Artificial Intelligence Systems Research Articles

Aptamer-functionalized graphene quantum dots combined with artificial intelligence detect bacteria for urinary tract infections.

Urinary tract infection is one of the most prevalent forms of bacterial infection, and prompt and efficient identification of pathogenic bacteria plays a pivotal role in the management of urinary tract infections. In this study, we propose a novel approach utilizing aptamer-functionalized graphene quantum dots integrated with an artificial intelligence detection system (AG-AI detection system) for rapid and highly sensitive detection of Escherichia coli (E. coli). Firstly, graphene quantum dots were modified with the aptamer that can specifically recognize and bind to E. coli. Therefore, the fluorescence intensity of graphene quantum dots was positively correlated with the concentration of E. coli. Finally, the fluorescence images were processed by artificial intelligence system to obtain the result of bacterial concentration. The AG-AI detection system, with wide linearity (103-109 CFU/mL) and a low detection limit (3.38×102 CFU/mL), can effectively differentiate between E. coli and other urinary tract infection bacteria. And the result of detection system is in good agreement with MALDI-TOF MS. The detection system is an accurate and effective way to detect bacteria in urinary tract infections.

Unveiling the Potential of Large Language Models in Transforming Chronic Disease Management: Mixed Methods Systematic Review.

Chronic diseases are a major global health burden, accounting for nearly three-quarters of the deaths worldwide. Large language models (LLMs) are advanced artificial intelligence systems with transformative potential to optimize chronic disease management; however, robust evidence is lacking. This review aims to synthesize evidence on the feasibility, opportunities, and challenges of LLMs across the disease management spectrum, from prevention to screening, diagnosis, treatment, and long-term care. Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) guidelines, 11 databases (Cochrane Central Register of Controlled Trials, CINAHL, Embase, IEEE Xplore, MEDLINE via Ovid, ProQuest Health & Medicine Collection, ScienceDirect, Scopus, Web of Science Core Collection, China National Knowledge Internet, and SinoMed) were searched on April 17, 2024. Intervention and simulation studies that examined LLMs in the management of chronic diseases were included. The methodological quality of the included studies was evaluated using a rating rubric designed for simulation-based research and the risk of bias in nonrandomized studies of interventions tool for quasi-experimental studies. Narrative analysis with descriptive figures was used to synthesize the study findings. Random-effects meta-analyses were conducted to assess the pooled effect estimates of the feasibility of LLMs in chronic disease management. A total of 20 studies examined general-purpose (n=17) and retrieval-augmented generation-enhanced LLMs (n=3) for the management of chronic diseases, including cancer, cardiovascular diseases, and metabolic disorders. LLMs demonstrated feasibility across the chronic disease management spectrum by generating relevant, comprehensible, and accurate health recommendations (pooled accurate rate 71%, 95% CI 0.59-0.83; I2=88.32%) with retrieval-augmented generation-enhanced LLMs having higher accuracy rates compared to general-purpose LLMs (odds ratio 2.89, 95% CI 1.83-4.58; I2=54.45%). LLMs facilitated equitable information access; increased patient awareness regarding ailments, preventive measures, and treatment options; and promoted self-management behaviors in lifestyle modification and symptom coping. Additionally, LLMs facilitate compassionate emotional support, social connections, and health care resources to improve the health outcomes of chronic diseases. However, LLMs face challenges in addressing privacy, language, and cultural issues; undertaking advanced tasks, including diagnosis, medication, and comorbidity management; and generating personalized regimens with real-time adjustments and multiple modalities. LLMs have demonstrated the potential to transform chronic disease management at the individual, social, and health care levels; however, their direct application in clinical settings is still in its infancy. A multifaceted approach that incorporates robust data security, domain-specific model fine-tuning, multimodal data integration, and wearables is crucial for the evolution of LLMs into invaluable adjuncts for health care professionals to transform chronic disease management. PROSPERO CRD42024545412; https://www.crd.york.ac.uk/PROSPERO/view/CRD42024545412.

The imperative of diversity and equity for the adoption of responsible AI in healthcare.

