The slicing based spreading analysis for melanoma prediction using reinforcement learning model DOI

Venkata Ashok K Gorantla,

Shiva Kumar Sriramulugari,

Amit Hasmukhbhai Mewada

et al.

Published: Dec. 14, 2023

the present study proposes a novel approach to skin lesion prediction, namely, slicing-based spreading analysis (SBSA) with reinforcement learning (RL) model. The aim of SBSA approach, as implemented in this study, is mine and capture key aspects data from different perspectives for more accurate classification. We additionally introduce RL models enhanced performance classification tasks. Specifically, our based on five phases: obtaining complete data, slicing collected repeating promotional process, training slices RL, finally, combining trained predicting type. A benchmark dataset 400 dermoscopic pictures was used test suggested melanoma identification. accuracy attained compared traditional like support vector machines (SVM), random forests (RF), multilayer perceptions (MLP) utilizing methodology. Results indicated that achieved better metrics than classic machine approaches. Furthermore, proposed models, an overall 94.56%, significantly outperforming other models. In conclusion, provides promising type prediction.

Language: Английский

Emerging Research Trends in Artificial Intelligence for Cancer Diagnostic Systems: A Comprehensive Review DOI Creative Commons
Sagheer Abbas,

Muhammad Waqas Asif,

Abdur Rehman

et al.

Heliyon, Journal Year: 2024, Volume and Issue: 10(17), P. e36743 - e36743

Published: Aug. 23, 2024

This review article offers a comprehensive analysis of current developments in the application machine learning for cancer diagnostic systems. The effectiveness approaches has become evident improving accuracy and speed detection, addressing complexities large intricate medical datasets. aims to evaluate modern techniques employed diagnostics, covering various algorithms, including supervised unsupervised learning, as well deep federated methodologies. Data acquisition preprocessing methods different types data, such imaging, genomics, clinical records, are discussed. paper also examines feature extraction selection specific diagnosis. Model training, evaluation metrics, performance comparison explored. Additionally, provides insights into applications discusses challenges related dataset limitations, model interpretability, multi-omics integration, ethical considerations. emerging field explainable artificial intelligence (XAI) diagnosis is highlighted, emphasizing XAI proposed improve diagnostics. These include interactive visualization decisions importance tailored enhanced interpretation, aiming enhance both transparency decision-making. concludes by outlining future directions, personalized medicine, advancements, guide researchers, clinicians, policymakers development efficient interpretable learning-based

Language: Английский

Citations

7
Emerging Technologies for Enhancing Robust Cybersecurity Measures for Business Intelligence in Healthcare 5.0 DOI

Abdur Rehman Sakhawat,

Areej Fatima, Sagheer Abbas

et al.

Advances in business information systems and analytics book series, Journal Year: 2024, Volume and Issue: unknown, P. 270 - 293

Published: Feb. 14, 2024

Healthcare 5.0 represents the next phase in healthcare evolution. It aims to harness creativity and expertise of professionals, integrating them with efficient, intelligent, precise technologies. This integration allows for resource-efficient patient-centered approaches, surpassing previous paradigms healthcare. To provide a comprehensive introduction 5.0, this chapter presents survey-based tutorial covering potential applications enabling technologies within domain. The takes approach introducing key concepts definitions 5.0. From perspective practitioners researchers, it explores that offers. Finally, several research challenges open issues require further development overcoming. These include effectively utilizing Business Intelligence as well implementing robust cybersecurity measures safeguard sensitive information.

Language: Английский

Citations

4
Information Modeling Technique to Decipher Research Trends of Federated Learning in Healthcare DOI Open Access

Rishu,

Vinay Kukreja, Shanmugasundaram Hariharan

et al.

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

Published: March 12, 2025

Aim The aim of this study is to determine the most prevalent types federated learning, discuss their uses in healthcare, highlight significant issues, and suggest methods for further research. Context When it comes handling distributed data, learning revolutionary, especially sensitive sectors like healthcare. In order improve outcomes growing number healthcare studies, there must be a method safely effectively analyze use enormous data. Objective purpose research large corpus 6,800 studies published between 2000 2024 apply topic modeling using Latent Semantic Analysis (LSA). Methods was analyzed LSA with goal identifying latent themes that capture spirit industry. provide an organized overview subject matter, five-topic solution devised. To guarantee relevance clarity, topics' coherence assessed. Results term frequency inverse document high-loading terms provided five major solutions. score achieved, i.e ., 0.789, indicating high level integration among identified topics. Different (FL), applications FL, key challenges possible associated FL have been analyzed. Conclusion This highlights significance privacy-preserving data analysis field, which may lead development creative solutions complex problems.

Language: Английский

Citations

0
A brain tumour classification on the magnetic resonance images using convolutional neural network based privacy‐preserving federated learning DOI

Şevket Ay,

Ekin Ekıncı, Zeynep Garip

et al.

International Journal of Imaging Systems and Technology, Journal Year: 2024, Volume and Issue: 34(1)

Published: Jan. 1, 2024

Abstract The healthcare industry has found it challenging to build a powerful global classification model due the scarcity and diversity of medical data. leading cause is privacy, which restricts data sharing among providers. Federated learning (FL) can contribute developing models by protecting privacy. This study tested various federated techniques in peer‐to‐peer setting classify brain Magnetic Resonance Images (MRI). authors propose aggregation strategies for FL, including Averaging (FedAvg), Quantum FL with FedAVG (QFedAvg) Fault Tolerant FedAvg (Ft‐FedAvg) differential privacy (Dp‐FedAvg). In each approach, custom Convolutional Neural Network (CNN) applied compute locally run nodes different parts same MRI dataset 10, 20 30 training test rounds. A central server CNN‐based three clients are included FL‐based tumour exchange combine weights on server, sent from local devices server. superiority performance proposed demonstrated comparing traditional methods metrics. Experimental results show that using approaches, FedAVg showed best 85.55% 84.60% success 10 rounds, respectively, while Ft‐FedAvg 85.80% rounds set. Statistical obtained approaches have superior regard accuracy robustness comparison others.

