Artificial Intelligence Decision Research Articles

Cui bono? Judicial decision-making in the era of AI: A qualitative study on the expectations of judges in Germany

Despite substantial artificial intelligence (AI) research in various domains, limited attention has been given to its impact on the judiciary, and studies directly involving judges are rare. We address this gap by using 20 in-depth interviews to investigate German judges’ perspectives on AI. The exploratory study examines (1) the integration of AI in court proceedings by 2040, (2) the impact of increased use of AI on the role and independence of judges, and (3) whether AI decisions should supersede human judgments if they were superior to them. The findings reveal an expected trend toward further court digitalization and various AI use scenarios. Notably, opinions differ on the influence of AI on judicial independence and the precedence of machine decisions over human judgments. Overall, the judges surveyed hold diverse perspectives without a clear trend emerging, although a tendency toward a positive and less critical evaluation of AI in the judiciary is discernible.

The Adoption of Artificial Intelligence in Bureaucratic Decision-making: A Weberian Perspective

This work questions AI´s role in bureaucratic decision-making. The Weberian conception of bureaucracy, based around the concept of an ideal bureaucracy in which authority is distributed, delegated, clearly delimited and hierarchical and that enshrines the following of formal rules, task specialization through division of labor, legal certainty and a predilection for efficiency in recordable and accountable decisions can serve as a framework to orient how governments should approach the adoption of artificial intelligence (AI) given the many problems associated with its careless deployment. Using theoretical analysis, this work explains Weberian ideas of bureaucracy and contrasts them with real-life cases of implementation of AI in bureaucratic decision-making, often with detrimental results for society. After identifying and framing issues related to AI, e.g., lack of transparency, attempts to shift accountability from humans to technology, the exacerbation of bias and potential for systemic discrimination, the paper proposes Weberian prescriptions that should help public administration make careful decisions about the adoption of AI and the consequences of its implementation. The article also engages with Weber critically, rejecting the notion that public administrators do not engage in politics and asserting that AI decision-making is necessarily political as well, as it entails exercising power over citizens.

TachAId-An interactive tool supporting the design of human-centered AI solutions.

In an era where Artificial Intelligence (AI) integration into business processes is crucial for maintaining competitiveness, there is a growing need for structured guidance on designing AI solutions that align with human needs. To this end, we present "technical assistance concerning human-centered AI development" (tachAId), an interactive advisory tool which comprehensively guides AI developers and decision makers in navigating the machine learning lifecycle with a focus on human-centered design. tachAId motivates and presents concrete technical advice to ensure human-centeredness across the phases of AI development. The tool's effectiveness is evaluated through a catalog of criteria for human-centered AI in the form of relevant challenges and goals, derived from existing methodologies and guidelines. Lastly, tachAId and one other comparable advisory tool were examined to determine their adherence to these criteria in order to provide an overview of the human-centered aspects covered by these tools and to allow interested parties to quickly assess whether the tools meet their needs.

The role of artificial intelligence in nuclear crisis decision making: a complement, not a substitute

ABSTRACT This article explores the nuanced interplay between artificial intelligence (AI) and human decision making in the high-stakes arena of nuclear crisis management. We argue that AI, despite its lack of emotional intelligence and experiential learning, presents unique opportunities to enhance decision-makers’ ability to navigate the complexities of nuclear crises. By juxtaposing AI's data-driven insights with human intelligence's depth in emotional and creative processes, we illustrate the complementary roles each can play in fostering empathy, understanding, and, ultimately, security dilemma sensibility (SDS). Through theoretical exploration and thought experiments on historical crises such as the Cuban Missile Crisis and the Able Archer incident, we demonstrate AI's potential in mitigating misperceptions and facilitating informed, empathetic responses that acknowledge the fears and intentions of adversaries. However, we also highlight the inherent limitations and ethical considerations of over-relying on AI, stressing the irreplaceable value of human judgment and the need for a balanced approach that leverages the strengths of both AI and human intelligence. Our conclusion underscores the importance of integrating AI as a tool within a broader strategy of crisis management that prioritises trust-building and direct communication among decision-makers to navigate the delicate dynamics of international security and diplomacy effectively.

Are businesses giving AI the support it needs to grow and scale?