Artificial Intelligence (AI) in healthcare holds transformative potential but faces critical challenges in ethical accountability and systemic inequities. Biases in AI models, such as lower diagnosis rates for Black women or gender stereotyping in Large Language Models, highlight the urgent need to address historical and structural inequalities in data and development processes. Disparities in clinical trials and datasets, often skewed toward high-income, English-speaking regions, amplify these issues. Moreover, the underrepresentation of marginalized groups among AI developers and researchers exacerbates these challenges. To ensure equitable AI, diverse data collection, federated data-sharing frameworks, and bias-correction techniques are essential. Structural initiatives, such as fairness audits, transparent AI model development processes, and early registration of clinical AI models, alongside inclusive global collaborations like TRAIN-Europe and CHAI, can drive responsible AI adoption. Prioritizing diversity in datasets and among developers and researchers, as well as implementing transparent governance will foster AI systems that uphold ethical principles and deliver equitable healthcare outcomes globally.

AI and Social Interactions: How Artificial Intelligence is Shaping Human Psychology and Social Dynamics

People's socializing tendencies have been profoundly changed by the use and acceptance of AI equipment. With an emphasis on motivated involvement and interaction actions, this research looks at how artificial intelligence (AI) affects people psychologically. The goal of the project is to provide a conceptual structure for understanding how innovations and advancements in AI technology affect the request, conduct, and intellect of individuals. Surveys and qualitative methods were used to test the study concepts. The results demonstrated a good correlation between empathy and artificial intelligence (AI), as well as a positive relationship between cognitive integration and each person's awareness level. The invention is encouraged by the psychological association between considerable data proficiency and the intermediate level of mental drive. To support personal growth, the research aims to shine a light on how AI technology influences unique creative attitudes and explain the conduct, drive, and cognitive aspects of AI systems from the social-psychological standpoint.

Detection of diabetic macular oedema patterns with fine-grained image categorisation on optical coherence tomography

PurposeTo develop an artificial intelligence (AI) system for detecting pathological patterns of diabetic macular oedema (DME) with fine-grained image categorisation using optical coherence tomography (OCT) images.MethodsThe development of the AI...

Artificial intelligence-assisted esophagogastroduodenoscopy improves procedure quality for endoscopists in early stages of training.

Completeness of esophagagogastroduodenoscopy (EGD) varies among endoscopists, leading to a high miss rate for gastric neoplasms. This study aimed to determine the effect of the Cerebro real-time artificial intelligence (AI) system on completeness of EGD for endoscopists in early stages of training. The AI system was built with CNN and Motion Adaptive Temporal Feature Aggregation (MA-TFA). A prospective sequential cohort study was conducted. Endoscopists were taught about the standardized EGD protocol to examine 27 sites. Then, each subject performed diagnostic EGDs per protocol (control arm). After completion of the required sample size, subjects performed diagnostic EGDs with assistance of the AI (study arm). The primary outcome was the rate of completeness of EGD. Secondary outcomes included overall inspection time, individual site inspection time, completeness of photodocumentation, and rate of positive pathologies. A total of 466 EGDs were performed with 233 in each group. Use of AI significantly improved completeness of EGD [mean (SD) (92.6% (6.2%) vs 71.2% (16.8%)]; P <0.001 (95% confidence interval 19.2%-23.8%, SD 0.012). There was no difference in overall mean (SD) inspection time [765.5 (338.4) seconds vs 740.4 (266.2); P =0.374]. Mean (SD) number of photos for photo-documentation significantly increased in the AI group [26.9 (0.4) vs 10.3 (4.4); P <0.001]. There was no difference in detection rates for pathologies in the two groups [8/233 (3.43%) vs 5/233 (2.16%), P =0.399]. Completeness of EGD examination and photodocumentation by endoscopists in early stages of are improved by the AI-assisted software Cerebro.

