Excessive documentation burden is linked to clinician burnout, thus motivating efforts to reduce burden. Generative artificial intelligence (AI) poses opportunities for burden reduction but requires rigorous assessment. We evaluated the ability of a large language model (LLM) (OpenAI's GPT-4) to interpret various intervention-response relationships presented on nursing flowsheets, assessing performance using MUC-5 evaluation metrics, and compared its assessments to those of nurse expert evaluators. ChatGPT correctly assessed 3 of 14 clinical scenarios, and partially correctly assessed 6 of 14, frequently omitting data from its reasoning. Nurse expert evaluators correctly assessed all relationships and provided additional language reflective of standard nursing practice beyond the intervention-response relationships evidenced in nursing flowsheets. Future work should ensure the training data used for electronic health record (EHR)-integrated LLMs includes all types of narrative nursing documentation that reflect nurses' clinical reasoning, and verification of LLM-based information summarization does not burden end-users.

The aging of society drives a rising demand for geriatric healthcare due to increased care needs and extended hospital stays in old age. Despite strained social security systems, ensuring high-quality medical care requires innovative solutions. Digitalization could be one of them, however older people, who are less digitally active, may not fully recognize its benefits. This study aims to assess digital participation among geriatric hospital patients and their views on continuous vital sign monitoring using wearables.

The survey was conducted at the geriatric hospital "Krankenhaus Rotes Kreuz Lübeck - Geriatriezentrum" to assess the digital participation of higher frailty patients requiring increased care. The questioning occurred between February 13th and March 10th, 2023. The questionnaire included demographic questions, questions about digital participation and digital skills, opinions on continuous monitoring, and a reflection on the impact of the coronavirus pandemic on internet use.

Of the 201 consecutively admitted patients, 52 were excluded from participation in the study based on the inclusion/exclusion criteria, mostly due to illness. Of the remaining 149 invited patients, 66 (44.2%) agreed to be interviewed, mostly females (76%) with an average age of 81.2 years (SD = 7.1). As a result, 68.2% of participants reported online activity, whereby females and those with low education or high age (p = 0.027) were offline more often. On average, 1-2 internet-enabled devices were used. Continuous vital sign monitoring was favoured by 32 participants and 61 expressed no concerns.

Our findings align with previous studies involving participants of comparable age, indicating comparable results, apart from disease-related participation restrictions. However, the significant proportion of patients who did not want to participate (55.7%) and the analysis of the reasons for nonparticipation suggest that the actual number of geriatric patients who do not engage online is higher. While this does not necessarily imply a complete rejection of digital products by this demographic, it highlights the need for greater emphasis on usability, feasibility, and clarification in future endeavors.

Building deep learning models that can rapidly segment whole slide images (WSIs) using only a handful of training samples remains an open challenge in computational pathology. The difficulty lies in the histological images themselves: many morphological structures within a slide are closely related and very similar in appearance, making it difficult to distinguish between them. However, a skilled pathologist can quickly identify the relevant phenotypes. Through years of training, they have learned to organize visual features into a hierarchical taxonomy (e.g., identifying carcinoma versus healthy tissue, or distinguishing regions within a tumor as cancer cells, the microenvironment, …). Thus, each region is associated with multiple labels representing different tissue types. Pathologists typically deal with this by analyzing the specimen at multiple scales and comparing visual features between different magnifications. Inspired by this multi-scale diagnostic workflow, we introduce the Navigator, a vision model that navigates through WSIs like a domain expert: it searches for the region of interest at a low scale, zooms in gradually, and localizes ever finer microanatomical classes. As a result, the Navigator can detect coarse-grained patterns at lower resolution and fine-grained features at higher resolution. In addition, to deal with sparsely annotated samples, we train the Navigator with a novel semi-supervised framework called S5CL v2. The proposed model improves the F1 score by up to 8% on various datasets including our challenging new TCGA-COAD-30CLS and Erlangen cohorts.

The aim of this study was to develop and validate a convolutional neural network (CNN) algorithm for the detection of impacted mandibular third molars in panoramic radiographs and the classification of the surgical extraction difficulty level. A dataset of 1730 panoramic radiographs was collected; 1300 images were allocated to training and 430 to testing. The performance of the model was evaluated using the confusion matrix for multiclass classification, and the actual scores were compared to those of two human experts. The area under the precision-recall curve of the YOLOv5 model ranged from 72% to 89% across the variables in the surgical difficulty index. The area under the receiver operating characteristic curve showed promising results of the YOLOv5 model for classifying third molars into three surgical difficulty levels (micro-average AUC 87%). Furthermore, the algorithm scores demonstrated good agreement with the human experts. In conclusion, the YOLOv5 model has the potential to accurately detect and classify the position of mandibular third molars, with high performance for every criterion in radiographic images. The proposed model could serve as an aid in improving clinician performance and could be integrated into a screening system.

