Privacy Framework Research Articles

ANEL: a novel efficient and lightweight authentication scheme for enhancing security in Vehicular Ad Hoc Networks (VANETs) using elliptic curve cryptography

Vehicular Ad-hoc Networks (VANETs) offer immense potential for revolutionizing modern transportation systems through real-time communication and interaction between vehicles and roadside infrastructure. These networks can significantly enhance traffic management, improve safety, and support various vehicular services. However, ensuring secure communication in VANETs remains a critical challenge due to the highly sensitive nature of the data exchanged, such as location and personal information. Current authentication schemes, including the ANEL scheme, exhibit vulnerabilities like data interception and tracking risks. In this paper, we propose a novel, efficient, and lightweight authentication scheme that leverages elliptic curve cryptography (ECC) and message mapping techniques to address these security issues. The proposed solution strengthens encryption, enhances privacy protection, and mitigates risks such as unauthorized access and man-in-the-middle attacks. Furthermore, the scheme offers scalability and distinguishes between regular and priority users, making it adaptable to various VANET applications. This approach significantly improves the overall security and privacy framework of VANET systems while ensuring the efficient use of computational resources in vehicular environments.

Running out the digital clock: transatlantic privacy politics and veto points in time

ABSTRACT Why does the European Commission continue to sign onto unstable international data flow agreements with the United States? After two agreements were struck down by the European Court of Justice, and clear signals that a new one could meet a similar fate, the European Commission agreed to the Data Privacy Framework. To make sense of this behavior, we connect traditional work on veto points with work on historical institutionalism to highlight an international negotiation strategy – running out the clock. While traditional veto points literature suggests multiple institutions with varying preferences will limit the set of potential policies or their adoption, we highlight how these dynamics change when considering veto points in time. We showcase our argument by building three historical narratives detailing the negotiation of the successive data flow agreements. Our findings have important implications for the future of the transatlantic privacy regime as well as negotiation dynamics.

Local Differential Privacy for correlated location data release in ITS

The ubiquity of location positioning devices has facilitated the implementation of various Intelligent Transportation System (ITS) applications that generate an enormous volume of location data. Recently, Local Differential Privacy (LDP) has been proposed as a rigorous privacy framework that permits the continuous release of aggregate location statistics without relying on a trusted data curator. However, the conventional LDP was built upon the assumption of independent data, which may not be suitable for inherently correlated location data. This paper investigates the quantification of potential privacy leakage in a correlated location data release scenario under a local setting, which has not been addressed in the literature. Our analysis shows that the privacy guarantee of LDP could be degraded in the presence of spatial–temporal and user correlations, albeit the perturbation is performed locally and independently by the users. This privacy guarantee is bounded by a privacy barrier that is affected by the intensity of correlations. We derive several important closed-form expressions and design efficient algorithms to compute such privacy leakage in a correlated location data. We subsequently propose a Δ-CLDP model that enhances the conventional LDP by incorporating the data correlations, and design a generic LDP data release framework that renders adaptive personalization of privacy preservation. Extensive theoretical analyses and simulations on scalable real datasets validate the security and performance efficiency of our work.

A Survey on User Profiling, Data Collection, and Privacy Issues of Internet Services

Users are usually required to share several types of data, including their personal data, as different providers strive to offer high-quality services that are often tailored to end-users’ preferences. However, when it comes to personalizing services, there are several challenges for meeting user’s needs and preferences. For content personalization and delivery of services to end users, services typically create user profiles. When user profiles are created, user data is collected and organized to meet the personalization requirements of the services. In this paper, we provide an overview of current research activities that focus on user profiling and ways to protect user data privacy. The paper presents different types of data that services collect from users on examples of commonly used Internet services. It proposes data categorization as a prerequisite for controlled data sharing between users and Internet services. Furthermore, it discusses how data generalization can be used for anonymization purposes on examples of the proposed data categories. Finally, it gives an overview of the privacy framework being developed and gives guidelines for future work focusing on data generalization methods in order to reduce user privacy risks.