Pulmonary health: From neonatology to gerontology

Pulmonary health is a key aspect of modern medicine, covering all age groups, from neonatology to gerontology. The article addresses the impact of genetic factors, environment, and human lifestyle on the development of respiratory diseases. Particular attention is paid to the role of genetic predisposition, embryonic disorders, and external factors such as air pollutants and tobacco smoking in the development of chronic lung diseases. Data on the structural and functional development of the respiratory system, its age-related changes, and the risks associated with the involution of the lung tissue are presented. An interdisciplinary approach, including cooperation between pediatricians, therapists, and pulmonologists, is essential for the early diagnosis and effective treatment of respiratory diseases. Particular attention is paid to the impact of new scientific advances, including genomic research and artificial intelligence systems, on diagnosing and treating respiratory diseases.

A Study on the Impact of AI on Artistic Creation

This research examines the revolutionary effect of Artificial Intelligence (AI) on creative work, with a focus on how new technologies are shifting conventional definitions of creativity, authorship, and artistic worth. As machine learning, deep learning, and neural networks continue to advance, AI systems like Generative Adversarial Networks (GANs) and deep neural networks can currently create original works of visual art, music, literature, and digital media. These advancements undermine the centuries-old assumption that creativity is a human-specific attribute, as machines start producing content that is as complex and valuable as human-created art. From AI-painted works of art selling for millions to AI-written music and literature receiving mainstream recognition, this technological advancement poses important questions regarding the future of artistic careers, intellectual property rights, and the ethical limits of machine-generated creativity. The report seeks to explore these changes, providing an inside look at the ways in which AI is remaking the art world and how this is shaping artists, spectators, and the wider cultural environment.

Classification and segmentation of hip fractures in x-rays: highlighting fracture regions for interpretable diagnosis

ObjectiveTo develop an artificial intelligence (AI) system capable of classifying and segmenting femoral fractures. To compare its performance against existing state-of-the-art methods.MethodsThis Institutional Review Board (IRB)-approved retrospective study did not require informed consent. 10,308 hip x-rays from 2618 patients were retrieved from the hospital PACS. 986 were randomly selected for annotation and randomly split into training, validation, and test sets at the patient level. Two radiologists segmented and classified femoral fractures based on their location (femoral neck, pertrochanteric region, or subtrochanteric region) and grade, using the Evans and Garden scales for neck and pertrochanteric regions, respectively. A YOLOv8 segmentation convolutional neural network (CNN) was trained to generate fracture masks and indicate their class and grade. Classification CNNs were trained in the same dataset for method comparison.ResultsOn the test set, YOLOv8 achieved a Dice coefficient of 0.77 (95% CI: 0.56–0.98) for segmenting fractures, an accuracy of 86.2% (95% CI: 80.77–90.55) for classification and grading, and an AUC of 0.981 (95% CI: 0.965–0.997) for fracture detection. These metrics are on par with or exceed those of previously published AI methods, demonstrating the efficacy of our approach.ConclusionsThe high accuracy and AUC values demonstrate the potential of the proposed neural network as a reliable tool in clinical settings. Further, it is the first to provide a precise segmentation of femoral fractures, as indicated by the Dice scores, which may enhance interpretability. A formal evaluation is planned to further assess its clinical applicability.Critical relevance statementThe proposed system offers high granularity in fracture classification and is the first to segment femoral fractures, ensuring interpretability.Key PointsWe present the first AI method that segments and grades femoral fractures.The method classifies fractures with fracture location and type.High accuracy and interpretability promise utility in clinical practice.Graphical

Artificial intelligence as a driver for improving the quality of management decisions of oil companies

Subject. This article discusses the importance of artificial intelligence systems in the management of oil companies. Objectives. The article aims to develop an algorithm for making a management decision based on the use of artificial intelligence tools and practical recommendations for its application. Methods. For the study, we used complex, comparative, and logical analyses. Results. The article presents an algorithm for making management decisions by the oil and gas company’s management team, involving the use of artificial intelligence systems. Relevance. The results of the study can be used by specialists of enterprises of the fuel and energy complex responsible for the introduction of artificial intelligence systems into the production process.

The Transformative Role of Artificial Intelligence in Plastic and Reconstructive Surgery: Challenges and Opportunities.

