Fair Algorithm Research Articles

Large Language Models in Otolaryngology Residency Admissions: A Random Sampling Analysis.

To investigate potential demographic bias in artificial intelligence (AI)-based simulations of otolaryngology, residency selection committee (RSC) members tasked with selecting one applicant among candidates with varied racial, gender, and sexual orientations. This study employed random sampling of simulated RSC member decisions using a novel Application Programming Interface (API) to virtually connect to OpenAI's Generative Pre-Trained Transformers (GPT-4 and GPT-4o). Simulated RSC members with diverse demographics were tasked with ranking to match 1 applicant among 10 with varied racial, gender, and sexual orientations. All applicants had identical qualifications; only demographics of the applicants and RSC members were varied for each simulation. Each RSC simulation ran 1000 times. Chi-square tests analyzed differences across categorical variables. GPT-4o simulations additionally requested a rationale for each decision. Simulated RSCs consistently showed racial, gender, and sexual orientation bias. Most applicant pairwise comparisons showed statistical significance (p < 0.05). White and Black RSCs exhibited greatest preference for applicants sharing their own demographic characteristics, favoring White and Black female applicants, respectively, over others (all pairwise p < 0.001). Asian male applicants consistently received lowest selection rates. Male RSCs favored White male and female applicants, while female RSCs preferred LGBTQIA+, White and Black female applicants (all p < 0.05). High socioeconomic status (SES) RSCs favored White female and LGBTQIA+ applicants, while low SES RSCs favored Black female and LGBTQIA+ applicants over others (all p < 0.001). Results from the newest iteration of the LLM, ChatGPT-4o, indicated evolved selection preferences favoring Black female and LGBTQIA+ applicants across all RSCs, with the rationale of prioritizing inclusivity given in >95% of such decisions. Utilizing publicly available LLMs to aid in otolaryngology residency selection may introduce significant racial, gender, and sexual orientation bias. Potential for significant and evolving LLM bias should be appreciated and minimized to promote a diverse and representative field of future otolaryngologists in alignment with current workforce data. N/A Laryngoscope, 2024.

Leveraging Artificial Intelligence to Optimize Transcranial Direct Current Stimulation for Long COVID Management: A Forward-Looking Perspective.

Long COVID (Coronavirus disease), affecting millions globally, presents unprecedented challenges to healthcare systems due to its complex, multifaceted nature and the lack of effective treatments. This perspective review explores the potential of artificial intelligence (AI)-guided transcranial direct current stimulation (tDCS) as an innovative approach to address the urgent need for effective Long COVID management. The authors examine how AI could optimize tDCS protocols, enhance clinical trial design, and facilitate personalized treatment for the heterogeneous manifestations of Long COVID. Key areas discussed include AI-driven personalization of tDCS parameters based on individual patient characteristics and real-time symptom fluctuations, the use of machine learning for patient stratification, and the development of more sensitive outcome measures in clinical trials. This perspective addresses ethical considerations surrounding data privacy, algorithmic bias, and equitable access to AI-enhanced treatments. It also explores challenges and opportunities for implementing AI-guided tDCS across diverse healthcare settings globally. Future research directions are outlined, including the need for large-scale validation studies and investigations of long-term efficacy and safety. The authors argue that while AI-guided tDCS shows promise for addressing the complex nature of Long COVID, significant technical, ethical, and practical challenges remain. They emphasize the importance of interdisciplinary collaboration, patient-centered approaches, and a commitment to global health equity in realizing the potential of this technology. This perspective article provides a roadmap for researchers, clinicians, and policymakers involved in developing and implementing AI-guided neuromodulation therapies for Long COVID and potentially other neurological and psychiatric conditions.

