Patient Privacy Research Articles

Factors Influencing Nurses’ Intentions to Protect Patient Privacy: A Survey-Based Study Using the Theory of Planned Behavior

Background and Purpose:Patient privacy and confidentiality are fundamental ethical principles in healthcare. Protecting patient privacy, which is accepted as a patient’s right, is one of the responsibilities of nurses. Few studies on patient privacy among nurses have used social cognitive approaches. The purpose of this study is to examine nurses’ intentions to protect patient privacy using the theory of planned behavior (TPB).Methods:This is a cross-sectional and correlational design study. The study sample consisted of 202 nurses working in the emergency departments, operating rooms, inpatient wards, and intensive care units of the hospitals. Research data were collected using a face-to-face questionnaire that included TPB components on patient privacy. The proposed research model was tested using structural equation modeling.Results:Attitude (β= .238,p< .05), subjective norm (β= .295,p< .05), and moral norm (β= .337,p< .05) toward patient privacy are positive predictors of intention. The moral norm is the most effective component of intention. Perceived behavioral control is not a significant predictor of intention (β= .049,p> .05).Implications for Practice:Norms that create a sense of moral obligation in nurses are a significant determinant in increasing the intention to protect patient privacy. Interventions that improve moral norms, attitudes, and subjective norms will increase the intention to protect privacy. Nurse managers should provide nurses with adequate skills, resources, and an appropriate work environment to ensure perceived behavioral control regarding patient privacy.

ASAP: A lightweight authenticated secure association protocol for IEEE 802.15.6 based medical BAN

Medical Body Area Networks (MBANs), a specialized subset of Wireless Body Area Networks (WBANs), are crucial for enabling medical data collection, processing, and transmission. The IEEE 802.15.6 standard governs these networks but falls short in practical MBAN scenarios. This paper introduces ASAP, a Lightweight Authenticated Secure Association Protocol integrated with IEEE 802.15.6. ASAP prioritizes patient privacy with randomized node ID generation and temporary shared keys, preventing node tracking and privacy violations. It optimizes network performance by consolidating Master Keys (MK), Pairwise Temporal Keys (PTK), and Group Temporal Keys (GTK) creation into a unified process, ensuring the efficiency of the standard four-message association protocol. ASAP enhances security by eliminating the need for pre-shared keys, reducing the attack surface, and improving forward secrecy. The protocol achieves mutual authentication without pre-shared keys or passwords and supports advanced cryptographic algorithms on nodes with limited processing capabilities. Additionally, it imposes connection initiation restrictions, requiring valid certificates for nodes, thereby addressing gaps in IEEE 802.15.6. Formal verification using Verifpal confirms ASAP's resilience against various attacks. Implementation results show ASAP's superiority over standard IEEE 802.15.6 protocols, establishing it as a robust solution for securing MBAN communications in medical environments.

Ethical dimensions of generative AI: a cross-domain analysis using machine learning structural topic modeling

Purpose The purpose of this study is to comprehensively examine the ethical implications surrounding generative artificial intelligence (AI). Design/methodology/approach Leveraging a novel methodological approach, the study curates a corpus of 364 documents from Scopus spanning 2022 to 2024. Using the term frequency-inverse document frequency (TF-IDF) and structural topic modeling (STM), it quantitatively dissects the thematic essence of the ethical discourse in generative AI across diverse domains, including education, healthcare, businesses and scientific research. Findings The results reveal a diverse range of ethical concerns across various sectors impacted by generative AI. In academia, the primary focus is on issues of authenticity and intellectual property, highlighting the challenges of AI-generated content in maintaining academic integrity. In the healthcare sector, the emphasis shifts to the ethical implications of AI in medical decision-making and patient privacy, reflecting concerns about the reliability and security of AI-generated medical advice. The study also uncovers significant ethical discussions in educational and financial settings, demonstrating the broad impact of generative AI on societal and professional practices. Research limitations/implications This study provides a foundation for crafting targeted ethical guidelines and regulations for generative AI, informed by a systematic analysis using STM. It highlights the need for dynamic governance and continual monitoring of AI’s evolving ethical landscape, offering a model for future research and policymaking in diverse fields. Originality/value The study introduces a unique methodological combination of TF-IDF and STM to analyze a large academic corpus, offering new insights into the ethical implications of generative AI across multiple domains.