Artificial intelligence (AI) is rapidly transforming industries, and financial services are no exception. This paper explores how AI can revolutionise both customer interactions and back-office operations within banks and insurance companies. AI-powered tools like chatbots and recommendation engines can answer questions 24/7, identify customer trends and suggest personalised products, all of which can significantly improve customer satisfaction. In the back office, AI automates repetitive tasks such as data analysis and document processing. This frees up human employees to focus on more complex and strategic activities, while also leading to cost savings and increased efficiency. However, implementing AI in financial services comes with its own set of challenges. Businesses must continuously invest in AI to keep pace with rapid technological advancements. Regulatory frameworks are struggling to adapt to the evolving nature of AI, which creates uncertainty for businesses. Additionally, ensuring transparency and fairness in AI decision making, especially with complex algorithms, is a major hurdle. The quality of data used to train AI models is also crucial. Biased data can lead to discriminatory outcomes, so maintaining high-quality data is essential. These challenges have led to the development of explainable AI (XAI) models, which can explain their reasoning, helping to address transparency concerns and meet regulatory requirements. Data quality management strategies are also required to ensure data is complete, accurate and unbiased for ethical AI implementation. On a global scale, collaboration among international organisations and governments is necessary to establish consistent and effective AI regulations. The paper concludes that the human element is invaluable in the age of AI. Human expertise is needed to program AI systems, address ethical considerations, and manage potential biases within the data. Furthermore, the workforce needs to be upskilled and reskilled to adapt to the changing landscape with AI. The paper recommends a ‘human-in-the-loop’ (HITL) approach, in which human judgment is combined with AI capabilities for responsible and efficient decision making. Collaboration and international harmonisation are key to ensuring responsible and ethical AI adoption in financial services.

The benefits and costs of explainable artificial intelligence in visual quality control: Evidence from fault detection performance and eye movements

AbstractVisual inspection tasks often require humans to cooperate with artificial intelligence (AI)‐based image classifiers. To enhance this cooperation, explainable artificial intelligence (XAI) can highlight those image areas that have contributed to an AI decision. However, the literature on visual cueing suggests that such XAI support might come with costs of its own. To better understand how the benefits and cost of XAI depend on the accuracy of AI classifications and XAI highlights, we conducted two experiments that simulated visual quality control in a chocolate factory. Participants had to decide whether chocolate molds contained faulty bars or not, and were always informed whether the AI had classified the mold as faulty or not. In half of the experiment, they saw additional XAI highlights that justified this classification. While XAI speeded up performance, its effects on error rates were highly dependent on (X)AI accuracy. XAI benefits were observed when the system correctly detected and highlighted the fault, but XAI costs were evident for misplaced highlights that marked an intact area while the actual fault was located elsewhere. Eye movement analyses indicated that participants spent less time searching the rest of the mold and thus looked at the fault less often. However, we also observed large interindividual differences. Taken together, the results suggest that despite its potentials, XAI can discourage people from investing effort into their own information analysis.

Reliability of AI Algorithms in Safety Applications

The integration of Artificial Intelligence (AI) algorithms into safety-critical applications has become increasingly prevalent across various domains, including autonomous vehicles, medical diagnosis, industrial automation, and aerospace systems. These applications rely heavily on AI to make decisions that directly affect human safety, economic stability, and operational efficiency. Given the critical nature of these tasks, it is essential to rigorously assess the reliability of AI algorithms to ensure they perform consistently and accurately under all conditions. Reliability, in this context, refers to the AI system's ability to function without failure over a specific period, under defined operational conditions. In safety-critical domains, even minor errors or inconsistencies in AI decision-making can lead to catastrophic outcomes, such as traffic accidents involving autonomous vehicles, incorrect medical diagnoses leading to improper treatments, or failures in industrial processes that may cause costly downtime or even human casualties. The increasing complexity and deployment of AI technologies in these domains highlight the urgent need for a comprehensive understanding and evaluation of AI reliability. This paper provides a detailed analysis of the design considerations and methodologies for enhancing the reliability of AI algorithms. The discussion begins by exploring the underlying principles of reliability in AI systems, focusing on both theoretical and practical perspectives. We examine key factors that influence reliability, including data quality, algorithmic robustness, model interpretability, and system integration. The paper then delves into various reliability assessment techniques, such as fault tolerance mechanisms, error detection and correction methods, redundancy, and validation processes. To provide a deeper understanding of reliability in AI, we introduce mathematical models and statistical evaluation techniques that quantify reliability metrics. For instance, reliability modeling using exponential distribution, Monte Carlo simulations for probabilistic reliability analysis, and error propagation studies using Jacobian matrices are presented. We also explore the use of machine learning-specific reliability metrics, such as the Area Under the Curve (AUC) in Receiver Operating Characteristic (ROC) analysis, which helps evaluate the performance of AI in critical decision-making contexts. Furthermore, this paper addresses the current challenges and limitations in ensuring AI reliability, including computational complexity, ethical considerations, and regulatory compliance issues. We highlight the difficulties in developing AI models that can maintain their reliability across diverse and unpredictable real-world scenarios. The potential for bias, lack of transparency in AI decision-making, and difficulties in explaining complex AI models also present significant hurdles that need to be addressed to enhance reliability. The findings and methodologies discussed in this paper aim to contribute to a deeper understanding of the complex landscape of AI reliability, providing a framework for researchers, practitioners, and policymakers to develop safer, more reliable AI systems that can be trusted to operate in environments where safety is paramount.