Transforming the Judicial System: The Impact of Machine Learning on Legal Processes and Outcomes

This paper explores the transformative impact of Machine Learning (ML) on legal processes and outcomes within the judicial system, aiming to analyze current applications, evaluate benefits and challenges, and assess ethical and practical implications. A comprehensive methodology is employed, including a literature review of academic articles, industry reports, and legal documents, case studies of specific ML tools like Correctional Offender Management Profiling for Alternative Sanctions (COMPAS) and ROSS Intelligence, and interviews with legal professionals, data scientists, and ethicists. Key findings reveal that ML enhances efficiency, accuracy, and accessibility in legal processes, with applications in recidivism prediction, legal research, e-discovery, and online dispute resolution. For instance, tools such as COMPAS aid in predicting recidivism rates, while ROSS Intelligence and CaseText streamline legal research through Natural Language Processing (NLP). However, significant challenges arise, particularly concerning data privacy, algorithmic bias, and the ethical implications of automated decision-making, as evidenced by criticisms of the COMPAS system for potential racial bias. The research underscores the necessity of interdisciplinary collaboration between legal experts and data scientists to develop robust and legally sound ML tools. The implications of these findings are profound, calling for clear policies and regulations to ensure transparency and fairness, the development of ethical frameworks to address bias and privacy concerns, and the provision of education and training for legal professionals in data science and ML. The paper concludes that continuous research and refinement of algorithms are essential to address emerging challenges and expand the beneficial applications of ML in the legal field, contributing to a more efficient, equitable, and accessible judicial system.

Use of Artificial Intelligence in Medical Devices for Post-Market Surveillance

Artificial intelligence (AI) has emerged as a transformative tool in post-market surveillance (PMS) for monitoring the safety and performance of medical products. This article explores the role of AI in optimizing PMS practices, focusing on its applications in signal detection, risk assessment, and regulatory compliance. By harnessing machine learning algorithms and big data analytics, AI facilitates the automated analysis of real-world evidence, including patient outcomes data, adverse event reports, and electronic health records. Through pattern recognition and anomaly detection, AI algorithms enable the early identification of potential safety issues and facilitate timely interventions to mitigate risks. Moreover, AI-driven PMS systems enhance regulatory oversight by providing regulators with comprehensive and actionable insights into product safety profiles and emerging trends. However, concerns regarding data privacy, algorithm bias, and interpretability underscore the need for transparent and ethically responsible AI deployment in PMS frameworks.

Innovation and challenges of artificial intelligence technology in personalized healthcare

As the burgeoning field of Artificial Intelligence (AI) continues to permeate the fabric of healthcare, particularly in the realms of patient surveillance and telemedicine, a transformative era beckons. This manuscript endeavors to unravel the intricacies of recent AI advancements and their profound implications for reconceptualizing the delivery of medical care. Through the introduction of innovative instruments such as virtual assistant chatbots, wearable monitoring devices, predictive analytic models, personalized treatment regimens, and automated appointment systems, AI is not only amplifying the quality of care but also empowering patients and fostering a more interactive dynamic between the patient and the healthcare provider. Yet, this progressive infiltration of AI into the healthcare sphere grapples with a plethora of challenges hitherto unseen. The exigent issues of data security and privacy, the specter of algorithmic bias, the requisite adaptability of regulatory frameworks, and the matter of patient acceptance and trust in AI solutions demand immediate and thoughtful resolution .The importance of establishing stringent and far-reaching policies, ensuring technological impartiality, and cultivating patient confidence is paramount to ensure that AI-driven enhancements in healthcare service provision remain both ethically sound and efficient. In conclusion, we advocate for an expansion of research efforts aimed at navigating the ethical complexities inherent to a technology-evolving landscape, catalyzing policy innovation, and devising AI applications that are not only clinically effective but also earn the trust of the patient populace. By melding expertise across disciplines, we stand at the threshold of an era wherein AI's role in healthcare is both ethically unimpeachable and conducive to elevating the global health quotient.