A critical review of machine learning applications in supply chain risk management

This article examines the transformative role of Machine Learning (ML) in Supply Chain Risk Management (SCRM), emphasizing its ability to enhance risk prediction and mitigation through real-time, data-driven insights. It demonstrates how ML applications like demand forecasting, inventory optimization, supplier risk assessment, and fraud detection improve supply chain resilience by analyzing large datasets to identify patterns, predict risks, and recommend proactive actions. However, the article also highlights key challenges, including data quality, availability, and privacy issues, which limit the effectiveness of ML models. Integrating ML with legacy systems poses additional technical and financial difficulties, particularly for smaller businesses. High implementation costs and scalability constraints further hinder widespread adoption. Ethical concerns, such as data bias and privacy, still need to be explored, raising questions about responsible ML use in supply chains. Future research should address these gaps by improving data governance frameworks for accuracy and privacy, developing scalable ML models suited to various supply chain environments, and exploring synergies with technologies like blockchain and the Internet of Things (IoT). These advancements will help realize ML’s full potential, fostering more agile, transparent, and resilient supply chains capable of navigating complex global risks. By tackling these challenges, ML can shift SCRM from reactive to proactive, ensuring long-term operational continuity and a competitive edge in the evolving global market.

Designing a utility evaluation framework for synthetic health data

ObjectivesSynthetic data (SD) promises to unlock health data for training, research, and innovation. However, where utility evaluation is performed, it is applied ad-hoc for a single task of interest. We produce an initial design for a robust benchmark across a range of tasks. ApproachWe undertook several projects as a prototyping experiment to gather requirements. These projects replicate previous studies performed on the Medical Information Mart for Intensive Care — a dataset used in more than 4,000 studies. We refine definitions, identify personas, draft a user statement, and collect requirements. ResultsDefinitions: We define utility as an extrinsic measure of SD on a larger system, most often through comparison to system performance on real data. This contrasts with fidelity, which measures the accuracy of SD through direct comparison to real data. Personas: Data custodian, User of SD, SD researcher. User statement: As a technical stakeholder, I need a reliable way to measure the utility of datasets and a benchmark to compare generation techniques. Requirements: SD researchers can focus on generation not evaluation; Supports comparison and leaderboards; Based on relevant and applications; Comprehensive across study types and applications; Future proof for population research requiring linking. ConclusionWe propose the following design: Data pipelines follow an extract-generate-evaluate workflow. Study types include cross-sectional and longitudinal. Applications include predictive modelling and clinical research. This results in a comprehensive utility benchmarking suite that complements current frameworks for fidelity and privacy of SD.

A Framework for Adversarial Streaming Via Differential Privacy and Difference Estimators

Classical streaming algorithms operate under the (not always reasonable) assumption that the input stream is fixed in advance. Recently, there is a growing interest in designing robust streaming algorithms that provide provable guarantees even when the input stream is chosen adaptively as the execution progresses. We propose a new framework for robust streaming that combines techniques from two recently suggested frameworks by Hassidim et al. (NeurIPS 2020) and by Woodruff and Zhou (FOCS 2021). These recently suggested frameworks rely on very different ideas, each with its own strengths and weaknesses. We combine these two frameworks into a single hybrid framework that obtains the “best of both worlds”, thereby solving a question left open by Woodruff and Zhou.

Contextual Acceptance of COVID-19 Mitigation Mobile Apps in the United States: Mixed Methods Survey Study on Postpandemic Data Privacy.

The COVID-19 pandemic gave rise to countless user-facing mobile apps to help fight the pandemic ("COVID-19 mitigation apps"). These apps have been at the center of data privacy discussions because they collect, use, and even retain sensitive personal data from their users (eg, medical records and location data). The US government ended its COVID-19 emergency declaration in May 2023, marking a unique time to comprehensively investigate how data privacy impacted people's acceptance of various COVID-19 mitigation apps deployed throughout the pandemic. This research aims to provide insights into health data privacy regarding COVID-19 mitigation apps and policy recommendations for future deployment of public health mobile apps through the lens of data privacy. This research explores people's contextual acceptance of different types of COVID-19 mitigation apps by applying the privacy framework of contextual integrity. Specifically, this research seeks to identify the factors that impact people's acceptance of data sharing and data retention practices in various social contexts. A mixed methods web-based survey study was conducted by recruiting a simple US representative sample (N=674) on Prolific in February 2023. The survey includes a total of 60 vignette scenarios representing realistic social contexts that COVID-19 mitigation apps could be used. Each survey respondent answered questions about their acceptance of 10 randomly selected scenarios. Three contextual integrity parameters (attribute, recipient, and transmission principle) and respondents' basic demographics are controlled as independent variables. Regression analysis was performed to determine the factors impacting people's acceptance of initial data sharing and data retention practices via these apps. Qualitative data from the survey were analyzed to support the statistical results. Many contextual integrity parameter values, pairwise combinations of contextual integrity parameter values, and some demographic features of respondents have a significant impact on their acceptance of using COVID-19 mitigation apps in various social contexts. Respondents' acceptance of data retention practices diverged from their acceptance of initial data sharing practices in some scenarios. This study showed that people's acceptance of using various COVID-19 mitigation apps depends on specific social contexts, including the type of data (attribute), the recipients of the data (recipient), and the purpose of data use (transmission principle). Such acceptance may differ between the initial data sharing and data retention practices, even in the same context. Study findings generated rich implications for future pandemic mitigation apps and the broader public health mobile apps regarding data privacy and deployment considerations.