Fine-grained vulnerability detection for medical sensor systems

The Internet of Things (IoT) has revolutionized the healthcare system by connecting medical sensors to the internet, while also posing challenges to the security of medical sensor networks (MSN). Given the extreme sensitivity of medical data, any vulnerability may result in data breaches and misuse, impacting patient safety and privacy. Therefore, safeguarding MSN security is critical. As medical sensor devices rely on smart healthcare software systems for data management and communication, precisely detecting system code vulnerabilities is essential to ensuring network security. Effective software vulnerability detection targets two key objectives: (i) achieving high accuracy and (ii) directly identifying vulnerable code lines for developers to fix. To address these challenges, we introduce Vulcoder, a novel vulnerability-oriented, encoder-driven model based on the Bidirectional Encoder Representations from Transformers (BERT) architecture. We propose a one-to-one mapping function to capture code semantics through abstract syntax trees (AST). Combined with multi-head attention, Vulcoder achieves precise function- and line-level detection of software vulnerabilities in MSN. This accelerates the vulnerability remediation process, thereby strengthening network security. Experimental results on various datasets demonstrate that Vulcoder outperforms previous models in identifying vulnerabilities within MSN. Specifically, it achieves a 1%–419% improvement in function-level prediction F1 scores and a 12.5%–380% increase in line-level localization precision. Therefore, Vulcoder helps enhance security defenses and safeguard patient privacy in MSN, facilitating the development of smart healthcare.

Generalizing Treatment Effect to a Target Population Without Individual Patient Data in a Real-World Setting.

The innovative use of real-world data (RWD) can answer questions that cannot be addressed using data from randomized clinical trials (RCTs). While the sponsors of RCTs have a central database containing all individual patient data (IPD) collected from trials, analysts of RWD face a challenge: regulations on patient privacy make access to IPD from all regions logistically prohibitive. In this research, we propose a double inverse probability weighting (DIPW) approach for the analysis sponsor to estimate the population average treatment effect (PATE) for a target population without the need to access IPD. One probability weighting is for achieving comparable distributions in confounders across treatment groups; another probability weighting is for generalizing the result from a subpopulation of patients who have data on the endpoint to the whole target population. The likelihood expressions for propensity scores and the DIPW estimator of the PATE can be written to only rely on regional summary statistics that do not require IPD. Our approach hinges upon the positivity and conditional independency assumptions, prerequisites to most RWD analysis approaches. Simulations are conducted to compare the performances of the proposed method against a modified meta-analysis and a regular meta-analysis.

Asthma management in the digital age.

Asthma affects 25 million people in the United States, and its prevalence is increasing. Access to care and adherence to prescribed asthma-treatment programs remain the principal formidable challenges for asthma management. Telemedicine offers substantial opportunities for improved asthma care of patients across the full range of socioeconomic strata. Ever-improving digital tools for asthma assessment and treatment are key components of telemedicine platforms for asthma management. These include a variety of remote patient-monitoring devices, digital inhaler systems, and mobile-health applications that facilitate ongoing assessment and adherence to treatment protocols. Digital tools for monitoring treatment focus on tracking medication use, inhalation technique, and physiological markers such as peak-flow rate and pulse-oximetry. Telemedicine visits allow for elements of assessment via video, approximating or duplicating many aspects of in-person visits, such as evaluating a patient's general appearance, breathing effort, and cough. Challenges remain in ensuring equitable access to these technologies, especially in rural and low-income areas, and in maintaining patient privacy and data security in digital platforms.

Enhancement of Health Services: Advanced Security Mechanism for Telemedicine in Public Health Practice

The sudden advancement of telemedicine in the healthcare industry presents potential hazards such as cyber-attacks and network susceptibilities, in addition to difficulties in patient privacy, data protection, and public health monitoring. To avoid illegal access, digital watermarking and precise medical image transfer are essential. Thus, for telemedicine in public health action, we introduce a unique security mechanism termed wavelet crypto fast decomposition with penguin optimization (WCFD-PO). Rivest-Shamir-Adelman (RSA), wavelet transforms (WT), and fast random singular value decomposition (FRSVD) encryption algorithms make up the suggested security methodology. By using the methods WT and FRSVD, one can decrease computing complexity and increase robustness. The PO approach offers a better-optimized scale factor in the fewest groups and repetitions as compared to the RSA approach, which is used for encrypting the watermarked image. The proposed mechanism is implemented by employing a Python application. Comprehensive testing indicates that our novel approach functions better than the current techniques, demonstrating the technique's higher efficiency. With the assistance of secure telemedicine solutions, this investigation offers physicians and tech programmers insightful information that will boost public health management.