Background/Objectives: This study comprehensively examines how artificial intelligence (AI) technologies are transforming clinical practice in plastic and reconstructive surgery across the entire patient care continuum, with the specific objective of identifying evidence-based applications, implementation challenges, and emerging opportunities that will shape the future of the specialty. Methods: A comprehensive narrative review was conducted analyzing the integration of AI technologies in plastic surgery, including preoperative planning, intraoperative applications, postoperative monitoring, and quality improvement. Challenges related to implementation, ethics, and regulatory frameworks were also examined, along with emerging technological trends that will shape future practice. Results: AI applications in plastic surgery demonstrate significant potential across multiple domains. In preoperative planning, AI enhances risk assessment, outcome prediction, and surgical simulation. Intraoperatively, AI-assisted robotics enables increased precision and technical capabilities beyond human limitations, particularly in microsurgery. Postoperatively, AI improves complication detection, pain management, and outcomes assessment. Despite these benefits, implementation faces challenges including data privacy concerns, algorithmic bias, liability questions, and the need for appropriate regulatory frameworks. Future directions include multimodal AI systems, federated learning approaches, and integration with extended reality and regenerative medicine technologies. Conclusions: The integration of AI into plastic surgery represents a significant opportunity to enhance surgical precision, improve outcome prediction, and expand the boundaries of what is surgically possible. However, successful implementation requires addressing ethical considerations and maintaining the human elements of surgical care. Plastic surgeons must actively engage with AI development to ensure these technologies address genuine clinical needs while aligning with the specialty's core values of restoring form and function, alleviating suffering, and enhancing quality of life.

Ethical implications related to processing of personal data and artificial intelligence in humanitarian crises: a scoping review

BackgroundHumanitarian organizations are rapidly expanding their use of data in the pursuit of operational gains in effectiveness and efficiency. Ethical risks, particularly from artificial intelligence (AI) data processing, are increasingly recognized yet inadequately addressed by current humanitarian data protection guidelines. This study reports on a scoping review that maps the range of ethical issues that have been raised in the academic literature regarding data processing of people affected by humanitarian crises.MethodsWe systematically searched databases to identify peer-reviewed studies published since 2010. Data and findings were standardized, grouping ethical issues into the value categories of autonomy, beneficence, non-maleficence, and justice. The study protocol followed Arksey and O’Malley’s approach and PRISMA reporting guidelines.ResultsWe identified 16,200 unique records and retained 218 relevant studies. Nearly one in three (n = 66) discussed technologies related to AI. Seventeen studies included an author from a lower-middle income country while four included an author from a low-income country. We identified 22 ethical issues which were then grouped along the four ethical value categories of autonomy, beneficence, non-maleficence, and justice. Slightly over half of included studies (n = 113) identified ethical issues based on real-world examples. The most-cited ethical issue (n = 134) was a concern for privacy in cases where personal or sensitive data might be inadvertently shared with third parties. Aside from AI, the technologies most frequently discussed in these studies included social media, crowdsourcing, and mapping tools.ConclusionsStudies highlight significant concerns that data processing in humanitarian contexts can cause additional harm, may not provide direct benefits, may limit affected populations’ autonomy, and can lead to the unfair distribution of scarce resources. The increase in AI tool deployment for humanitarian assistance amplifies these concerns. Urgent development of specific, comprehensive guidelines, training, and auditing methods is required to address these ethical challenges. Moreover, empirical research from low and middle-income countries, disproportionally affected by humanitarian crises, is vital to ensure inclusive and diverse perspectives. This research should focus on the ethical implications of both emerging AI systems, as well as established humanitarian data management practices.Trial registrationNot applicable.