Balancing confidentiality and care coordination: challenges in patient privacy.

In the digital age, maintaining patient confidentiality while ensuring effective care coordination poses significant challenges for healthcare providers, particularly nurses. To investigate the challenges and strategies associated with balancing patient confidentiality and effective care coordination in the digital age. A cross-sectional study was conducted in a general hospital in Egypt to collect data from 150 nurses across various departments with at least six months of experience in patient care. Data were collected using six tools: Demographic Form, HIPAA Compliance Checklist, Privacy Impact Assessment (PIA) Tool, Data Sharing Agreement (DSA) Framework, EHR Privacy and Security Assessment Tool, and NIST Cybersecurity Framework. Validity and Reliability were ensured through pilot testing and factor analysis. Participants were primarily aged 31-40 years (45%), with 75% female and 60% staff nurses. High compliance was observed in the HIPAA Compliance Checklist, especially in Administrative Safeguards (3.8 ± 0.5), indicating strong management and training processes, with an overall score of 85 ± 10. The PIA Tool showed robust privacy management, with Project Descriptions scoring 4.5 ± 0.3 and a total score of 30 ± 3. The DSA Framework had a mean total score of 20 ± 2, with Data Protection Measures scoring highest at 4.0 ± 0.4. The EHR assessments revealed high scores in Access Controls (4.4 ± 0.3) and Data Integrity Measures (4.3 ± 0.3), with an overall score of 22 ± 1.5. The NIST Cybersecurity Framework had a total score of 18 ± 2, with the highest scores in Protect (3.8) and lower in Detect (3.6). Strong positive correlations were found between HIPAA Compliance and EHR Privacy (r = 0.70, p < 0.05) and NIST Cybersecurity (r = 0.55, p < 0.05), reflecting effective data protection practices. The study suggests that continuous improvement in privacy practices among healthcare providers, through ongoing training and comprehensive privacy frameworks, is vital for enhancing patient confidentiality and supporting effective care coordination.

User Privacy Protection via Windows Registry Hooking and Runtime Encryption.

The Windows registry contains a plethora of information in a hierarchical database. It includes system-wide settings, user preferences, installed programs, and recently accessed files and maintains timestamps that can be used to construct a detailed timeline of user activities. However, these data are unencrypted and thus vulnerable to exploitation by malicious actors who gain access to this repository. To address this security and privacy concern, we propose a novel approach that efficiently encrypts and decrypts sensitive registry data in real time. Our developed proof-of-concept program intercepts interactions between the registry's application programming interfaces (APIs) and other Windows applications using an advanced hooking technique. This enables the proposed system to be transparent to users without requiring any changes to the operating system or installed software. Our approach also implements the data protection API (DPAPI) developed by Microsoft to securely manage each user's encryption key. Ultimately, our research provides an enhanced security and privacy framework for the Windows registry, effectively fortifying the registry against security and privacy threats while maintaining its accessibility to legitimate users and applications.

Commodifying Data: Analyzing Legal Regulations on Foreign Digital Payment Systems A Comparative Analysis Between the United States and the People’s Republic of China

This paper provides a comprehensive study on the intersection of digital payment platforms, particularly those originating from the People's Republic of China (PRC), such as Alipay and WeChat Pay, with the legal frameworks for data privacy in the United States. The research investigates whether the data collection policies of these Chinese digital payment systems comply with or violate existing U.S. laws designed to protect personal data. Given the rapid global expansion of these platforms and their necessity for facilitating modern economic activities, the paper emphasizes the significant trade-off users face by surrendering personal data to access digital payment services. This surrender of data, equated to a "pay-to-play" system, raises critical questions about the adequacy of current legal protections for users' data privacy in the U.S. This inquiry is crucial in understanding the broader implications of the global rise of digital payment platforms on privacy and consumer protection laws, particularly as these platforms become increasingly integrated into the international market. The paper aims to contribute to the ongoing discourse on digital privacy and the need for modernized legal frameworks to address the challenges posed by the digital economy's global nature.