Artificial Intelligence (AI) holds immense potential to transform healthcare, yet progress is often hindered by the reliance on large labeled datasets and unimodal data. Multimodal Foundation Models (FMs), particularly those leveraging Self-Supervised Learning (SSL) on multimodal data, offer a paradigm shift towards label-efficient, holistic patient modeling. However, the rapid emergence of these complex models has created a fragmented landscape. Here, we provide a systematic review of multimodal FMs for medical imaging applications. Through rigorous screening of 1,144 publications (2012-2024) and in-depth analysis of 48 studies, we establish a unified terminology and comprehensively assess the current state-of-the-art. Our review aggregates current knowledge, critically identifies key limitations and underexplored opportunities, and culminates in actionable guidelines for researchers, clinicians, developers, and policymakers. This work provides a crucial roadmap to navigate and accelerate the responsible development and clinical translation of next-generation multimodal AI in healthcare.

The rapid progress in the development of artificial intelligence (AI) is having a substantial impact on health care (HC) delivery and the physician-patient interaction.

This scoping review aims to offer a thorough analysis of the current status of integrating AI into medical practice as well as the apprehensions expressed by HC professionals (HCPs) over its application.

This scoping review used the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines to examine articles that investigated the apprehensions of HCPs about medical AI. Following the application of inclusion and exclusion criteria, 32 of an initial 217 studies (14.7%) were selected for the final analysis. We aimed to develop an attitude range that accurately captured the unfavorable emotions of HCPs toward medical AI. We achieved this by selecting attitudes and ranking them on a scale that represented the degree of aversion, ranging from mild skepticism to intense fear. The ultimate depiction of the scale was as follows: skepticism, reluctance, anxiety, resistance, and fear.

In total, 3 themes were identified through the process of thematic analysis. National surveys performed among HCPs aimed to comprehensively analyze their current emotions, worries, and attitudes regarding the integration of AI in the medical industry. Research on technostress primarily focused on the psychological dimensions of adopting AI, examining the emotional reactions, fears, and difficulties experienced by HCPs when they encountered AI-powered technology. The high-level perspective category included studies that took a broad and comprehensive approach to evaluating overarching themes, trends, and implications related to the integration of AI technology in HC. We discovered 15 sources of attitudes, which we classified into 2 distinct groups: intrinsic and extrinsic. The intrinsic group focused on HCPs' inherent professional identity, encompassing their tasks and capacities. Conversely, the extrinsic group pertained to their patients and the influence of AI on patient care. Next, we examined the shared themes and made suggestions to potentially tackle the problems discovered. Ultimately, we analyzed the results in relation to the attitude scale, assessing the degree to which each attitude was portrayed.

The solution to addressing resistance toward medical AI appears to be centered on comprehensive education, the implementation of suitable legislation, and the delineation of roles. Addressing these issues may foster acceptance and optimize AI integration, enhancing HC delivery while maintaining ethical standards. Due to the current prominence and extensive research on regulation, we suggest that further research could be dedicated to education.

Background: Technology is rapidly being developed to improve healthcare outcomes. However, the attitudes and perceptions of occupational therapists (OTs) on artificial intelligence (AI) in healthcare are not yet known. Purpose: This study aims to: explore Canadian OTs' (a) understanding and knowledge on AI, (b) opinions and perspectives on AI, and (c) perceptions of potential benefits and risks AI might bring to occupational therapy practice in Canada. Method: A sequential explanatory mixed method approach was used to gather perspectives of Canadian registered OTs. Two hundred and eighty-two survey respondents and 15 focus group participants took part in the study. Findings: Three main themes emerged: "AI Knowledge and Implementation," "Use of AI in Occupational Therapy," and "Human vs. Machine." OTs have various levels of understanding of AI, and its capabilities within practice and are open to AI use in practice. Although ethical concerns must be addressed, OTs do not perceive AI to pose a threat to employment. Conclusion: OTs have the ability to implement and guide policy changes for technology adoption, and understanding their current perspectives creates opportunities to advocate for change in the field. Further education is needed to better prepare professionals for clinical usage of AI.