SYNERGIES IN THE EVOLUTION OF ARTIFICIAL INTELLIGENCE AND FASHION

This study provides an in-depth analysis of the intersection between artificial intelligence (AI) and fashion, focusing on the potentials and ethical challenges related to the intensive use of large datasets. Using methodologies such as “Envisioning” and the “Scenario Planning Model”, the aim is to develop clear and structured approaches to the deployment of AI in the fashion industry, exploring how this technology can transform and enrich the creative process, extending human capabilities. Among the main ethical challenges are training on culturally connoted data, which risks introducing formative bias, and the need to explain AI processes and decisions in understandable terms (explainability). The possible formalization of AI methods to make their processes and results formal is also discussed. The potential capabilities of these technologies to overcome the limits of human intellect are also considered. The conclusions highlight the importance of a balance between technological innovation and ethical considerations to promote ongoing dialogue among diverse stakeholders, ensuring that AI in the fashion industry evolves inclusively, ethically, and responsibly, fostering a future where technology and human creativity coexist harmoniously.

Cultural-Centric Online English Teaching with Artificial Intelligence Technology and Decision Support System

Cultural-Centric Online English Teaching with Artificial Intelligence Technology and Decision Support System

THE IMPACT OF ARTIFICIAL INTELLIGENCE IN DECISION MAKING: A COMPREHENSIVE REVIEW

Artificial Intelligence (AI) has emerged as a transformative technology with the potential to revolutionize decision making across various domains. This research article explores the impact of AI in decision making and its implications for individuals, organizations, and society. The article begins by providing an overview of AI and its key components, such as machine learning and natural language processing. It then discusses the role of AI in enhancing decision-making processes by automating tasks, augmenting human capabilities, and providing data-driven insights. The article highlights the benefits of AI in improving decision accuracy, efficiency, and scalability, while also acknowledging the challenges and risks associated with its implementation. These challenges include ethical considerations, biases in AI algorithms, and potential job displacement. The article further explores the importance of transparency, accountability, and interpretability in AI decision-making systems. Additionally, it discusses the role of human-AI collaboration and the need for interdisciplinary approaches to ensure the responsible and ethical deployment of AI in decision making. Drawing on case studies and empirical research, the article provides concrete examples of how AI is transforming decision making in various fields, such as finance, healthcare, and transportation. Finally, the article concludes by discussing future directions and recommendations for policymakers, organizations, and individuals to harness the full potential of AI in decision making while addressing its ethical, social, and economic implications. KEYWORDS: Artificial Intelligence, Decision Making, Machine Learning, Automation, Augmentation, Ethical Considerations, Bias, Transparency, Human-AI Collaboration, Responsible AI.