Harnessing the Potential of Artificial Intelligence in Managing Viral Hepatitis

Viral hepatitis continues to be a serious global health concern, impacting millions of people, putting a strain on healthcare systems across the world, and causing significant morbidity and mortality. Traditional diagnostic, prognostic, and therapeutic procedures to address viral hepatitis are successful but have limits in accuracy, speed, and accessibility. Artificial intelligence (AI) advancement provides substantial opportunities to overcome these challenges. This study investigates the role of AI in revolutionizing viral hepatitis care, from early detection to therapy optimization and epidemiological surveillance. A comprehensive literature review was conducted using predefined keywords in the Nature, PLOS ONE, PubMed, Frontiers, Wiley Online Library, BMC, Taylor &amp; Francis, Springer, ScienceDirect, MDPI, IEEE Xplore Digital Library, and Google Scholar databases. Peer-reviewed publications written in English between January 2019 and August 2024 were examined. The data of the selected research papers were synthesized and analyzed using thematic and narrative analysis techniques. The use of AI-driven algorithms in viral hepatitis control involves many significant aspects. AI improves diagnostic accuracy by integrating machine learning (ML) models with serological, genomic, and imaging data. It enables tailored treatment plans by assessing patient-specific characteristics and predicting therapy responses. AI-powered technologies aid in epidemiological modeling, and AI-powered systems effectively track treatment adherence, identify medication resistance, and control complications associated with chronic hepatitis infections. It is vital in identifying new antiviral medicines and vaccines, speeding the development pipeline through high-throughput screening and predictive modeling. Despite its transformational promise, using AI in viral hepatitis care presents various challenges, including data privacy concerns, the necessity for extensive and varied datasets, and the possibility of algorithmic biases. Ethical considerations, legal frameworks, and multidisciplinary collaboration are required to resolve these issues and ensure AI technology’s safe and successful use in clinical practice. Exploiting the full AI’s potential for viral hepatitis management provides unparalleled prospects to improve patient outcomes, optimize public health policies, and, eventually, and alleviate the disease’s negative impact worldwide. This study seeks to provide academics, medics, and policymakers with the fundamental knowledge they need to harness AI’s potential in the fight against viral hepatitis.

Predicting serious postoperative complications and evaluating racial fairness in machine learning algorithms for metabolic and bariatric surgery

BackgroundPredicting the risk of complications is critical in metabolic and bariatric surgery (MBS). ObjectivesTo develop machine learning (ML) models to predict serious postoperative complications of MBS and evaluate racial fairness of the models. SettingMetabolic and Bariatric Surgery Accreditation and Quality Improvement Program (MBSAQIP) national database, United States. MethodsWe developed logistic regression, random forest (RF), gradient-boosted tree (GBT), and XGBoost model using the MBSAQIP Participant Use Data File from 2016 to 2020. To address the class imbalance, we randomly undersampled the complication-negative class to match the complication-positive class. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC), precision, recall, and F1 score. Fairness across White and non-White patient groups was assessed using equal opportunity difference and disparate impact metrics. ResultsA total of 40,858 patients were included after undersampling the complication-negative class. The XGBoost model was the best-performing model in terms of AUROC; however, the difference was not statistically significant. While the F1 score and precision did not vary significantly across models, the RF exhibited better recall compared to the logistic regression. Surgery type was the most important feature to predict complications, followed by operative time. The logistic regression model had the best fairness metrics for race. ConclusionsThe XGBoost model achieved the highest AUROC, albeit without a statistically significant difference. The RF may be useful when recall is the primary concern. Undersampling of the privileged group may improve the fairness of boosted tree models.

Ethical considerations in AI use for SMEs and supply chains: Current challenges and future directions

The integration of Artificial Intelligence (AI) in Small and Medium-sized Enterprises (SMEs) and supply chains has revolutionized operational efficiencies and decision-making processes. This review paper aims to explore the ethical implications of AI adoption in these sectors, identifying current challenges and proposing future directions for ethical AI deployment. Through an extensive review of existing literature, the paper examines key ethical concerns such as data privacy, bias in AI algorithms, transparency, and the socio-economic impact on the workforce. The findings indicate that SMEs face unique ethical challenges due to their limited resources and expertise, which exacerbate issues related to AI implementation. Additionally, supply chains grapple with transparency and accountability, necessitating immediate attention to ensure ethical practices. The review concludes that establishing a robust ethical framework is crucial for guiding AI integration in SMEs and supply chains. It recommends the development of standardized ethical guidelines, enhanced stakeholder engagement, and increased investment in AI literacy and infrastructure. Future research should focus on creating adaptable ethical models that evolve alongside technological advancements and industry needs, ensuring that AI contributes to sustainable and equitable growth. This paper contributes to the ongoing discourse on ethical AI, offering actionable insights for policymakers, business leaders, and researchers dedicated to fostering responsible AI practices in SMEs and supply chains. Keywords: Ethical AI, SMEs (Small and Medium-sized Enterprises), Supply Chains, Algorithmic Bias, Data Privacy, Transparency, Accountability, AI Trust, Regulatory Compliance, Stakeholder Engagement, Job Displacement, Explainable AI (XAI), Workforce Reskilling, Sustainable AI Practices, AI Ethics Frameworks.