Considering that most patients with low or no significant risk factors can safely undergo noncardiac surgery without additional cardiac evaluation, and given the excessive evaluations often performed in patients undergoing intermediate or higher risk noncardiac surgeries, practical preoperative risk assessment tools are essential to reduce unnecessary delays for urgent outpatient services and manage medical costs more efficiently.

This study aimed to use the Observational Medical Outcomes Partnership Common Data Model to develop a predictive model by applying machine learning algorithms that can effectively predict major adverse cardiac and cerebrovascular events (MACCE) in patients undergoing noncardiac surgery.

This retrospective observational network study collected data by converting electronic health records into a standardized Observational Medical Outcomes Partnership Common Data Model format. The study was conducted in 2 tertiary hospitals. Data included demographic information, diagnoses, laboratory results, medications, surgical types, and clinical outcomes. A total of 46,225 patients were recruited from Seoul National University Bundang Hospital and 396,424 from Asan Medical Center. We selected patients aged 65 years and older undergoing noncardiac surgeries, excluding cardiac or emergency surgeries, and those with less than 30 days of observation. Using these observational health care data, we developed machine learning-based prediction models using the observational health data sciences and informatics open-source patient-level prediction package in R (version 4.1.0; R Foundation for Statistical Computing). A total of 5 machine learning algorithms, including random forest, were developed and validated internally and externally, with performance assessed through the area under the receiver operating characteristic curve (AUROC), the area under the precision-recall curve, and calibration plots.

All machine learning prediction models surpassed the Revised Cardiac Risk Index in MACCE prediction performance (AUROC=0.704). Random forest showed the best results, achieving AUROC values of 0.897 (95% CI 0.883-0.911) internally and 0.817 (95% CI 0.815-0.819) externally, with an area under the precision-recall curve of 0.095. Among 46,225 patients of the Seoul National University Bundang Hospital, MACCE occurred in 4.9% (2256/46,225), including myocardial infarction (907/46,225, 2%) and stroke (799/46,225, 1.7%), while in-hospital mortality was 0.9% (419/46,225). For Asan Medical Center, 6.3% (24,861/396,424) of patients experienced MACCE, with 1.5% (6017/396,424) stroke and 3% (11,875/396,424) in-hospital mortality. Furthermore, the significance of predictors linked to previous diagnoses and laboratory measurements underscored their critical role in effectively predicting perioperative risk.

Our prediction models outperformed the widely used Revised Cardiac Risk Index in predicting MACCE within 30 days after noncardiac surgery, demonstrating superior calibration and generalizability across institutions. Its use can optimize preoperative evaluations, minimize unnecessary testing, and streamline perioperative care, significantly improving patient outcomes and resource use. We anticipate that applying this model to actual electronic health records will benefit clinical practice.

This systematic literature review examines the quality of early clinical evaluation of artificial intelligence (AI) decision support systems (DSS) reported in glaucoma care. Artificial Intelligence applications within glaucoma care are increasing within the literature. For such DSS, there needs to be standardised reporting to enable faster clinical adaptation. In May 2022, a checklist to facilitate reporting of early AI studies (DECIDE-AI) was published and adopted by the EQUATOR network.

The Cochrane Library, Embase, Ovid MEDLINE, PubMed, SCOPUS, and Web of Science Core Collection were searched for studies published between January 2020 and May 2023 that reported clinical evaluation of DSS for the diagnosis of glaucoma or for identifying the progression of glaucoma driven by AI. PRISMA guidelines were followed (PROSPERO registration: CRD42023431343). Study details were extracted and were reviewed against the DECIDE-AI checklist. The AI-Specific Score, Generic-Item Score, and DECIDE-AI Score were generated.

A total of 1,552 records were screened, with 19 studies included within the review. All studies discussed an early clinical evaluation of AI use within glaucoma care, as defined by the a priori study protocol. Overall, the DECIDE-AI adherence score was low, with authors under reporting the AI specific items (30.3%), whilst adhering well to the generic reporting items (84.7%).

Overall, reporting of important aspects of AI studies was suboptimal. Encouraging editors and authors to incorporate the checklist will enhance standardised reporting, bolstering the evidence base for integrating AI DSS into glaucoma care workflows, thus help improving patient care and outcomes.