Trust's Significance in Human-AI Communication and Decision-Making

With artificial intelligence (AI) continuing to pervade many aspects of society, it is critical to comprehend the dynamics of trust in AI decision-making and human-AI interaction. This study explores the many facets of trust and looks at how important it is in influencing user attitudes, actions, and the general effectiveness of AI systems. In order to understand the complex interactions between intelligent machines and people, the research incorporates multidisciplinary viewpoints from the fields of psychology, human-computer interaction, and ethics. The first area of inquiry is what influences the creation of first faith in AI. [1]We investigate how consumers' desire to trust AI-driven technology is influenced by system transparency, explain ability, and user experience through empirical study. The creation of design concepts intended to build a foundation of trust in AI systems is informed by insights gained at this stage. The second aspect of the study focuses on how trust changes over time in extended encounters between humans and artificial intelligence. We study the dynamics of trust-building and erosion by monitoring user experiences and system performance. This helps to clarify the critical points and factors that affect the trust's trajectory. This long-term viewpoint aids in the creation of adaptable artificial intelligence systems that can adapt to changing user demands and address issues with trust. The third line of investigation concerns the function of trust in AI-influenced decision-making processes. We evaluate the extent to which users depend on AI-generated insights and the influence of trust on decision outcomes using experimental scenarios and real-world case studies. This stage clarifies the fine balance needed to maximise the collaboration between AI and humans and emphasises the significance of matching AI suggestions with user values. The research concludes with an examination of the consequences of trust in AI for wider societal contexts, with a focus on ethical issues. We look at accountability frameworks, the potential fallout from blind trust, and the moral obligations of AI engineers in creating reliable systems. In order to foster a symbiotic relationship between humans and intelligent systems in a world increasingly driven by AI, this thorough investigation of the role of trust in human-AI interaction and decision-making ultimately aims to provide actionable insights for the design, implementation, and governance of AI technologies.

The Potential Utilisation of Artificial Intelligence (AI) in Enterprises

Artificial intelligence (AI) is a part of computer science that aims to create and develop intelligent machines. For AI to function and perform, it involves the development of algorithms, models, and systems that enable computers to learn from the data, identify patterns, and make predictions. Research indicates that AI can increase private industry output, decision-making, and effectiveness. By identifying how enterprises utilize AI, what is the impact of implementing AI in enterprise operations, and how does it affect the efficiency, productivity, and overall performance of the business? This exploratory research begins with insight into understanding AI and addresses how AI has been utilized and implemented in businesses, ethical and societal considerations, and potential benefits and challenges that businesses may confront. Ten in-depth interviews with leading industry experts using AI, followed by focus groups, generated information. opinions, and key insights. By utilizing the software Nvivo, this quantitative research presents key themes and content findings to assist in educating enterprise personnel during the AI decision-making and implementation stages.

AI in Autonomous Vehicles

The integration of Artificial Intelligence (AI) in autonomous vehicles represents a transformative leap towards safer, more efficient, and technologically advanced transportation systems. This abstract provides a comprehensive overview of the dynamic landscape where AI converges with autonomous vehicles, examining the synergies, challenges, and far-reaching implications of this groundbreaking integration AI-Powered Perception and Decision-Making: Delving into the technological core, the abstract explores how AI empowers autonomous vehicles with sophisticated perception systems, such as computer vision and sensor fusion. It discusses the role of machine learning algorithms in real-time decision-making, enabling vehicles to adapt to dynamic road conditions and unforeseen circumstances. Challenges in Autonomy: Recognizing the complexity of autonomous systems, the abstract addresses challenges such as handling edge cases, ensuring robustness against adversarial attacks, and navigating regulatory and ethical considerations. It emphasizes the importance of addressing these challenges to foster public trust and acceptance of autonomous vehicles. Human-AI Interaction in Autonomous Vehicles: Examining the interface between humans and AI-driven vehicles, the abstract discusses the importance of designing intuitive and trustworthy communication channels. It explores advancements in natural language processing and gesture recognition, fostering seamless collaboration between humans and autonomous systems. Regulatory Landscape and Ethical Considerations: Recognizing the pivotal role of regulations, the abstract discusses the evolving regulatory landscape for autonomous vehicles. It delves into ethical considerations surrounding AI decisions in critical situations, underscoring the need for a harmonized approach to ensure responsible AI deployment in autonomous driving scenarios.

Artificial Intelligence Sensing: Effective Flavor Blueprinting of Tea Infusions for a Quality Control Perspective.

