Anonymization Process Research Articles

Providing Delayed, In-Person Collected Feedback From Residents to Teaching Faculty: Lessons Learned.

Background Teaching faculty request timely feedback from residents to improve their skills. Yet even with anonymous processes, this upward feedback can be difficult to obtain as residents raise concerns about identification and repercussions. Objective To examine faculty perception of the quality and content of feedback from residents after increasing anonymity and sacrificing timeliness. Methods Between 2011 and 2017, an associate program director at a large internal medicine residency program met briefly with each resident individually to obtain feedback about their teaching faculty shortly after their rotation. To improve anonymity, residents were promised their feedback would not be released until they graduated. In 2019, all feedback was collated and released at one time to faculty. We administered 3 timed, voluntary, anonymous, 36-item closed-ended surveys to faculty asking about the content and value, and to self-identify whether the feedback was praise, constructive, or criticism. Results Exactly 189 faculty participated with 140 completing all 3 surveys (74.1% response rate). Faculty reported this feedback content to be of higher quality (81.0%, 81 of 100) and quantity (82.4%, 84 of 102) in contrast to prior feedback. More than 85.4% (88 of 103) of faculty agreed this feedback was more specific. Faculty identified less praise (median 35.0% vs median 50.0%, P<.001) and more negative constructive feedback (median 20.0% vs median 5.0%, P<.001) compared to prior feedback. More than 82% (116 of 140) of faculty reported it would change their behavior, but 3 months after receiving the feedback, only 63.6% (89 or 140) felt the same way (P<.001). Faculty were divided on the necessity of a time delay, with 41.4% (58 of 140) believing it reduced the feedback's value. Despite the delay, 32.1% (45 of 140) felt they could identify residents. Conclusions Offering a substantial delay in feedback delivery increased anonymity and enabled residents to furnish more nuanced and constructive comments; however, faculty opinions diverged on whether this postponement was valuable.

Big data anonymization using Spark for enhanced privacy protection

This article introduces an advanced solution for anonymizing large-scale sensitive data, addressing the limitations of traditional approaches when applied to vast datasets. By leveraging the Spark distributed computing framework, we propose a method that parallelizes the data anonymization process, enhancing efficiency and scalability. Utilizing Spark's resilient distributed datasets (RDD), the approach integrates two primary operations, Map_RDD and ReduceByKey_RDD, to execute the anonymization tasks. Our comprehensive experimental evaluation demonstrates our solution's effectiveness and improved performance in preserving data privacy while balancing data utility and confidentiality. A significant contribution of our study is the development of a wide array of solutions for data owners, particularly notable for a 500 MB dataset at an anonymity level of K=100, where our methodology produces 832 unique solutions. This study also opens avenues for future research in applying different privacy models within the Spark ecosystem, such as l-diversity and t-closeness.

Privacy Preserving Using K Member Gaussian Kernel Fuzzy C Means and Self Adaptive Honey Badger for Online Social Networks

An online social network (OSN) gives users a strong platform to communicate and exchange information. Protecting publicly published data from individual identification is the primary problem in sharing social network databases. The most popular method for protecting privacy is anonymizing data, which involves deleting or altering some information while maintaining as much of the original data as feasible. This work presents a combination anonymizing algorithm, which is based on k member Gaussian kernel fuzzy c means clustering and self-adaptive honey badger optimization technique (KGKFCM-SAHBO). As part of the suggested anonymization process, the various users are divided into C clusters, each of which has at least K users, using a K-member Gaussian kernel fuzzy c means clustering technique. After that, the primary clusters are further optimized using the self-adaptive honey badger optimization approach (SAHBO) to further anonymize the data and network graph. Using the Yelp dataset, the experimental findings demonstrate the efficiency of the proposed model and evaluate a number of parameters, including execution time, degree of anonymization, and information loss. The experimental results show that the proposed strategy reduced information loss and improved the degree of anonymization when compared to existing methodologies.