Reliable generation of privacy-preserving synthetic electronic health record time series via diffusion models.

Electronic health records (EHRs) are rich sources of patient-level data, offering valuable resources for medical data analysis. However, privacy concerns often restrict access to EHRs, hindering downstream analysis. Current EHR deidentification methods are flawed and can lead to potential privacy leakage. Additionally, existing publicly available EHR databases are limited, preventing the advancement of medical research using EHR. This study aims to overcome these challenges by generating realistic and privacy-preserving synthetic EHRs time series efficiently. We introduce a new method for generating diverse and realistic synthetic EHR time series data using denoizing diffusion probabilistic models. We conducted experiments on 6 databases: Medical Information Mart for Intensive Care III and IV, the eICU Collaborative Research Database (eICU), and non-EHR datasets on Stocks and Energy. We compared our proposed method with 8 existing methods. Our results demonstrate that our approach significantly outperforms all existing methods in terms of data fidelity while requiring less training effort. Additionally, data generated by our method yield a lower discriminative accuracy compared to other baseline methods, indicating the proposed method can generate data with less privacy risk. The proposed model utilizes a mixed diffusion process to generate realistic synthetic EHR samples that protect patient privacy. This method could be useful in tackling data availability issues in the field of healthcare by reducing barrier to EHR access and supporting research in machine learning for health. The proposed diffusion model-based method can reliably and efficiently generate synthetic EHR time series, which facilitates the downstream medical data analysis. Our numerical results show the superiority of the proposed method over all other existing methods.

Study of Patients' Privacy during the COVID-19 Pandemic in Iranian Health Care Settings.

Patient's privacy protection is a challenging ethical issue. The complex situation of the COVID-19 pandemic was a probable predictor of breaching confidentiality. This study aimed to assess the viewpoints of COVID-19-confirmed patients, who were hospitalized, and their healthcare providers about the compliance of different aspects of patient's privacy. This cross-sectional study included 3433 COVID-19-confirmed patients who were hospitalized in Kerman, between 2020 and 2021, and about 1228 related physicians, nurses, and paraclinical staff. Two separate validated researcher-made questionnaires were developed, each including subscales for physical, informational, and spatial privacy, as well as a satisfaction rate of privacy protection. The data were analyzed using SPSS software version 26, with independent samples t test, Mann-Whitney-U, Kruskal Wallis, and Multiple Linear Regression tests at a 95% confidence interval. The mean percentages of the patients' privacy scores in physical, spatial, and informational areas were significantly lower (P<0.001) than the average of the medical staff's scores in all three areas (Difference: 10.27%, 14.83%, and 4.91%, respectively). Physical and spatial privacy scores could be predicted based on the participants' classification, patients or medical staff, and sex. The mean patients' satisfaction score was 9.25% lower than the medical staff's (P<0.001). Moreover, only academic hospitals showed a statistically significant difference between the patient's satisfaction with privacy protection and medical staff's viewpoints (P<0.001). Although this study indicated the benefits of protecting patients' privacy in the healthcare setting, patients' privacy scores and satisfaction were lower than their healthcare providers. The pandemic conditions might have been an obstacle to preserving patients' rights. These findings demonstrated the importance of sensitizing healthcare providers to manage these ethical challenges in a complicated critical state such as the COVID-19 pandemic.

Cross-domain additive learning of new knowledge rather than replacement.