Regulatory compliance in the age of data privacy: A comparative study of the Nigerian and U.S. legal landscapes

This comparative study delves into the intricate regulatory compliance mechanisms governing data privacy in Nigeria and the United States. Employing a qualitative methodology, this research synthesizes existing literature and case studies to highlight the pivotal role of data protection regulations in safeguarding personal information in an increasingly digital world. The analysis begins with an exploration of the conceptual framework of data privacy, emphasizing key principles such as transparency, accountability, and data minimization. It then meticulously examines the Nigerian Data Protection Regulation (NDPR) and the diverse U.S. legal landscape, including the California Consumer Privacy Act (CCPA) and the Health Insurance Portability and Accountability Act (HIPAA). The main findings reveal significant differences in regulatory approaches, with Nigeria's centralized enforcement by the National Information Technology Development Agency (NITDA) contrasting sharply with the U.S.'s fragmented state and federal oversight. Case studies underscore the challenges faced by both jurisdictions in ensuring compliance, ranging from resource constraints in Nigeria to the complexities of navigating multiple regulatory requirements in the U.S. The study concludes that while both countries emphasize the importance of data privacy, their divergent legal landscapes reflect unique cultural, economic, and operational contexts. The research recommends harmonizing data protection standards, enhancing enforcement mechanisms, and fostering greater public awareness to address the evolving challenges of data privacy. This study contributes to the global discourse on data privacy by offering insights into the strengths and weaknesses of different regulatory frameworks, thereby providing a roadmap for policymakers, businesses, and other stakeholders to navigate the complexities of data privacy compliance. Keywords: Data Privacy, Regulatory Compliance, Nigeria, United States, Data Protection, Legal Frameworks.

On the Connection Between the ABS Perturbation Methodology and Differential Privacy

This paper explores analytical connections between the perturbation methodology of the Australian Bureau of Statistics (ABS) and the differential privacy (DP) framework. We consider a single static counting query function and find the analytical form of the perturbation distribution with symmetric support for the ABS perturbation methodology. We then analytically measure the DP parameters, namely the (ε, δ) pair, for the ABS perturbation methodology under this setting. The results and insights obtained about the behaviour of (ε, δ) with respect to the perturbation support and variance are used to judiciously select the variance of the perturbation distribution give a good δ in the DP framework for a given desired ε and perturbation support. Finally, we propose a simple sampling scheme to implement the perturbation probability matrix in the ABS Cellkey method. The post sampling (ε, δ) pair is numerically analysed as a function of the Cellkey size. It is shown that the best results are obtained for a larger Cellkey size, because the (ε, δ) pair ost-sampling measures remain almost identical when we compare sampling and theoretical results.

Evaluation based on multi-criteria decision-making methods and spherical fuzzy framework for security and privacy in metaverse technologies: A case study

Integrating the metaverse technology with the transportation system has several security and privacy issues. This study assesses the 12 security solutions to select the best one to overcome security and privacy issues (such as data theft, unauthorized access, and theft of personal data) when integrating the transportation system with metaverse technology. A suggested methodology is conducted by experts and decision-makers using linguistic terms and spherical fuzzy numbers to express their opinions on evaluating the criteria and alternatives. Selecting the best security solution (alternative) is critical because it includes several conflict security criteria, such as data theft, authentication, security attacks, and others. This paper introduces a methodology for multi-criteria decision-making (MCDM) in a spherical fuzzy (SF) environment. The MCDM method dealt with various conflicting criteria, and SF dealt with uncertainty and vague information while evaluating the criteria and alternatives. The suggested methodology consists of two main phases. The first phase introduces the analytic hierarchy process (SF-AHP) method to compute the criteria weights. The second phase introduces the Weighted Aggregates Sum Product Assessment (SF-WASPAS) method to rank and select the best alternative. The results show the end-to-end authentication protocol is the best alternative (security solution). This study conducted a sensitivity analysis of the stability of the rank by changing the criteria’s weights. The sensitivity analysis results show that the end-to-end authentication protocol is the best alternative (security solution) in different cases. We compare the suggested methodology with six other MCDM methods: SF-TOPSIS, SF-VIKOR, SF-MABAC, SF-CODAS, SF-MARCOS, and SF-COPRAS to show the effectiveness of the proposed method. The results show that the presented methodology is robust compared to other MCDM methods.