Promoting equality, diversity and inclusion in research and funding: reflections from a digital manufacturing research network

BackgroundEqual, diverse, and inclusive teams lead to higher productivity, creativity, and greater problem-solving ability resulting in more impactful research. However, there is a gap between equality, diversity, and inclusion (EDI) research and practices to create an inclusive research culture. Research networks are vital to the research ecosystem, creating valuable opportunities for researchers to develop their partnerships with both academics and industrialists, progress their careers, and enable new areas of scientific discovery. A feature of a network is the provision of funding to support feasibility studies – an opportunity to develop new concepts or ideas, as well as to ‘fail fast’ in a supportive environment. The work of networks can address inequalities through equitable allocation of funding and proactive consideration of inclusion in all of their activities.MethodsThis study proposes a strategy to embed EDI within research network activities and funding review processes. This paper evaluates 21 planned mitigations introduced to address known inequalities within research events and how funding is awarded. EDI data were collected from researchers engaging in a digital manufacturing network activities and funding calls to measure the impact of the proposed method.ResultsQuantitative analysis indicates that the network’s approach was successful in creating a more ethnically diverse network, engaging with early career researchers, and supporting researchers with care responsibilities. However, more work is required to create a gender balance across the network activities and ensure the representation of academics who declare a disability. Preliminary findings suggest the network’s anonymous funding review process has helped address inequalities in funding award rates for women and those with care responsibilities, more data are required to validate these observations and understand the impact of different interventions individually and in combination.ConclusionsIn summary, this study offers compelling evidence regarding the efficacy of a research network's approach in advancing EDI within research and funding. The network hopes that these findings will inform broader efforts to promote EDI in research and funding and that researchers, funders, and other stakeholders will be encouraged to adopt evidence-based strategies for advancing this important goal.

XimSwap: Many-to-Many Face Swapping for TinyML

The unprecedented development of deep learning approaches for video processing has caused growing privacy concerns. To ensure data analysis while maintaining privacy, it is essential to address how to protect individuals’ identities. One solution is to anonymize data at the source, avoiding the transmission or storage of information that could lead to identification. This study introduces XimSwap, a novel deep learning technique for real-time video anonymization, which can remove facial identification features directly on edge devices with minimal computational resources. Our approach offers a comprehensive solution that guarantees privacy by design. This novel method for implementing face-swapping ensures that the pose and expression of a target face remain unchanged and can be used on embedded devices with very limited computational resources. By incorporating style transfer layers into convolutional ones and optimizing the network’s operation, we achieved a reduction of over 98% in the required operations and parameters compared with state-of-the-art architectures. Our approach also significantly reduces RAM usage, making it possible to implement the anonymization process on tiny edge devices, including microcontrollers, such as the STM32H743.

The Costs of Anonymization: Case Study Using Clinical Data.

Sharing data from clinical studies can accelerate scientific progress, improve transparency, and increase the potential for innovation and collaboration. However, privacy concerns remain a barrier to data sharing. Certain concerns, such as reidentification risk, can be addressed through the application of anonymization algorithms, whereby data are altered so that it is no longer reasonably related to a person. Yet, such alterations have the potential to influence the data set's statistical properties, such that the privacy-utility trade-off must be considered. This has been studied in theory, but evidence based on real-world individual-level clinical data is rare, and anonymization has not broadly been adopted in clinical practice. The goal of this study is to contribute to a better understanding of anonymization in the real world by comprehensively evaluating the privacy-utility trade-off of differently anonymized data using data and scientific results from the German Chronic Kidney Disease (GCKD) study. The GCKD data set extracted for this study consists of 5217 records and 70 variables. A 2-step procedure was followed to determine which variables constituted reidentification risks. To capture a large portion of the risk-utility space, we decided on risk thresholds ranging from 0.02 to 1. The data were then transformed via generalization and suppression, and the anonymization process was varied using a generic and a use case-specific configuration. To assess the utility of the anonymized GCKD data, general-purpose metrics (ie, data granularity and entropy), as well as use case-specific metrics (ie, reproducibility), were applied. Reproducibility was assessed by measuring the overlap of the 95% CI lengths between anonymized and original results. Reproducibility measured by 95% CI overlap was higher than utility obtained from general-purpose metrics. For example, granularity varied between 68.2% and 87.6%, and entropy varied between 25.5% and 46.2%, whereas the average 95% CI overlap was above 90% for all risk thresholds applied. A nonoverlapping 95% CI was detected in 6 estimates across all analyses, but the overwhelming majority of estimates exhibited an overlap over 50%. The use case-specific configuration outperformed the generic one in terms of actual utility (ie, reproducibility) at the same level of privacy. Our results illustrate the challenges that anonymization faces when aiming to support multiple likely and possibly competing uses, while use case-specific anonymization can provide greater utility. This aspect should be taken into account when evaluating the associated costs of anonymized data and attempting to maintain sufficiently high levels of privacy for anonymized data. German Clinical Trials Register DRKS00003971; https://drks.de/search/en/trial/DRKS00003971. RR2-10.1093/ndt/gfr456.