In medical clinical scenarios for reasons such as patient privacy, information protection and data migration, when domain adaptation is needed for real scenarios, the source-domain data is often inaccessible and only the pre-trained source model on the source-domain is available. Existing solutions for this type of problem tend to forget the rich task experience previously learned on the source domain after adapting, which means that the model simply overfits the target-domain data when adapting and does not learn robust features that facilitate real task decisions. We address this problem by exploring the particular application of source-free domain adaptation in medical image segmentation and propose a two-stage additive source-free adaptation framework. We generalize the domain-invariant features by constraining the core pathological structure and semantic consistency between different perspectives. And we reduce the segmentation generated by locating and filtering elements that may have errors through Monte-Carlo uncertainty estimation. We conduct comparison experiments with some other methods on a cross-device polyp segmentation and a cross-modal brain tumor segmentation dataset, the results in both the target and source domains verify that the proposed method can effectively solve the domain offset problem and the model retains its dominance on the source domain after learning new knowledge of the target domain.This work provides valuable exploration for achieving additive learning on the target and source domains in the absence of source data and offers new ideas and methods for adaptation research in the field of medical image segmentation.

Explainable Artificial Intelligence for Early Prediction of Pressure Injury Risk.

Hospital-acquired pressure injuries (HAPIs) have a major impact on patient outcomes in intensive care units (ICUs). Effective prevention relies on early and accurate risk assessment. Traditional risk-assessment tools, such as the Braden Scale, often fail to capture ICU-specific factors, limiting their predictive accuracy. Although artificial intelligence models offer improved accuracy, their "black box" nature poses a barrier to clinical adoption. To develop an artificial intelligence-based HAPI risk-assessment model enhanced with an explainable artificial intelligence dashboard to improve interpretability at both the global and individual patient levels. An explainable artificial intelligence approach was used to analyze ICU patient data from the Medical Information Mart for Intensive Care. Predictor variables were restricted to the first 48 hours after ICU admission. Various machine-learning algorithms were evaluated, culminating in an ensemble "super learner" model. The model's performance was quantified using the area under the receiver operating characteristic curve through 5-fold cross-validation. An explainer dashboard was developed (using synthetic data for patient privacy), featuring interactive visualizations for in-depth model interpretation at the global and local levels. The final sample comprised 28 395 patients with a 4.9% incidence of HAPIs. The ensemble super learner model performed well (area under curve = 0.80). The explainer dashboard provided global and patient-level interactive visualizations of model predictions, showing each variable's influence on the risk-assessment outcome. The model and its dashboard provide clinicians with a transparent, interpretable artificial intelligence-based risk-assessment system for HAPIs that may enable more effective and timely preventive interventions.

Beyond human perception: Revolutionizing ophthalmology with artificial intelligence and deep learning

The purpose of the study was to provide a comprehensive overview of the transformative applications of artificial intelligence (AI) in ophthalmology, with a focus on its impact on screening, diagnosis, and treatment planning. A comprehensive literature search was conducted to identify relevant studies on the applications of AI in ophthalmology. PubMed, Embase, and Scopus were searched using appropriate keywords, with inclusion criteria focusing on studies related to image analysis, diagnostic algorithms, predictive models, and treatment planning. Limited to English-language articles, both original research and review articles were considered, while studies emphasizing nonophthalmic applications of AI or lacking sufficient detail were excluded. AI algorithms, powered by deep learning models, have demonstrated remarkable accuracy in the automated screening and detection of various ocular diseases. The potential implications of AI include revolutionizing screening programs for early identification of individuals at risk, facilitating timely interventions, and improving patient outcomes. The integration of AI with teleophthalmology and remote monitoring systems has the potential to alleviate the burden on health-care systems, particularly in underserved areas. The applications of AI in ophthalmology hold significant potential for transforming the field by enhancing diagnostic accuracy, optimizing treatment strategies, and increasing access to eye care. However, successful implementation requires addressing challenges such as diverse and representative datasets, ensuring interpretability and explainability of AI models, and addressing ethical considerations related to patient privacy and data security. Collaborative efforts between ophthalmologists, data scientists, and regulatory bodies are deemed crucial to fully leverage the potential of AI in ophthalmology.

Challenges and recommendations in the implementation of audiovisual telemedicine communication: a systematic review.