Responsible AI for cardiovascular disease detection: Towards a privacy-preserving and interpretable model

Background and objectiveCardiovascular disease (CD) is a major global health concern, affecting millions with symptoms like fatigue and chest discomfort. Timely identification is crucial due to its significant contribution to global mortality. In healthcare, artificial intelligence (AI) holds promise for advancing disease risk assessment and treatment outcome prediction. However, machine learning (ML) evolution raises concerns about data privacy and biases, especially in sensitive healthcare applications. The objective is to develop and implement a responsible AI model for CD prediction that prioritize patient privacy, security, ensuring transparency, explainability, fairness, and ethical adherence in healthcare applications. MethodsTo predict CD while prioritizing patient privacy, our study employed data anonymization involved adding Laplace noise to sensitive features like age and gender. The anonymized dataset underwent analysis using a differential privacy (DP) framework to preserve data privacy. DP ensured confidentiality while extracting insights. Compared with Logistic Regression (LR), Gaussian Naïve Bayes (GNB), and Random Forest (RF), the methodology integrated feature selection, statistical analysis, and SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) for interpretability. This approach facilitates transparent and interpretable AI decision-making, aligning with responsible AI development principles. Overall, it combines privacy preservation, interpretability, and ethical considerations for accurate CD predictions. ResultsOur investigations from the DP framework with LR were promising, with an area under curve (AUC) of 0.848 ± 0.03, an accuracy of 0.797 ± 0.02, precision at 0.789 ± 0.02, recall at 0.797 ± 0.02, and an F1 score of 0.787 ± 0.02, with a comparable performance with the non-privacy framework. The SHAP and LIME based results support clinical findings, show a commitment to transparent and interpretable AI decision-making, and aligns with the principles of responsible AI development. ConclusionsOur study endorses a novel approach in predicting CD, amalgamating data anonymization, privacy-preserving methods, interpretability tools SHAP, LIME, and ethical considerations. This responsible AI framework ensures accurate predictions, privacy preservation, and user trust, underscoring the significance of comprehensive and transparent ML models in healthcare. Therefore, this research empowers the ability to forecast CD, providing a vital lifeline to millions of CD patients globally and potentially preventing numerous fatalities.

Privacy decision‐making and the effects of privacy choice architecture: Experiments toward the design of behaviorally‐aware privacy regulation

AbstractThe current notice and choice privacy framework fails to empower individuals in effectively making their own privacy choices. In this Article we offer evidence from three novel experiments showing that at the core of this failure is a cognitive error. Notice and choice caters to a heuristic that people employ to make privacy decisions. This heuristic is meant to distinguish between a party's good or bad intent in face‐to‐face‐situations. In the online context, it distorts privacy decision‐making and leaves potential disclosers vulnerable to exploitation.From our experimental evidence exploring the heuristic's effect, we conclude that privacy law must become more behaviorally aware. Specifically, privacy law must be redesigned to intervene in the cognitive mechanisms that keep individuals from making better privacy decisions. A behaviorally‐aware privacy regime must centralize, standardize and simplify the framework for making privacy choices.To achieve these goals, we propose a master privacy template which requires consumers to define their privacy preferences in advance—doing so avoids presenting the consumer with a concrete counterparty, and this, in turn, prevents them from applying the intent heuristic and reduces many other biases that affect privacy decision‐making. Our data show that blocking the heuristic enables consumers to consider relevant privacy cues and be considerate of externalities their privacy decisions cause.The master privacy template provides a much more effective platform for regulation. Through the master template the regulator can set the standard for automated communication between user clients and website interfaces, a facility which we expect to enhance enforcement and competition about privacy terms.

Privacy Essentials

Following a series of legislative changes around privacy over the past 25 years, this study highlights data protection regulations and the complexities of applying these frameworks. To address this, we created a privacy framework to guide organisations in what steps they need to undertake to achieve compliance with the UK GDPR, highlighting the existing privacy frameworks for best practice and the requirements from the Information Commissioners Office. We applied our framework to a UK charity sector; to account for the specific nuances that working in a charity brings, we worked closely with local charities to understand their requirements, and interviewed privacy experts to develop a framework that is readily accessible and provides genuine value. Feeding the results into our privacy framework, a decision tree artefact has been developed for compliance. The artefact has been tested against black-box tests, System Usability Tests and UX Honeycomb tests. Results show that Privacy Essentials! provides the foundation of a data protection management framework and offers organisations the catalyst to start, enhance, or even validate a solid and effective data privacy programme.

Revolutionizing Sleep Health: The Emergence and Impact of Personalized Sleep Medicine.