Artificial Intelligence in Psychiatry: A Comprehensive Literature Review

IntroductionThe incorporation of artificial intelligence (AI) in healthcare, especially in mental health services, offers potential advancements in efficiency and personalization. As AI technologies like machine learning and natural language processing (NLP) continue to evolve, it’s vital to evaluate their applications in psychiatry comprehensively.ObjectivesThis review aims to summarize and characterize studies that used AI, particularly machine learning and NLP, in mental health. Additionally, it endeavors to understand how these technologies may enhance diagnostic tools, symptom monitoring, and delivery of personalized treatment in psychiatry.MethodsAdhering to PRISMA guidelines, a systematic search was executed across multiple medical databases, including PubMed, Scopus, ScienceDirect, and PsycINFO. Keywords encompassed machine learning, data mining, psychiatry, and mental health. Exclusion criteria included non-English papers, anonymization process descriptions, case studies, conference papers, and other reviews. Data from various segments in the provided information were synthesized to capture the broader picture of AI’s application in psychiatry.ResultsFrom the 327 articles initially identified, 58 were chosen for detailed review. Studies predominantly revolved around three main populations: patients in medical databases, emergency room visitors, and social media users. The primary applications of AI entailed symptom extraction, illness severity classification, therapy effectiveness comparison, and psychopathological insights derivation. Data sources mainly included medical records and social media, with Python emerging as the preferred platform for most studies.ConclusionsWhile AI shows immense promise in revolutionizing mental health care, its current applications largely confirm existing clinical hypotheses. Ethical concerns, such as patient privacy and data biases, remain paramount. Future work should delve deeper into these challenges while further exploring AI’s potential in clinical psychiatry practice.Disclosure of InterestNone Declared

Software development to anonymize patient information from clinical computed tomography images

A free software for digital imaging and communications in medicine (DICOM) image anonymization is needed to protect patient information from third parties. This study aimed to develop a software for the anonymization of patient information in computed tomography (CT) image. There was a total of 17 informations to be anonymized, such as Patient’s Name, Other Patient Names, Patient’s ID, Patient’s Birth Date, Patient’s Sex, Patient’s Age, Study ID, Study Description, Series Description, Institution Name, Institution Adress, Referring Phisician’s Name, Consulting Physician’s Name, Performing Physician’s Name, Name of Physician(s) Reading Study, Operator’s Name, and Protocol Name. In every information, its initial value was replaced with a dummy value with the string value of "N/A". For testing, patient CT images from four different hospitals with different scanners were collected. It is found that each scanner had different information stored in DICOM information. However, the anonymization process in the four hospitals works well with accuracy of 100%. The developed software can anonymize DICOM images flexibly and successfully. This software can be used for anonymization of patient information in order to protect patient information.