Algorithmic learning or learner autonomy? Rethinking AI’s role in digital education

PurposeThe purpose of this study is to critically investigate the effects of artificial intelligence (AI)-driven learning environments on learner autonomy, knowledge co-construction and digital equity in Thailand’s online education context. It aims to analyze how algorithmic learning tools and AI-based systems shape learners' educational experiences, particularly focusing on whether these technologies enhance or constrain learner autonomy and collaborative interactions. Additionally, it evaluates ethical and social implications, particularly the digital divide, algorithmic biases and the transparency of AI-mediated educational processes. The study seeks to provide actionable insights and recommendations for policymakers, educators and developers to optimize AI integration for inclusive lifelong learning.Design/methodology/approachThis research adopts a qualitative case study methodology involving in-depth, semi-structured interviews, participant observations and detailed document analysis from 30 diverse participants, including urban and rural learners, educators and instructional designers. Employing actor-network theory (ANT) and posthumanist theoretical frameworks, the study critically examines interactions between human and non-human actors within AI-driven educational environments. This approach allows for comprehensive exploration of participants' lived experiences, perceptions and interactions with AI technologies. Triangulation of multiple data sources ensures depth and reliability, providing nuanced insights into how AI systems influence learner autonomy, collaborative learning processes and educational equity.FindingsThe study identifies AI-driven learning platforms as having dual impacts on learner autonomy, simultaneously enabling greater self-directed learning through personalized guidance and real-time feedback while imposing algorithmic constraints that restrict intellectual exploration. AI-facilitated knowledge co-construction effectively organizes and structures peer interactions yet often results in depersonalized, transactional communication lacking deep human engagement. Furthermore, significant digital inequities are evident, with rural learners disproportionately disadvantaged due to limited technological infrastructure and algorithmic biases embedded within AI-driven platforms. These inequities highlight a critical need for inclusive, transparent and ethical design practices to ensure equitable access and meaningful learner agency.Research limitations/implicationsThe qualitative, context-specific nature of the case study limits the generalizability of findings beyond Thailand’s online educational environment. The rapidly evolving nature of AI technologies suggests that findings reflect a specific temporal context, necessitating continuous research updates. Participant selection bias may have influenced findings, with participants potentially holding strong pre-existing opinions on AI-driven education. Future studies should consider longitudinal analyses to understand long-term impacts, comparative cross-cultural research to validate findings and deeper exploration of algorithmic biases, fairness and ethical considerations in AI-driven educational tools, particularly how they affect diverse socio-economic groups.Practical implicationsPractically, this research underscores the necessity of designing AI-driven learning platforms that genuinely empower learner autonomy and critical thinking rather than restrict or overly dictate educational pathways. Educational practitioners and technology developers should implement transparent AI decision-making processes, enabling learners to understand and critically evaluate algorithmic recommendations. Policies must prioritize digital infrastructure expansion and digital literacy programs to reduce urban-rural digital disparities. Furthermore, pedagogical practices should integrate human-centric approaches to complement AI tools, ensuring that AI technologies serve to enhance, rather than replace, essential human interactions within collaborative learning environments.Social implicationsThe study emphasizes that AI-driven education holds significant potential for democratizing lifelong learning by offering personalized and adaptive learning experiences. However, it also warns of the risks associated with reinforcing existing socio-economic and digital inequalities, particularly between urban and rural learners. Addressing algorithmic biases and digital divides through targeted policy interventions, infrastructural investments and inclusive educational practices is critical for fostering equity and inclusivity in education. By ensuring that AI platforms are accessible and beneficial for all learner demographics, societies can leverage technology as a tool for broad-based educational empowerment and social development rather than a mechanism perpetuating exclusion.Originality/valueThis research contributes original insights into the nuanced impacts of AI-driven learning on learner autonomy, collaborative interactions and digital equity within the Thai educational context – areas relatively underexplored in existing literature. By uniquely integrating actor-network theory and posthumanist perspectives, it offers a sophisticated understanding of the interplay between humans and technologies in educational settings. The empirical findings deliver crucial evidence-based recommendations for educators, policymakers and developers aiming to ethically and inclusively leverage AI in education. Ultimately, the study enriches ongoing discussions on balancing technological advancement with human-centric, equitable and socially responsible educational practices.

The Role of Artificial Intelligence in the Evaluation of Prostate Pathology.