Personalized sleep medicine represents a transformative shift in healthcare, emphasizing individualized approaches to optimizing sleep health, considering the bidirectional relationship between sleep and health. This field moves beyond conventional methods, tailoring care to the unique physiological and psychological needs of individuals to improve sleep quality and manage disorders. Key to this approach is the consideration of diverse factors like genetic predispositions, lifestyle habits, environmental factors, and underlying health conditions. This enables more accurate diagnoses, targeted treatments, and proactive management. Technological advancements play a pivotal role in this field: wearable devices, mobile health applications, and advanced diagnostic tools collect detailed sleep data for continuous monitoring and analysis. The integration of machine learning and artificial intelligence enhances data interpretation, offering personalized treatment plans based on individual sleep profiles. Moreover, research on circadian rhythms and sleep physiology is advancing our understanding of sleep's impact on overall health. The next generation of wearable technology will integrate more seamlessly with IoT and smart home systems, facilitating holistic sleep environment management. Telemedicine and virtual healthcare platforms will increase accessibility to specialized care, especially in remote areas. Advancements will also focus on integrating various data sources for comprehensive assessments and treatments. Genomic and molecular research could lead to breakthroughs in understanding individual sleep disorders, informing highly personalized treatment plans. Sophisticated methods for sleep stage estimation, including machine learning techniques, are improving diagnostic precision. Computational models, particularly for conditions like obstructive sleep apnea, are enabling patient-specific treatment strategies. The future of personalized sleep medicine will likely involve cross-disciplinary collaborations, integrating cognitive behavioral therapy and mental health interventions. Public awareness and education about personalized sleep approaches, alongside updated regulatory frameworks for data security and privacy, are essential. Longitudinal studies will provide insights into evolving sleep patterns, further refining treatment approaches. In conclusion, personalized sleep medicine is revolutionizing sleep disorder treatment, leveraging individual characteristics and advanced technologies for improved diagnosis, treatment, and management. This shift towards individualized care marks a significant advancement in healthcare, enhancing life quality for those with sleep disorders.

A Two-Stage Differential Privacy Scheme for Federated Learning Based on Edge Intelligence.

The issue of data privacy protection must be considered in distributed federated learning (FL) so as to ensure that sensitive information is not leaked. In this article, we propose a two-stage differential privacy (DP) framework for FL based on edge intelligence. Various levels of privacy preservation can be provided according to the degree of data sensitivity. In the first stage, the randomized response mechanism is used to perturb the original feature data by the user terminal for data desensitization, and the user can self-regulate the level of privacy preservation. In the second stage, noise is added to the local models by the edge server to further guarantee the privacy of the models. Finally, the model updates are aggregated in the cloud. In order to evaluate the performance of the proposed end-edge-cloud FL framework in terms of training accuracy and convergence, extensive experiments are conducted on a real electrocardiogram (ECG) signal dataset. Bi-directional long-short-term memory (BiLSTM) neural network is adopted to training classification model. The effect of different combinations of feature perturbation and noise addition on the model accuracy is analyzed depending on different privacy budgets and parameters. The experimental results demonstrate that the proposed privacy-preserving framework provides good accuracy and convergence while ensuring privacy.

A blockchain-empowered federated learning-based framework for data privacy in lung disease detection system

Lung diseases are one of the prime reasons for mortality globally, having an estimated five million per year fatal cases worldwide. This is a growing global concern so early detection using a Computed Tomography (CT) scan is crucial to prevent loss that grabs the attention of cutting-edge technologies to bring the concept called “Smart Healthcare”. However, the paucity and heterogeneity of medical data across the globe make it challenging to develop a global classification framework, while the other concerns that arise from legal and privacy leakage become an obstacle for data sharing as single source data is hardly enough to represent universal. Federated Learning has issued a solution to licensing research and data heterogeneity concerns allowing collaborative and on-device learning without sharing raw data. FL faces security issues such as Denial-of-service, Reverse engineering attacks, etc, where it is impossible to track the data and store it securely. The study proposes an innovative framework that combines Blockchain technology and Federated Learning (FL) to enable collaborative model training while preserving data privacy. Through this approach, patient data is authenticated using blockchain, and FL facilitates on-device learning without sharing raw data. The framework utilizes the DenseNet-201 model for lung disease classification, with model parameter aggregation using the FedAvg algorithm and storage on the blockchain via IPFS. Finally, we have conducted a substantial investigation with Python and its widely used libraries, like TensorFlow and Scikit-Learn to demonstrate that the algorithm accurately detects lung diseases and attained an accuracy, precision, recall, and F1-score of 90%.