Tea infusions are the most consumed beverages in the world after water; their pleasant yet peculiar flavor profile drives consumer choice and acceptance and becomes a fundamental benchmark for the industry. Any qualification method capable of objectifying the product's sensory features effectively supports industrial quality control laboratories in guaranteeing high sample throughputs even without human panel intervention. The current study presents an integrated analytical strategy acting as an Artificial Intelligence decision tool for black tea infusion aroma and taste blueprinting. Key markers validated by sensomics are accurately quantified in a wide dynamic range of concentrations. Thirteen key aromas are quantitatively assessed by standard addition with in-solution solid-phase microextraction sampling followed by GC-MS. On the other hand, nineteen key taste and quality markers are quantified by external standard calibration and LC-UV/DAD. The large dynamic range of concentration for sensory markers is reflected in the selection of seven high-quality teas from different geographical areas (Ceylon, Darjeeling Testa Valley and Castleton, Assam, Yunnan, Azores, and Kenya). The strategy as a sensomics-based expert system predicts teas' sensory features and acts as an AI smelling and taste machine suitable for quality controls.

Transparency in AI Decision Making: A Survey of Explainable AI Methods and Applications

Artificial Intelligence (AI) systems have become pervasive in numerous facets of modern life, wielding considerable influence in critical decision-making realms such as healthcare, finance, criminal justice, and beyond. Yet, the inherent opacity of many AI models presents significant hurdles concerning trust, accountability, and fairness. To address these challenges, Explainable AI (XAI) has emerged as a pivotal area of research, striving to augment the transparency and interpretability of AI systems. This survey paper serves as a comprehensive exploration of the state-of-the-art in XAI methods and their practical applications. We delve into a spectrum of techniques, spanning from model-agnostic approaches to interpretable machine learning models, meticulously scrutinizing their respective strengths, limitations, and real-world implications. The landscape of XAI is rich and varied, with diverse methodologies tailored to address different facets of interpretability. Model-agnostic approaches offer versatility by providing insights into model behavior across various AI architectures. In contrast, interpretable machine learning models prioritize transparency by design, offering inherent understandability at the expense of some predictive performance. Layer-wise Relevance Propagation (LRP) and attention mechanisms delve into the inner workings of neural networks, shedding light on feature importance and decision processes. Additionally, counterfactual explanations open avenues for exploring what-if scenarios, elucidating the causal relationships between input features and model outcomes. In tandem with methodological exploration, this survey scrutinizes the deployment and impact of XAI across multifarious domains. Successful case studies showcase the practical utility of transparent AI in healthcare diagnostics, financial risk assessment, criminal justice systems, and more. By elucidating these use cases, we illuminate the transformative potential of XAI in enhancing decision-making processes while fostering accountability and fairness. Nevertheless, the journey towards fully transparent AI systems is fraught with challenges and opportunities. As we traverse the current landscape of XAI, we identify pressing areas for further research and development. These include refining interpretability metrics, addressing the scalability of XAI techniques to complex models, and navigating the ethical dimensions of transparency in AI decision-making.Through this survey, we endeavor to cultivate a deeper understanding of transparency in AI decision-making, empowering stakeholders to navigate the intricate interplay between accuracy, interpretability, and ethical considerations. By fostering interdisciplinary dialogue and inspiring collaborative innovation, we aspire to catalyze future advancements in Explainable AI, ultimately paving the way towards more accountable and trustworthy AI systems.

Explainable deep learning diagnostic system for prediction of lung disease from medical images

Around the globe, respiratory lung diseases pose a severe threat to human survival. Based on a central goal to reduce contiguous transmission from infected to healthy persons, several technologies have evolved for diagnosing lung pathologies. One of the emerging technologies is the utility of Artificial Intelligence (AI) based on computer vision for processing wide varieties of medical imaging but AI methods without explainability are often treated as a black box. Based on a view to demystifying the rationale influencing AI decisions, this paper designed and developed a novel low-cost explainable deep-learning diagnostic tool for predicting lung disease from medical images. For this, we investigated explainable deep learning (DL) models (conventional DL and vision transformers (ViTs)) for performing prediction of the existence of pneumonia, COVID19, or no-disease from both original and data augmentation (DA)-based medical images (from two chest X-ray datasets). The results show that our experimental consideration of the DA that combines the impact of cropping, rotation, and horizontal flipping (CROP+ROT+HF) for transforming input images and then passed as input to an Inception-V3 architecture yielded a performance that surpasses all the ViTs and other conventional DL approaches in most of the evaluated performance metrics. Overall, the results suggest that the utility of data augmentation schemes aided the DL methods to yield higher classification accuracies. Furthermore, we compared five different class activation mapping (CAM) algorithms (GradCAM, GradCAM++, EigenGradCAM, AblationCAM, and RandomCAM). The result shows that most of the examined CAM algorithms were effective in identifying the attention region containing the existence of pneumonia or COVID-19 from the medical images (chest X-rays). Our developed low-cost AI diagnostic tool (pilot system) can assist medical experts and radiographers in proffering early diagnosis of lung disease. For this, we selected five to seven deep learning models and the explainable algorithms were deployed on a novel web interface implemented via a Gradio framework.