Artificial intelligence (AI) is an emerging tool in diagnostic pathology, including prostate pathology. This review summarizes the possibilities offered by AI and also discusses the challenges and risks. AI has the potential to assist in the diagnosis and grading of prostate cancer. Diagnostic safety can be enhanced by avoiding the accidental underdiagnosis of small lesions. Another possible benefit is a greater degree of standardization of grading. AI for clinical use needs to be trained on large, high-quality data sets that have been assessed by experienced pathologists. A problem with the use of AI in prostate pathology is the plethora of benign mimics of prostate cancer and morphological variants of cancer that are too unusual to allow sufficient training of AI. AI systems need to be able to account for variations in local routines for cutting, staining, and scanning of slides. We also need to be aware of the risk that users will rely too much on the output of an AI system, leading to diagnostic errors and loss of clinical competence. The reporting pathologist must ultimately be responsible for accepting or rejecting the diagnosis proposed by AI.

Custom AI Protocols for Post-Surgical Recovery

Post-surgical recovery plays a crucial role in determining patient outcomes, yet traditional recovery protocols often fail to account for the individual variations in patient response. These standardized approaches can lead to longer recovery times, increased complications, and lower patient satisfaction. Artificial Intelligence (AI) offers a promising solution to these challenges by enabling personalized, data-driven recovery plans that can be dynamically adjusted based on real-time patient data. This paper explores the role of AI in transforming post-surgical recovery, focusing on the development and implementation of custom AI protocols. AI systems can integrate data from various sources such as wearable devices, electronic health records (EHRs), and mobile health applications to monitor key recovery metrics like pain levels, mobility, and wound healing. By continuously analyzing these data points, AI can tailor recovery protocols to individual patients, optimizing treatment plans, improving adherence, and predicting potential complications. This dynamic approach has the potential to reduce recovery time, prevent readmissions, and enhance overall patient satisfaction. However, the integration of AI in post-surgical recovery also presents challenges, including concerns about data privacy, algorithmic bias, and the need for clinical validation. Ethical considerations regarding informed consent and transparency are also discussed. The paper concludes by exploring future directions for AI in healthcare, including the integration of AI with robotic surgery, telehealth platforms, and the development of regulatory frameworks to ensure the safe and equitable adoption of AI in recovery protocols.

Real-Time AI Systems for ICU Risk Monitoring

Intensive Care Units (ICUs) play a critical role in the management of patients with life-threatening conditions. Early identification of risks such as sepsis, respiratory failure, and cardiac arrest can significantly improve patient outcomes. Traditional monitoring systems, however, rely on periodic assessments and manual interventions, which may not be sufficient for timely risk detection. Artificial Intelligence (AI) has emerged as a powerful tool in overcoming these limitations, enabling real-time, data-driven risk prediction and management. This paper explores the role of AI in transforming ICU risk monitoring through real-time predictive analytics. By leveraging machine learning, deep learning, and sensor technologies, AI systems can process vast amounts of patient data, offering early detection of complications, prioritizing high-risk patients, and supporting decision-making. The paper discusses the integration of AI into existing ICU frameworks, the components of AI-powered monitoring systems, their real-world applications, and the impact on patient care. Despite its transformative potential, the paper also addresses the challenges of data privacy, system integration, accuracy, and clinician adoption. Finally, the paper explores future directions, including the potential for AI to integrate with other ICU technologies and predict long-term outcomes. Through continuous innovation and rigorous validation, AI-driven monitoring has the potential to revolutionize ICU care by enhancing patient safety, optimizing resource allocation, and improving clinical outcomes.

AN EXPLORATORY STUDY OF THE IMPACT OF ARTIFICIAL INTELLIGENCE ON THE JUDICIARY PRODUCTIVITY AND EFFICIENCY

This exploratory study investigates the application of artificial intelligence (AI) within the Sergipe State Court, focusing on its impact on efficiency, productivity, and user satisfaction. AI technologies, such as machine learning algorithms and natural language processing (NLP), have streamlined judicial processes, including automated case prioritization, enhanced legal research, and predictive analytics for informed decision-making. The study highlights AI’s potential to ensure consistent and fair judgments, increasing satisfaction among judicial service users in Sergipe. However, it also addresses significant risks, including algorithmic biases and concerns about AI’s trustworthiness and transparency. These risks emphasize the need for careful oversight and monitoring to prevent adverse impacts on justice. According to the exploratory research, the main benefit expected by users is the significant increase in efficiency and productivity, however, the main identified risk lies in concerns about algorithmic biases and trust in AI systems, a factor considered essential or very relevant by a significant portion of respondents. Regarding the services to be prioritized for enhancement with AI, the grouping of documents by similarity emerges as primordial, considered essential or very relevant by most employees. The study concludes with recommendations for integrating AI to improve operational efficiency and the credibility of the Sergipe State Court.