This systematic review aims to identify the elements of doctor-patient communication in telemedicine, emerging challenges, and proposed recommendations. Four databases, including Science Direct, PubMed, Cochrane, and ProQuest, were searched using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The inclusion criteria consisted of original research papers, availability of free full text, and publications during the past 10 years. A total of 13 articles completed the selection process and satisfied the established criteria. The issues and recommendations of telemedicine communication were categorized into three distinct groups: pre-consultation, during-consultation, and post-consultation. Preparation encompasses the arranging of visual elements, safeguarding patient privacy and confidentiality, and addressing any technical challenges that may arise. The consultation encompasses nonverbal behavior, empathy, the doctor-patient connection, and a physical examination. Post-telemedicine consultations refer to medical appointments that occur after a telemedicine session, typically involving follow-up medical interactions. Telemedicine presents unique challenges in doctor-patient consultations that differ from face-to-face interactions. Therefore, clinicians must acquire communication skills specific to telemedicine to ensure effective consultations and achieve optimal health results.

Healthcare Provider's Preferences on Open Versus Restricted Visiting Hours in Surgical Intensive Care Unit.

The increasing consideration of open visiting hours in intensive care units (ICUs) has presented implementation challenges. Recognizing the invaluable insights of healthcare providers, this study aimed to assess their perspectives on the potential impact of altering ICU visiting policies on patient care. In a single-institution, cross-sectional study, a five-point Likert survey was administered to healthcare providers working in a surgical ICU to assess their perceptions of open and restricted visiting hours. Descriptive statistics were used to analyze the responses. Ninety-three healthcare providers participated, including nurses (53%), resident physicians (23%), and other healthcare professionals. The majority (81%) did not believe that open ICU visiting hours would enhance patient-centered care. Providers expressed concerns that open visitation could lead to increased provider stress (89%), physical barriers to patient access (80%), disruption in workflow (90%), and provider burnout (76%). Conversely, they perceived that restricted visitation would preserve patient privacy (87%), promote better healing for patients (64%), reduce provider stress (81%), and improve workflow (95%). Nurses were more likely than resident physicians to perceive that open ICU hours would alleviate familial stress (p=0.022). Additionally, both nurses and other healthcare providers recognized the potential for open ICU hours to improve communication between patient families and healthcare teams (p<0.05). Notably, there were no significant differences among providers in perceptions of the impact of open versus restricted visitation on provider stress, workflow, or other factors (p>0.05). This survey reveals a negative perception of open ICU visiting hours among healthcare providers in a surgical ICU, with concerns that such a change could increase stress for both patients and providers. These findings underscore the potential challenges and negative impacts of implementing open visiting hours in this setting, emphasizing the need for careful consideration and further research to inform policy decisions and visitation practices.

Genomic Data Fusion using Paillier Cryptosystem

The proposed work performs secure data fusion using homomorphic encryption, specifically the Paillier cryptosystem. The Paillier cryptosystem allows computation to be performed on encrypted data without decrypting it first, thus ensuring the privacy and security of the computation. The experiment measures the algorithm's performance based on execution time, memory usage, security, accuracy, and scalability. The data-level Paillier cryptosystem approach is generally slower than the feature-level fusion method due to its more complex operations and computations. Scalability is limited by the time required for encryption, homomorphic addition, and decryption. Improving scalability can be achieved by parallelizing the encryption and decryption steps, optimizing the homomorphic addition algorithm, or using more efficient cryptographic primitives. The article compares the performance of the Paillier cryptosystem with differential privacy in terms of their advantages and disadvantages. By adopting a preemptive approach to data fusion security, healthcare organizations can minimize the risk of data breaches and protect patient privacy. Data fusion security is an important factor when dealing with medical records. In the field of medical records, data fusion refers to the method of combining multiple sources of data into a distinct record. This can include data from electronic health records (EHRs), medical imaging devices, wearable devices, and other sources. There are several security considerations that must be addressed when fusing data from multiple sources.

Privacy in Community Pharmacies in Saudi Arabia: A Cross-Sectional Study.