Artificial intelligence-based decision support software to improve the efficacy of acute stroke pathway in the NHS: an observational study.

In a drip-and-ship model for endovascular thrombectomy (EVT), early identification of large vessel occlusion (LVO) and timely referral to a comprehensive center (CSC) are crucial when patients are admitted to an acute stroke center (ASC). Several artificial intelligence (AI) decision-aid tools are increasingly being used to facilitate the rapid identification of LVO. This retrospective cohort study aimed to evaluate the impact of deploying e-Stroke AI decision support software in the hyperacute stroke pathway on process metrics and patient outcomes at an ASC in the United Kingdom. Except for the deployment of e-Stroke on 01 March 2020, there were no significant changes made to the stroke pathway at the ASC. The data were obtained from a prospective stroke registry between 01 January 2019 and 31 March 2021. The outcomes were compared between the 14 months before and 12 months after the deployment of AI (pre-e-Stroke cohort vs. post-e-Stroke cohort) on 01 March 2020. Time window analyses were performed using Welch's t-test. Cochran-Mantel-Haenszel test was used to compare changes in disability at 3 months assessed by modified Rankin Score (mRS) ordinal shift analysis, and Fisher's exact test was used for dichotomised mRS analysis. In the pre-e-Stroke cohort, 19 of 22 patients referred received EVT. In the post-e-Stroke cohort, 21 of the 25 patients referred were treated. The mean door-in-door-out (DIDO) and door-to-referral times in pre-e-Stroke vs. post-e-Stroke cohorts were 141 vs. 79 min (difference 62 min, 95% CI 96.9-26.8 min, p < 0.001) and 71 vs. 44 min (difference 27 min, 95% CI 47.4-5.4 min, p = 0.01), respectively. The adjusted odds ratio (age and NIHSS) for mRS ordinal shift analysis at 3 months was 3.14 (95% CI 0.99-10.51, p = 0.06) and the dichotomized mRS 0-2 at 3 months was 16% vs. 48% (p = 0.04) in the pre- vs. post-e-Stroke cohorts, respectively. In this single-center study in the United Kingdom, the DIDO time significantly decreased since the introduction of e-Stroke decision support software into an ASC hyperacute stroke pathway. The reduction in door-in to referral time indicates faster image interpretation and referral for EVT. There was an indication of an increased proportion of patients regaining independent function after EVT. However, this should be interpreted with caution given the small sample size. Larger, prospective studies and further systematic real-world evaluation are needed to demonstrate the widespread generalisability of these findings.

Navigating Now and Next: Recent Advances and Future Horizons in Robotic Radical Prostatectomy.

Robotic-assisted radical prostatectomy (RARP) has become the leading approach for radical prostatectomy driven by innovations aimed at improving functional and oncological outcomes. The initial advancement in this field was transperitoneal multiport robotics, which has since undergone numerous technical modifications. These enhancements include the development of extraperitoneal, transperineal, and transvesical approaches to radical prostatectomy, greatly facilitated by the advent of the Single Port (SP) robot. This review offers a comprehensive analysis of these evolving techniques and their impact on RARP. Additionally, we explore the transformative role of artificial intelligence (AI) in digitizing robotic prostatectomy. AI advancements, particularly in automated surgical video analysis using computer vision technology, are unprecedented in their scope. These developments hold the potential to revolutionize surgeon feedback and assessment and transform surgical documentation, and they could lay the groundwork for real-time AI decision support during surgical procedures in the future. Furthermore, we discuss future robotic platforms and their potential to further enhance the field of RARP. Overall, the field of minimally invasive radical prostatectomy for prostate cancer has been an incubator of innovation over the last two decades. This review focuses on some recent developments in robotic prostatectomy, provides an overview of the next frontier in AI innovation during prostate cancer surgery, and highlights novel robotic platforms that may play an increasing role in prostate cancer surgery in the future.