A new problem of evil?

Abstract This article examines whether artificial intelligence (AI)-driven harm can be classified as moral or natural evil, or whether a new category – artificial evil – is needed. Should AI’s harm be seen as a product of human design, thus maintaining moral responsibility for its creators, or whether AI’s autonomous actions resemble natural evil, where harm arises unintentionally? The concept of artificial evil, combining elements of both moral and natural evil, is presented to better address this dilemma. Just as AI is still a form of intelligence (albeit non-biological), artificial evil would still be evil in the sense that it results in real harm or suffering – it is just that this harm is produced by AI systems rather than by nature or human moral agents directly. The discussion further extends into the realm of defence or theodicy, drawing parallels with the Free Will Defence, questioning if AI autonomy may be justified even if it results in harm, much like human free will. Ultimately, the article calls for a re-evaluation of our ethical frameworks and glossary of terms to address the emerging challenges of AI autonomy and its moral implications.

Prediction of Degradation of Concrete Surface Layer Using Neural Networks Applied to Ultrasound Propagation Signals

The aim of this article is the development of a new artificial intelligence (AI) system for the condition assessment of concrete structures. To study the process of concrete degradation, the so-called spatiotemporal waveform profiles were obtained, which are sets of ultrasonic signals acquired by stepwise surface profiling of the concrete surface. The recorded signals at three frequencies, 50, 100 and 200 kHz, were analyzed and informative areas of the signals were identified. The type of the created neural network is a multilayer perceptron. Stochastic gradient descent was chosen as the learning algorithm. Measurement datasets (test, training and validation) were created to determine two factors of interest—the degree of material degradation (three gradations of material weakening) and the thickness (depth) of the degraded layer varied gradually from 3 to 40 mm from the surface. This article proves that the training datasets were sufficient to obtain acceptable results. The built networks correctly predicted the degree of degradation for all elements of the test dataset. The relative error in prediction of a thickness of degraded layer did not exceed 3% in the case of a thickness of 25 mm. It is shown that the results for the Fourier amplitude spectra are significantly worse than the results of neural networks built based on information about the measured signals themselves.

ACM-Assessor: An Artificial Intelligence System for Assessing Angle Closure Mechanisms in Ultrasound Biomicroscopy

Primary angle-closure glaucoma (PACG), characterized by angle closure (AC) with insidious and irreversible progression, requires precise assessment of AC mechanisms for accurate diagnosis and treatment. This study developed an artificial intelligence system, ACM-Assessor, to evaluate AC mechanisms in ultrasound biomicroscopy (UBM) images. A dataset of 8482 UBM images from 1160 patients was retrospectively collected. ACM-Assessor comprises models for pixel-to-physical spacing conversion, anterior chamber angle boundary segmentation, and scleral spur localization, along with three binary classification models to assess pupillary block (PB), thick peripheral iris (TPI), and anteriorly located ciliary body (ALCB). The integrated assessment model classifies AC mechanisms into pure PB, pure non-PB, multiple mechanisms (MM), and others. ACM-Assessor’s evaluation encompassed external testing (2266 images), human–machine competition and assisting beginners’ assessment (an independent test set of 436 images). ACM-Assessor achieved accuracies of 0.924 (PB), 0.925 (TPI), 0.947 (ALCB), and 0.839 (integrated assessment). In man–machine comparisons, the system’s accuracy was comparable to experts (p &gt; 0.05). With model assistance, beginners’ accuracy improved by 0.117 for binary classification and 0.219 for integrated assessment. ACM-Assessor demonstrates expert-level accuracy and enhances beginners’ learning in UBM analysis.