Privacy in healthcare is a fundamental right essential to maintain patient confidentiality and trust. Community pharmacies in Saudi Arabia (SA) play a critical role in the healthcare system by providing accessible services and serving as initial points of contact for medical advice. However, the open nature of these settings poses significant challenges in maintaining patient privacy. This cross-sectional study used electronic surveys distributed across various online platforms. The target sample included Saudi adults, with a sample size of 385 participants to achieve 80% statistical power at a 95% confidence interval. The survey comprised demographic questions and sections evaluating perceptions of privacy, the importance of privacy, and personal experiences regarding privacy in community pharmacies. Descriptive statistics and logistic regression models were used for the analysis. A total of 511 responses were obtained. The mean age was 33.5 years, with an almost equal distribution of males (49.71%) and females (50.29%). Most participants held a bachelor's degree or higher (78.67%). Privacy perceptions varied, with only 9.0% strongly agreeing that there was a private space for consultations, while 64.0% felt that the design of community pharmacies did not adequately consider patient privacy, and 86.9% reported that conversations could be overheard. Privacy concerns were notable, with almost one-half of the participants (47.6%) having concerns about privacy and 56.6% doubting the confidentiality of their health information. Moreover, 17.6% reported being asked for unnecessary personal information when buying medication, and 56.2% admitted to avoiding discussing a health problem with the pharmacist due to privacy concerns. Experiences of privacy breaches were reported by 15.7% of respondents. Logistic regression analysis revealed that the availability of private space in the pharmacy and patients feeling that the pharmacy respects their privacy were associated with a lower likelihood of avoiding discussions with pharmacists due to privacy concerns (OR = 0.758, CI = 0.599-0.0957 and OR = 0.715, CI = 0.542-0.945 respectively) Conversely, greater privacy concerns and previous privacy breaches significantly increased the likelihood of avoiding discussions with pharmacists in the community pharmacy (OR = 1.657, CI = 1.317-2.102 and OR = 4.127, CI = 1.886-9.821 respectively). This study highlights the significant concerns regarding privacy practices in community pharmacies in SA. Thus, there is a need for standards to improve privacy in community pharmacies, such as mandating the need for private consultation areas and enhanced staff training on handling privacy-related issues. Addressing the issue of privacy is crucial for maintaining patient trust, improving healthcare service quality, and ensuring effective patient-pharmacist interactions.

My Holistic Data Share: A WEB3 Data Share Application: Extending Beyond Finance to Privacy-Protected Decentralized Share of Multi-Dimensional Data to Enhance Global Healthcare

Abstract WEB3.0 technologies on network architectures, distributed ledgers, and decentralized artificial intelligence represents a transformative shift in how data are handled, stored and shared. These innovations promise to significantly enhance consumer data privacy rights by addressing fundamental vulnerabilities associated with traditional centralized systems and self-custody wallets. Data breaches in traditional systems operated mainly by third parties are more common, resulting in significant data leaks because of centralized storage and excessive data movement, sometimes unnecessarily. Healthcare data breaches have been a growing concern globally. Several hospitals faced operational halts on account of the impact of ransomware on patient care and privacy. WEB3 Wallets are a vital component emerging as a significant force for global financial inclusion, especially in developing economies. They promote inclusion, reduce costs and empower individuals through self-custody. Though major improvements are needed in these wallets, their use is rising steadily. The global cryptocurrency user base is expected to reach over 500 million by 2025, with substantial growth in emerging markets, according to a report by Statista in 2023. This paper introduces a concept beyond cryptocurrencies and finance into everyday real-world use cases that need combinatorial access to a person’s holistic data, including financial and health records, genomic data, and advanced directives, among others, that need to be privacy protected and shared with specific actors identified for their roles in the WEB3 ecosystem through decentralized identifiers and non-fungible token badges identifying particular recipients. The author introduced the concept at ETHBoston in April 2024, won accolades for a primitive implementation using underlying threshold cryptography technologies, and enhanced it into a conceptual holistic data share application for global healthcare, as presented in this paper.