Prospective deployment of an automated implementation solution for artificial intelligence translation to clinical radiation oncology.

Artificial intelligence (AI)-based technologies embody countless solutions in radiation oncology, yet translation of AI-assisted software tools to actual clinical environments remains unrealized. We present the Deep Learning On-Demand Assistant (DL-ODA), a fully automated, end-to-end clinical platform that enables AI interventions for any disease site featuring an automated model-training pipeline, auto-segmentations, and QA reporting. We developed, tested, and prospectively deployed the DL-ODA system at a large university affiliated hospital center. Medical professionals activate the DL-ODA via two pathways (1): On-Demand, used for immediate AI decision support for a patient-specific treatment plan, and (2) Ambient, in which QA is provided for all daily radiotherapy (RT) plans by comparing DL segmentations with manual delineations and calculating the dosimetric impact. To demonstrate the implementation of a new anatomy segmentation, we used the model-training pipeline to generate a breast segmentation model based on a large clinical dataset. Additionally, the contour QA functionality of existing models was assessed using a retrospective cohort of 3,399 lung and 885 spine RT cases. Ambient QA was performed for various disease sites including spine RT and heart for dosimetric sparing. Successful training of the breast model was completed in less than a day and resulted in clinically viable whole breast contours. For the retrospective analysis, we evaluated manual-versus-AI similarity for the ten most common structures. The DL-ODA detected high similarities in heart, lung, liver, and kidney delineations but lower for esophagus, trachea, stomach, and small bowel due largely to incomplete manual contouring. The deployed Ambient QAs for heart and spine sites have prospectively processed over 2,500 cases and 230 cases over 9 months and 5 months, respectively, automatically alerting the RT personnel. The DL-ODA capabilities in providing universal AI interventions were demonstrated for On-Demand contour QA, DL segmentations, and automated model training, and confirmed successful integration of the system into a large academic radiotherapy department. The novelty of deploying the DL-ODA as a multi-modal, fully automated end-to-end AI clinical implementation solution marks a significant step towards a generalizable framework that leverages AI to improve the efficiency and reliability of RT systems.

Reader bias in breast cancer screening related to cancer prevalence and artificial intelligence decision support—a reader study

ObjectivesThe aim of our study was to examine how breast radiologists would be affected by high cancer prevalence and the use of artificial intelligence (AI) for decision support.Materials and methodThis reader study was based on selection of screening mammograms, including the original radiologist assessment, acquired in 2010 to 2013 at the Karolinska University Hospital, with a ratio of 1:1 cancer versus healthy based on a 2-year follow-up. A commercial AI system generated an exam-level positive or negative read, and image markers. Double-reading and consensus discussions were first performed without AI and later with AI, with a 6-week wash-out period in between. The chi-squared test was used to test for differences in contingency tables.ResultsMammograms of 758 women were included, half with cancer and half healthy. 52% were 40–55 years; 48% were 56–75 years. In the original non-enriched screening setting, the sensitivity was 61% (232/379) at specificity 98% (323/379). In the reader study, the sensitivity without and with AI was 81% (307/379) and 75% (284/379) respectively (p < 0.001). The specificity without and with AI was 67% (255/379) and 86% (326/379) respectively (p < 0.001). The tendency to change assessment from positive to negative based on erroneous AI information differed between readers and was affected by type and number of image signs of malignancy.ConclusionBreast radiologists reading a list with high cancer prevalence performed at considerably higher sensitivity and lower specificity than the original screen-readers. Adding AI information, calibrated to a screening setting, decreased sensitivity and increased specificity.Clinical relevance statementRadiologist screening mammography assessments will be biased towards higher sensitivity and lower specificity by high-risk triaging and nudged towards the sensitivity and specificity setting of AI reads. After AI implementation in clinical practice, there is reason to carefully follow screening metrics to ensure the impact is desired.Key Points• Breast radiologists’ sensitivity and specificity will be affected by changes brought by artificial intelligence.• Reading in a high cancer prevalence setting markedly increased sensitivity and decreased specificity.• Reviewing the binary reads by AI, negative or positive, biased screening radiologists towards the sensitivity and specificity of the AI system.Graphical abstract