Anonymizing eye-tracking stimuli with stable diffusion

Casual users nowadays can create almost arbitrary image content by providing textual prompts to generative machine-learning models. These models rapidly improve image quality with each new generation, providing means to create photos, paintings in different styles, and even videos. One feature of such models is the ability to take an image as input and adjust content according to a prompt. A visual obfuscation of content can be achieved for static images and videos by slightly changing persons, text, and other objects. The potential of this technique can be applied in eye-tracking experiments for post-hoc dissemination of analysis results and visualization. In this work, we discuss how the technique could serve to anonymize stimuli (e.g., for double-blind reviews, remove product placements, etc.) and protect the privacy of people visible in the stimuli. We further investigate how the application of this anonymization process influences visual saliency and the depiction of stimuli in visualization techniques. Our results show that slight image transformations do not drastically change the saliency of a scene but obfuscate objects and faces while keeping important image structures for context.

A Secure Face Verification Scheme Based on Fully Homomorphic Encryption with Anonymity

With the widespread adoption of cloud computing, the face verification process often requires the client to upload the face to an untrusted cloud server to obtain the verification results. Privacy leakage issues may arise if the client’s private information is not protected. This paper proposes a secure and anonymous face verification scheme using fully homomorphic encryption technology and SealPIR. Our scheme is a three-party solution that requires a third-party server trusted by the client. This scheme not only prevents the client’s facial data from being obtained by untrusted data servers but also prevents the data server from learning the index corresponding to the face that the client wants to verify. In a single-face verification process, the client only needs to perform one upload operation and one download operation, with a communication volume of 264 KB. We can complete a privacy-protected anonymous face verification process in 84.91 ms.

CONSIDERAŢII ISTORICO-TEORETICE LIBERTARIENE DESPRE “OBICEIUL PĂMÂNTULUI” SPECIFIC SATELOR DEVĂLMAȘE ROMÂNEȘTI – SEC. IV- XIX-LEA. UN STUDIU DE ISTORIE SOCIALĂ ȘI FILOSOFIA CULTURII. PARTEA a II- a

There is today among jurists or a widespread tendency to consider that the institutions of positive law, in their capacity as late creations of the modern state, can be considered, based on their ideal, formal, general, logical and rational, as independent realities, which can be studied in themselves, separated both by the living conditions of the society to which they apply, and by the manifestations of the popular spirit of the latter. However, modern legal norms are not established solely as a result of direct state action. Other sources are involved in their construction. Thus, at the base of the legal norms we discover the moral-practical doctrines, the jurisprudence, the tradition, but especially the customs. It is a series of primary sources through which the norms maintain their connection with the conditions and needs of social life, which, moreover, they norm. This finding leads us to say that legal institutions and norms are not the exclusive products of a well-determined state legislator, a logical, objective and general production of his imagination, will and especially reason, but the creation of an anonymous and continuous process of the whole cultural, the expression of a more or less long historical evolution of the human community on which it is reflected.

Consider anonymous recruitment process when mulling new board members

In many nonprofits, the senior leadership, staff and board are constantly on the lookout for individuals who might make good additions to the board when an opening comes up. And when a possible nominee is identified, it's often standard practice to reach out the person, letting them know they might be contacted for consideration.

Data Anonymization Process Challenges and Context Missions

Data anonymization is one of the solutions allowing companies to comply with the GDPR directive in terms of data protection. In this context, developers must follow several steps in the process of data anonymization in development and testing environments. Indeed, real personal and sensitive data must not leave the production environment which is very secure. Often, anonymization experts are faced with difficulties including the lack of data flows and mapping between data sources, the non-cooperation of the database project teams (refusal to change) or even the lack of skills of these teams present due to the age of the systems developed by experienced teams who unfortunately left the project. Other problems are lack of data models. The aim of this paper is to discuss an anonymization process of databases of banking applications and present our context-based recommendations to overcome the different issues met and the solutions to improve methodologies of data anonymization process.