Twenty-Five Years of Progress-Lessons Learned From JMIR Publications to Address Gender Parity in Digital Health Authorships: Bibliometric Analysis.

Digital health research plays a vital role in advancing equitable health care. The diversity of research teams is thereby instrumental in capturing societal challenges, increasing productivity, and reducing bias in algorithms. Despite its importance, the gender distribution within digital health authorship remains largely unexplored. This study aimed to investigate the gender distribution among first and last authors in digital health research, thereby identifying predicting factors of female authorship. This bibliometric analysis examined the gender distribution across 59,980 publications from 1999 to 2023, spanning 42 digital health journals indexed in the Web of Science. To identify strategies ensuring equality in research, a detailed comparison of gender representation in JMIR journals was conducted within the field, as well as against a matched sample. Two-tailed Welch 2-sample t tests, Wilcoxon rank sum tests, and chi-square tests were used to assess differences. In addition, odds ratios were calculated to identify predictors of female authorship. The analysis revealed that 37% of first authors and 30% of last authors in digital health were female. JMIR journals demonstrated a higher representation, with 49% of first authors and 38% of last authors being female, yielding odds ratios of 1.96 (95% CI 1.90-2.03; P<.001) and 1.78 (95% CI 1.71-1.84; P<.001), respectively. Since 2008, JMIR journals have consistently featured a greater proportion of female first authors than male counterparts. Other factors that predicted female authorship included having female authors in other relevant positions and gender discordance, given the higher rate of male last authors in the field. There was an evident shift toward gender parity across publications in digital health, particularly from the publisher JMIR Publications. The specialized focus of its sister journals, equitable editorial policies, and transparency in the review process might contribute to these achievements. Further research is imperative to establish causality, enabling the replication of these successful strategies across other scientific fields to bridge the gender gap in digital health effectively.

Addressing bias in bagging and boosting regression models.

As artificial intelligence (AI) becomes widespread, there is increasing attention on investigating bias in machine learning (ML) models. Previous research concentrated on classification problems, with little emphasis on regression models. This paper presents an easy-to-apply and effective methodology for mitigating bias in bagging and boosting regression models, that is also applicable to any model trained through minimizing a differentiable loss function. Our methodology measures bias rigorously and extends the ML model's loss function with a regularization term to penalize high correlations between model errors and protected attributes. We applied our approach to three popular tree-based ensemble models: a random forest model (RF), a gradient-boosted model (GBT), and an extreme gradient boosting model (XGBoost). We implemented our methodology on a case study for predicting road-level traffic volume, where RF, GBT, and XGBoost models were shown to have high accuracy. Despite high accuracy, the ML models were shown to perform poorly on roads in minority-populated areas. Our bias mitigation approach reduced minority-related bias by over 50%.

Can large language models estimate public opinion about global warming? An empirical assessment of algorithmic fidelity and bias

Large language models (LLMs) can be used to estimate human attitudes and behavior, including measures of public opinion, a concept referred to as algorithmic fidelity. This study assesses the algorithmic fidelity and bias of LLMs in estimating public opinion about global warming. LLMs were conditioned on demographics and/or psychological covariates to simulate survey responses. Findings indicate that LLMs can effectively reproduce presidential voting behaviors but not global warming opinions unless the issue relevant covariates are included. When conditioned on both demographic and covariates, GPT-4 demonstrates improved accuracy, ranging from 53% to 91%, in predicting beliefs and attitudes about global warming. Additionally, we find an algorithmic bias that underestimates the global warming opinions of Black Americans. While highlighting the potential of LLMs to aid social science research, these results underscore the importance of conditioning, model selection, survey question format, and bias assessment when employing LLMs for survey simulation.

