CHD) is the most common congenital disability affecting healthy development and growth, even resulting in pregnancy termination or fetal death. Recently, deep learning techniques have made remarkable progress to assist diagnosing CHD. One very popular method directly classifying ultrasound images, recognized as abnormal normal, which tends focus more on global features neglects semantic knowledge of anatomical structures. The other approach segmentation-based diagnosis, requires a large number pixel-level annotation masks for training. However, detailed segmentation costly unavailable. Based above analysis, we propose SKGC, universal framework identify normal four-chamber (4CH) guided by few masks, while improving accuracy remarkably. SKGC consists semantic-level extraction module (SKEM), multi-knowledge fusion (MFM), classification (CM). SKEM responsible obtaining high-level knowledge, serving an abstract representation structures that obstetricians on. MFM lightweight but efficient fuses with original specific images. CM classifies fused can be replaced any advanced classifier. Moreover, design new loss function enhances constraint between foreground background predictions, quality knowledge. Experimental results collected real-world NA-4CH publicly FEST datasets show achieves impressive performance best 99.68% 95.40%, respectively. Notably, improves from 74.68% 88.14% using only 10 labeled masks.

Language: Английский

---

Grid-Metaverse: The Path From Digital Twins and Prototype Tests on DC Microgrids

Published: June 1, 2023

With the development of cutting-edge technologies and efforts business giants, Metaverse is becoming increasingly reachable. In addition to fields healthcare, education cultural tourism, will also have a profound impact on power grid. The digital twin (DT) regarded as foundation Metaverse, but its high cost hinders broad application DTs in grids. this paper, we propose path build direct current (DC) microgrid, which representative block future grid with penetration renewable energy sources, forming Grid-Metaverse that can significantly reduce improve interactivity. A model-based DT, originated from physical model, data-driven using graph neural network data one, are built together illustrate show their roles Grid-Metaverse. Moreover, prototype tests, threats denial-of-service (DoS) false injection (FDI) attacks validated through developed DTs.

Language: Английский

---

Medical Metaverse: A New Virtual Health Experience

Published: Dec. 4, 2023

Ahstract- This work examines the impact of medical metaverse on traditional medicine and integration various technologies that have greatly enhanced its capabilities. Technologies such as artificial intelligence (AI), blockchain, Internet Things (IoT), augmented reality (AR), virtual (VR), 5G, big data, natural language processing, digital twins facilitated healthcare delivery, allowing clinicians to diagnose patients regardless distance receive real-time data. The these has also improved outcomes created new experiences. paper highlights how adoption by enhancing patient experiences outcomes, especially in disease management specialized care settings. However, use is not without challenges, this offers solutions address them. Despite potential issues, a way delivering services. Further development optimization are necessary realize full metaverse, but it promising exciting area worth exploring.

Language: Английский

---