CONSIDERAŢII ISTORICO-TEORETICE LIBERTARIENE DESPRE “OBICEIUL PĂMÂNTULUI” SPECIFIC SATELOR DEVĂLMAȘE ROMÂNEȘTI ÎNTRE SECOLELE AL IV-LEA ȘI AL XIX-LEA. UN STUDIU DE ISTORIE SOCIALĂ ȘI FILOSOFIA CULTURII. PARTEA I

There is today among jurists or a widespread tendency to consider that the institutions of positive law, in their capacity as late creations of the modern state, can be considered, based on their ideal, formal, general, logical and rational, as independent realities, which can be studied in themselves, separated both by the living conditions of the society to which they apply, and by the manifestations of the popular spirit of the latter. However, modern legal norms are not established solely as a result of direct state action. Other sources are involved in their construction. Thus, at the base of the legal norms we discover the moral-practical doctrines, the jurisprudence, the tradition, but especially the customs. It is a series of primary sources through which the norms maintain their connection with the conditions and needs of social life, which, moreover, they norm. This finding leads us to say that legal institutions and norms are not the exclusive products of a well-determined state legislator, a logical, objective and general production of his imagination, will and especially reason, but the creation of an anonymous and continuous process of the whole cultural, the expression of a more or less long historical evolution of the human community on which it is reflected.

Ethical considerations on artificial intelligence in dentistry: A framework and checklist

ObjectiveArtificial Intelligence (AI) refers to the ability of machines to perform cognitive and intellectual human tasks. In dentistry, AI offers the potential to enhance diagnostic accuracy, improve patient outcomes and streamline workflows. The present study provides a framework and a checklist to evaluate AI applications in dentistry from this perspective. MethodsLending from existing guidance documents, an initial draft of the checklist and an explanatory paper were derived and discussed among the groups members. ResultsThe checklist was consented to in an anonymous voting process by 29 Topic Group Dental Diagnostics and Digital Dentistry, ITU/WHO Focus Group AI on Health's members. Overall, 11 principles were identified (diversity, transparency, wellness, privacy protection, solidarity, equity, prudence, law and governance, sustainable development, accountability, and responsibility, respect of autonomy, decision-making). ConclusionsProviders, patients, researchers, industry, and other stakeholders should consider these principles when developing, implementing, or receiving AI applications in dentistry. Clinical SignificanceWhile AI has become increasingly commonplace in dentistry, there are ethical concerns around its usage, and users (providers, patients, and other stakeholders), as well as the industry should consider these when developing, implementing, or receiving AI applications based on comprehensive framework to address the associated ethical challenges.

Quality assessment of breast cancer studies conducted with the Delphi technique.

The Delphi technique is a consensus method aiming to obtain statistical estimations from a qualitative approach, through an iterative process that leads to consensus within experts. The main characteristics of the technique include iteration, anonymity, feedback, and consensus reaching. When high-quality, quantitative evidence on a particular topic is insufficient, the Delphi technique can be used for making decisions in clinical scenarios. However, the quality of studies on breast cancer conducted with this technique, has not been assessed. We aim to evaluate the quality of studies on breast cancer which used the Delphi technique as their method. A quality assessment tool (Quali-D) was created through consensus among experts on the Delphi technique. Then, the tool was applied to studies on breast cancer which used the Delphi technique as their method. Studies conducted through the Delphi technique mainly assessed for quality indicators and expressed needs in patients with breast cancer. High-quality characteristics were reported in 63.89% of the studies. 98.61% used the Delphi technique due to lack of a more adequate method to solve their research question. 98.61% summarized and presented results in a clear way. In 91.67% of the studies, at least two rounds were conducted. 86.11% described the methods for expert selection in a complete manner. Only 54.17% of the studies reported an anonymous process and 4.17% of the studies disclosed conflicts of interest thoroughly. A variety of topics were assessed through the Delphi technique in cases where no other technique would have been more appropriate for assessing these issues. Significant limitations are present in terms of anonymity and full disclosure of conflicts of interest. We found that the quality of studies conducted with the Delphi technique regarding breast cancer is overall good. However, the limitations of each study must be considered when applying their results to clinical practice.