The Role of Artificial Intelligence in Achieving the United Nations Sustainable Development Goals

The United Nations&amp;#39; 2030 Agenda for Sustainable Development aims to tackle poverty, inequality, and environmental degradation and foster economic growth. This study investigates the transformative potential of artificial intelligence (AI) in achieving these Sustainable Development Goals (SDGs). Analyzing data from 44 sources, the research highlights AI&amp;#39;s capacity to address critical challenges in healthcare, education, environmental management, economic growth, and gender equality. AI applications in renewable energy, waste management, disease detection, personalized education, and gender equality are examined. The study also emphasizes the ethical issues associated with AI, such as algorithmic bias, data privacy breaches, and job displacement. To fully leverage AI&amp;#39;s potential, it is essential to develop intelligent automation governance systems, foster interdisciplinary research combining AI and sustainability, and promote public-private partnerships. Additionally, enhancing public AI literacy and implementing eco-friendly AI policies are crucial. The study advocates for a holistic ethical framework to maximize AI&amp;#39;s benefits while mitigating risks, promoting cross-disciplinary collaboration, and establishing ethical AI standards. By doing so, AI can significantly contribute to a more inclusive, equitable, and sustainable future.

Exploring AI's Role in Supporting Diversity and Inclusion Initiatives in Multicultural Marketplaces

This paper explores the relationship between diversity, inclusion, and artificial intelligence (AI) in multicultural workplaces and markets. Growing recognition of the value of diversity and inclusion (D&amp;I) initiatives is changing organizational practices globally. Despite progress, challenges such as unconscious bias, resistance to change, and difficulty in measuring progress still exist.This study is based on a comprehensive literature review that adheres to academic ethical standards. These findings highlight the potential of AI to improve recruitment processes, personalize marketing strategies, and develop inclusive technologies. However, challenges such as bias in AI algorithms and limited data diversity need to be addressed.The paper emphasizes the need for ethical AI frameworks, diverse representation in AI development, and transparency in the use of AI to promote justice. AI has the potential to foster a more equitable society by encouraging diversity and inclusion. Further research is needed to explore the long-term impact of AI on diversity and inclusion and to develop ethical AI frameworks tailored to different industries and cultural contexts.The study also explores how AI can support D&amp;I initiatives by providing data-driven insights, automating processes, and offering personalized interventions to reduce bias and increase inclusivity. By reviewing recent research and case studies, this paper offers practical recommendations for leveraging AI in D&amp;I efforts, aiming to build a more inclusive and equitable workplace culture.

Reinvention mediates impacts of skin tone bias in algorithms: implications for technology diffusion

Abstract Two studies examine how skin tone bias in image recognition algorithms impacts users’ adoption and usage of image recognition technology. We employed a diffusion of innovations framework to explore perceptions of compatibility, complexity, observability, relative advantage, and reinvention to determine their influence on participants' utilization of image recognition algorithms. Despite being more likely to encounter algorithm bias, individuals with darker skin tones perceived image recognition algorithms as having greater levels of compatibility and relative advantage, being more observable, and less complex and thus used them more extensively compared to those with lighter skin tones. Individuals with darker skin tones also displayed higher levels of reinvention behaviors, suggesting a potential adaptive response to counteract algorithm biases.

Identifying and Addressing Bias in Artificial Intelligence

Identifying and Addressing Bias in Artificial Intelligence

AI-powered revolution in plant sciences: advancements, applications, and challenges for sustainable agriculture and food security

Artificial intelligence (AI) is revolutionizing plant sciences by enabling precise plant species identification, early disease diagnosis, crop yield prediction, and precision agriculture optimization. AI uses machine learning and image recognition to aid ecological research and biodiversity conservation. It plays a crucial role in plant breeding by accelerating the development of resilient, high-yielding crops with desirable traits. AI models using climate and soil data contribute to sustainable agriculture and food security. In plant phenotyping, AI automates the measurement and analysis of plant characteristics, enhancing our understanding of plant growth. Ongoing research aims to improve AI models’ robustness and interpretability while addressing data privacy and algorithmic biases. Interdisciplinary collaboration is essential to fully harness AI’s potential in plant sciences for a sustainable, food-secure future.