AI Governance in Healthcare: Best Practices, Solutions, and Unresolved Issues

What challenges are presented by AI adoption in healthcare? What role should AI governance play in regard to the opportunities and risks for generative AI (large language models like ChatGPT-4, etc.) in health? How do we responsibly evaluate new technology and ensure guardrails for its deployment, but without stifling innovation? How do we address societal fears and biases? What level of reliability and robustness should algorithms achieve before being adopted with confidence across healthcare? How can AI support health equity and expand access? We will explore these questions and more with some of the nation’s leading experts on these topics. Learning Objectives Understand key opportunities and challenges for AI governance in healthcare, particularly in regard to rapidly emerging generative AI applications Discuss various responsible AI and AI governance frameworks that seek to ensure safety and efficacy in AI adoption Explore the role of government, legislation, and regulators in supporting safe and equitable adoption of AI in healthcare Explore opportunities for government, industry, and academic partnerships to accelerate AI validation, development, and adoption of safety standards Consider the potential impact of AI on the healthcare workforce and provider and patient experiences Identify the risks we should be most concerned with and what approaches can we take to mitigate those risks? By the end of the presentation, participants should have a sense of the opportunities and challenges of adopting AI in healthcare, key risks and frameworks for mitigating them, and the potential role of government, academia, and industry in addressing these issues Generative AI application examples Patient Communication and Engagement Intelligent chatbots interacting with patients, answering their queries, providing health advice, and helping them manage their conditions Mental Health Support Provide psychological support to patients, helping them manage stress, anxiety, depression, and other mental health issues. Provide interventions based on CBT and other therapeutic techniques Automated Medical Documentation Note-taking can allowing doctors to focus more on patient care Education and Training Create educational content and personalized training for medical students and healthcare professionals, potentially simulating things like patient cases and providing guidance on diagnosis, etc. Synthesis and Summarization of Medical Research Make it easier for people to keep up with the latest developments Public Health Communication In a public health emergency, help craft health messages, FAQs, and timely and accurate updates for the public Generative AI challenges and limitations Ensuring the accuracy of AI-generated information Maintaining patient privacy Handling the ethical implications of AI and human-machine interactions in healthcare “Hallucinations” (false or inaccurate information) can occur when model has limited knowledge of a particular domain “Distraction” (losing track of earlier inputs) can occur when inputs go beyond allowable token count for model “context window” and models start to produce responses not relevant to queries and context.

Improving privacy-preserving multi-faceted long short-term memory for accurate evaluation of encrypted time-series MRI images in heart disease

In therapeutic diagnostics, early diagnosis and monitoring of heart disease is dependent on fast time-series MRI data processing. Robust encryption techniques are necessary to guarantee patient confidentiality. While deep learning (DL) algorithm have improved medical imaging, privacy and performance are still hard to balance. In this study, a novel approach for analyzing homomorphivally-encrypted (HE) time-series MRI data is introduced: The Multi-Faceted Long Short-Term Memory (MF-LSTM). This method includes privacy protection. The MF-LSTM architecture protects patient’s privacy while accurately categorizing and forecasting cardiac disease, with accuracy (97.5%), precision (96.5%), recall (98.3%), and F1-score (97.4%). While segmentation methods help to improve interpretability by identifying important region in encrypted MRI images, Generalized Histogram Equalization (GHE) improves image quality. Extensive testing on selected dataset if encrypted time-series MRI images proves the method’s stability and efficacy, outperforming previous approaches. The finding shows that the suggested technique can decode medical image to expose visual representation as well as sequential movement while protecting privacy and providing accurate medical image evaluation.

Artificial intelligence in psychodermatology: A brief report of applications and impact in clinical practice.

This report evaluates the potential of artificial intelligence (AI) in psychodermatology, emphasizing its ability to enhance diagnostic accuracy, treatment efficacy, and personalized care. Psychodermatology, which explores the connection between mental health and skin disorders, stands to benefit from AI's advanced data analysis and pattern recognition capabilities. A literature search was conducted on PubMed and Google Scholar, spanning from 2004 to 2024, following PRISMA guidelines. Studies included demonstrated AI's effectiveness in predicting treatment outcomes for body dysmorphic disorder, identifying biomarkers in psoriasis and anxiety disorders, and refining therapeutic strategies. The review identified several studies highlighting AI's role in improving treatment outcomes and diagnostic accuracy in psychodermatology. AI was effective in predicting outcomes for body dysmorphic disorder and identifying biomarkers related to psoriasis and anxiety disorders. However, challenges such as limited dermatologist knowledge, integration difficulties, and ethical concerns regarding patient privacy were noted. AI holds significant promise for advancing psychodermatology by improving diagnostic precision, treatment effectiveness, and personalized care. Nonetheless, realizing this potential requires large-scale clinical validation, enhanced dataset diversity, and robust ethical frameworks. Future research should focus on these areas, with interdisciplinary collaboration essential for overcoming current challenges and optimizing patient care in psychodermatology.