Anonymous Methods Based on Multi-Attribute Clustering and Generalization Constraints

The dissemination and sharing of data sheets in IoT applications presents privacy and security challenges that can be addressed using the k-anonymization algorithm. However, this method needs improvement, for example, in areas related to its overgeneralization and its insufficient attribute diversity constraints during the anonymization process. To address these issues, this study proposes a multi-attribute clustering and generalization constraints (k,l)-anonymization method that can be applied to multidimensional data tables. The algorithm first used a greedy strategy to rank the attributes by width first, derived the division into dimensions to construct a multidimensional generalization hierarchy, and then selected the attributes with the most significant width values as the priority generalization attributes. Next, the k-nearest neighbor (KNN) clustering method was introduced to determine the initial clustering center by the width-first results, divide the quasi-identifier attributes into KNN clusters according to a distance metric, and generalize the quasi-identifier attributes in the equivalence class using a hierarchical generalization structure. Then, the proposed method re-evaluated the attributes to be generalized before each generalization operation. Finally, the algorithm employed an improved frequency–diversity constraint to generalize sensitive attributes in order to ensure that there were at least l records that were mutually dissimilar and closest in the equivalence class. While limiting the frequency threshold for the occurrence of sensitive attributes, the sensitive attribute values remained similar within the group, thus achieving protection of anonymity for all the attributes.

An anonymization-based privacy-preserving data collection protocol for digital health data.

Digital health data collection is vital for healthcare and medical research. But it contains sensitive information about patients, which makes it challenging. To collect health data without privacy breaches, it must be secured between the data owner and the collector. Existing data collection research studies have too stringent assumptions such as using a third-party anonymizer or a private channel amid the data owner and the collector. These studies are more susceptible to privacy attacks due to third-party involvement, which makes them less applicable for privacy-preserving healthcare data collection. This article proposes a novel privacy-preserving data collection protocol that anonymizes healthcare data without using a third-party anonymizer or a private channel for data transmission. A clustering-based k-anonymity model was adopted to efficiently prevent identity disclosure attacks, and the communication between the data owner and the collector is restricted to some elected representatives of each equivalent group of data owners. We also identified a privacy attack, known as "leader collusion", in which the elected representatives may collaborate to violate an individual's privacy. We propose solutions for such collisions and sensitive attribute protection. A greedy heuristic method is devised to efficiently handle the data owners who join or depart the anonymization process dynamically. Furthermore, we present the potential privacy attacks on the proposed protocol and theoretical analysis. Extensive experiments are conducted in real-world datasets, and the results suggest that our solution outperforms the state-of-the-art techniques in terms of privacy protection and computational complexity.

You Can't Always Get What You Want: Towards User-Controlled Privacy on Android

Mobile applications (hereafter, apps) collect a plethora of information regarding the user behavior and his device through third-party analytics libraries. However, the collection and usage of such data raised several privacy concerns, mainly because the end-user - i.e., the actual owner of the data - is out of the loop in this collection process. Also, the existing privacy-enhanced solutions that emerged in the last years follow an ”all or nothing” approach, leaving the user the sole option to accept or completely deny access to privacy-related data. This work has the two-fold objective of assessing the privacy impact of mobile analytics libraries and proposing a data anonymization methodology that offers a trade-off between the utility and privacy of the collected data and enables complete control over the sharing process. To achieve that, we present an empirical privacy assessment on the analytics libraries used in the 4500 most-used Android apps of the Google Play Store in late 2020. Then, we propose an empowered anonymization methodology, based on MobHide (Caputo <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">et al.</i> , 2020), that gives the end-user complete control over the collection and anonymization process. Finally, we empirically demonstrate the applicability and effectiveness of our solution thanks to HideDroid, a fully-fledged anonymization app for the Android ecosystem.

Methods and tools for healthcare data anonymization: a literature review

Healthcare is a rapidly evolving field. Such development creates opportunities to provide better quality, evidence-based treatment, however, increasing privacy violations. Anonymization can be applied to share data safely. This paper is a literature review of methods and tools for anonymization that includes 1930 papers, 32 of which were selected for the final evaluation. This article is an updated and extended version of the research conducted by our team in 2020 (Vovk 2021a). We found that despite the variety of methods, there is no single method that fits all cases. Certain methods are more widely used and create a background for advanced and secure methods. Our research shows that anonymization methods are used in tools for simplification and automation. Among challenges related to the use of all anonymization methods, is the proper balance between risk and data utility and highlight that the anonymization process is not limited to only technical measures but interdisciplinary.