Language: Английский

Citations

3
A survey of recent advances in analysis of skin images DOI
Pragya Gupta, Jagannath Nirmal, Ninad Mehendale

et al.

Evolutionary Intelligence, Journal Year: 2024, Volume and Issue: 17(5-6), P. 4155 - 4178

Published: Aug. 25, 2024

Language: Английский

Citations

2
Federated learning for solar energy applications: A case study on real-time fault detection DOI
Ibtihal Ait Abdelmoula, Hicham Oufettoul,

Nassim Lamrini

et al.

Solar Energy, Journal Year: 2024, Volume and Issue: 282, P. 112942 - 112942

Published: Sept. 21, 2024

Language: Английский

Citations

2
Federated Learning in Real-Time Medical IoT: Optimizing Privacy and Accuracy for Chronic Disease Monitoring DOI Creative Commons

Et al. Dhairyashil Patil

Deleted Journal, Journal Year: 2024, Volume and Issue: 19(3), P. 32 - 42

Published: Jan. 25, 2024

The rising occurrence of long-term illnesses requires inventive and effective healthcare solutions, the incorporation Internet Things (IoT) technologies holds significant potential in revolutionizing conventional medical monitoring. This study presents an innovative method called Adaptive Federated Learning for Chronic Disease Prediction (AFL-CDP), which is specifically designed real-time applications. main objective to enhance both privacy accuracy surveillance chronic diseases. AFL-CDP utilizes federated learning, a decentralized approach machine learning that allows model training on multiple edge devices without need transfer raw data central server. not only mitigates concerns related sensitive but also improves precision predictive models by assimilating information from various sources. adaptability enables ongoing improvement using changing patient data, resulting personalized timely forecasts In order improve IoT with limited resources, integrates utilization SPECK, advanced technique preserving privacy. SPECK secure aggregation encryption mechanisms safeguard throughout process, guaranteeing confidentiality while integrity model. Ensuring security are utmost importance, particularly field IoT. proposed methodology assessed dataset consists purpose monitoring model's performance evaluated Area Under Curve (AUC) metric, achieves impressive AUC 94.37%. showcases efficacy framework capturing fundamental patterns varied data. To summarize, this strong applications, highlighting significance combination offers thorough satisfies strict demands high level precision, establishing basis enhanced results interventions.

Language: Английский

Citations

1
Unveiling the Spectrum of UV-Induced DNA Damage in Melanoma: Insights from AI-Based Analysis of Environmental Factors, Repair Mechanisms, and Skin Pigment Interactions DOI Creative Commons
Maram Fahaad Almufareh

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 64837 - 64860

Published: Jan. 1, 2024

Language: Английский

Citations

1
Novel paradigm of therapeutic intervention for skin cancer: challenges and opportunities DOI Creative Commons

Modassir Nasim,

Mariya Khan,

Rabea Parveen

et al.

Future Journal of Pharmaceutical Sciences, Journal Year: 2024, Volume and Issue: 10(1)

Published: Sept. 1, 2024

Abstract Background Skin cancer continues to be an imperative global health issue, urging continuous exploration of treatment methodologies. Conventional treatments for skin include surgical interventions, immunotherapy, targeted therapy, chemotherapy, and radiation therapy. However, these methods often present obstacles like resistance, systemic toxicity, limited effectiveness in advanced stages, infection risk, pain, long recovery, impact on healthy tissue. Main body the abstract Nanomedicine holds promise by facilitating precise drug administration, early detection, heightened therapeutic efficiency via localized delivery systems. The integration nanomedicine into alleviation therapies demonstrates optimistic outcomes, including refined delivery, augmented bioavailability, minimized adverse effects, potential theranostic applications. Recent breakthroughs have propelled advancements treatment, showing significant transforming paradigm. presents review provides comprehensive aspects existing their challenges, spotlighting recent nanomedicine. Short conclusion This delineates landscape treatments, underscores constraints, highlights strides that transform paradigm ultimately elevating patient prognosis. Importantly, emphasizes substantial challenges hinder clinical translation nanomedicines suggests possible remedies surpass them. Graphic

Language: Английский

Citations

1
Fed-RHLP: Enhancing Federated Learning with Random High-Local Performance Client Selection for Improved Convergence and Accuracy DOI Open Access

Pramote Sittijuk,

Kreangsak Tamee

Symmetry, Journal Year: 2024, Volume and Issue: 16(9), P. 1181 - 1181

Published: Sept. 9, 2024

We introduce the random high-local performance client selection strategy, termed Fed-RHLP. This approach allows opportunities for higher-performance clients to contribute more significantly by updating and sharing their local models global aggregation. Nevertheless, it also enables lower-performance participate collaboratively based on proportional representation determined probability of roulette wheel (RW). Improving symmetry in federated learning involves IID Data: is naturally present, making model updates easier aggregate Non-IID asymmetries can impact fairness. Solutions include data balancing, adaptive algorithms, robust aggregation methods. Fed-RHLP enhances allowing representation, which performance. fosters inclusivity collaboration both scenarios. In this work, through experiments, we demonstrate that offers accelerated convergence speed improved accuracy aggregating final model, effectively mitigating challenges posed Data distribution

Language: Английский

Citations

1