Game-changing intelligence: Unveiling the societal impact of artificial intelligence in game software

Efficient artificial intelligence (AI) in gaming software optimizes development, improves user experiences, and increases industrial production, all of which contribute positively to economic development and technical advancement. Possible drawbacks include employment displacement, ethical problems, algorithmic biases, and excessive reliance on AI; everyone has an impact on society’s dynamics and principles. In this study, we proposed a novel method called Fine-Tuned Ring Toss Game Optimization Adaptive Artificial Neural Network (FRTGO-AANN) to improve user experience in game software, examine social implications to promote ethical growth, explore AI transformational significance in games, and shape new game-play. In this research, we employ a dataset of 250 video game projects. The collected data are undergoing feature extraction by using a Principal Component Analysis (PCA) method. We used a few parameters for our suggested and current approaches to analyze the research’s findings. Our proposed FRTGO-AANN method achieves superior results like accuracy (96.7 %), precision (92.5 %), F1-score (95.3 %), recall (90.6 %).AI in gaming software represents a game changer, transforming user experience, narrative intricacy, along with the world of virtual reality. It has a profound social influence, altering entertainment and promoting innovation, along with difficult ethical considerations.

Harmonizing Spirituality and Algorithmic Fairness in Contemporary Work Environments

Harmonizing Spirituality and Algorithmic Fairness in Contemporary Work Environments

Artificial intelligence in education – opportunities and threats

The article aims to present a comprehensive view of the current state and future of artificial intelligence (AI), with particular emphasis on its opportunities and threats in the context of education. The first section introduces the topic, highlighting the revolutionary nature of AI in today’s world, and discussing the complex definition of artificial intelligence. It then outlines the opportunities for the use of AI in education, such as individualization of the learning process, customization of teaching materials, and support for students with special educational needs. An analysis of potential threats is also conducted, including ethical issues, student data security, algorithmic biases, and the risk of students becoming dependent on AI-based tools. Finally, the article emphasizes the need for a responsible approach and the necessity of regulation, ethical guidelines, and supervision to effectively harness the potential of artificial intelligence in education. The article also encourages further research into the effectiveness of AI implementation in education, especially in the Polish context.

Artificial Intelligence And Cancer Care in Africa

AI's potential to revolutionize oncology through enhanced diagnostics, treatment planning, and patient monitoring is well-documented globally. However, in Africa, its adoption has been slower, albeit steadily progressing. This commentary explores the integration of artificial Intelligence in cancer care across Africa, assessing its current state, challenges and future directions. It highlights significant AI innovations in cancer diagnostics, such as DataPathology, PapsAI, MinoHealth, and Hurone AI, which utilize AI for tissue analysis, cervical cell imaging, disease forecasting, and remote patient monitoring. Despite these advancements, several challenges impede AI's full integration into African healthcare systems. Key issues include data privacy and security, algorithm bias, and insufficient regulatory frameworks. The review emphasizes the necessity of robust data protection policies, representative datasets to mitigate biases, and clear guidelines for AI deployment tailored to the African context. Emerging AI technologies in Africa, such as AI-enhanced telemedicine, mobile health applications, predictive analytics, and virtual tumor boards, show promise in overcoming geographic and resource limitations. These innovations can facilitate remote consultations, continuous patient monitoring, and multidisciplinary collaborations, thereby improving cancer care accessibility and outcomes. Conclusively, recommendations for enhancing AI integration in African cancer care, including investing in data infrastructure, capacity building for healthcare professionals, and fostering international collaborations are discussed. Addressing ethical and regulatory challenges is crucial to ensure responsible and effective use of AI technologies. By leveraging AI, Africa can significantly improve cancer care delivery, reduce mortality rates, and enhance patient quality of life.