The integration of big data and artificial intelligence (AI) in healthcare, particularly through the analysis of electronic health records (EHR), presents significant opportunities for improving diagnostic accuracy and patient outcomes. However, the challenge of processing and accurately labeling vast amounts of unstructured data remains a critical bottleneck, necessitating efficient and reliable solutions. This study investigates the ability of domain specific, fine-tuned large language models (LLMs) to classify unstructured EHR texts with typographical errors through named entity recognition tasks, aiming to improve the efficiency and reliability of supervised learning AI models in healthcare.

Turkish clinical notes from pediatric emergency room admissions at Hacettepe University İhsan Doğramacı Children's Hospital from 2018 to 2023 were analyzed. The data were preprocessed with open source Python libraries and categorized using a pretrained GPT-3 model, "text-davinci-003," before and after fine-tuning with domain-specific data on respiratory tract infections (RTI). The model's predictions were compared against ground truth labels established by pediatric specialists.

Out of 24,229 patient records classified as poorly labeled, 18,879 were identified without typographical errors and confirmed for RTI through filtering methods. The fine-tuned model achieved a 99.88% accuracy, significantly outperforming the pretrained model's 78.54% accuracy in identifying RTI cases among the remaining records. The fine-tuned model demonstrated superior performance metrics across all evaluated aspects compared to the pretrained model.

Fine-tuned LLMs can categorize unstructured EHR data with high accuracy, closely approximating the performance of domain experts. This approach significantly reduces the time and costs associated with manual data labeling, demonstrating the potential to streamline the processing of large-scale healthcare data for AI applications.

Artificial intelligence (AI), a branch of computer science, has been of growing research interest since its introduction to healthcare disciplines in the 1970s. Research has demonstrated that the application of such technologies has allowed for greater task accuracy and efficiency in medical disciplines such as diagnostics, treatment protocols and clinical decision-making. Application in pharmacy practice is reportedly narrower in scope; with greater emphasis placed on stock management and day-to-day function optimisation than enhancing patient outcomes. Despite this, new studies are underway to explore how AI technologies may be utilised in areas such as pharmacist interventions, medication adherence, and personalised medicine. Objective/s: The aim of this study was to identify current use of AI in measuring performance outcomes in pharmacy practice.

A scoping review was conducted in accordance with PRISMA Extension for Scoping Reviews (PRISMA-ScR). A comprehensive literature search was conducted in MEDLINE, Embase, IPA (International Pharmaceutical Abstracts), and Web of Science databases for articles published between January 1, 2018 to September 11, 2023, relevant to the aim. The final search strategy included the following terms: ("artificial intelligence") AND ("pharmacy" OR "pharmacist" OR "pharmaceutical service" OR "pharmacy service"). Reference lists of identified review articles were also screened.

The literature search identified 560 studies, of which seven met the inclusion criteria. These studies described the use of AI in pharmacy practice. All seven studies utilised models derived from machine learning AI techniques. AI identification of prescriptions requiring pharmacist intervention was the most frequent (n = 4), followed by screening services (n = 2), and patient-facing mobile applications (n = 1). These results indicated a workflow- and productivity-focused application of AI within current pharmacy practice, with minimal intention for direct patient health outcome improvement. Despite this, the review also revealed AI's potential in data collation and analytics to aid in pharmacist contribution towards the healthcare team and improvement of health outcomes.

This scoping review has identified, from the literature available, three main areas of focus, (1) identification and classification of atypical or inappropriate medication orders, (2) improving efficiency of mass screening services, and (3) improving adherence and quality use of medicines. It also identified gaps in AI's current utility within the profession and its potential for day-to-day practice, as our understanding of general AI techniques continues to advance.

To understand whether art and technology (mainly conversational agents) may help oncology patients to experience a more humanized journey.

This narrative review encompasses a comprehensive examination of the existing literature in this field by a multicenter, multidisciplinary, and multiprofessional team aiming to analyze the current developments and potential future directions of using art and technology for patient engagement.

We identified three major themes of patient engagement with art and three major themes of patient engagement with technologies. Two real-case scenarios are reported from our experience to practically envision how findings from the literature can be implemented in different contexts.

Art therapy and technologies can be ancillary supports for healthcare professionals but are not substitutive of their expertise and responsibilities. Such tools may help to convey a more empathetic and uplifting patient journey if properly integrated within clinical practice, whereby the humanistic touch of medicine remains pivotal.

Community-based high intensity services for people living with severe and enduring mental illnesses face critical workforce shortages and workflow efficiency challenges. The expectation to monitor complex, dynamic patient data from ever-expanding electronic health records leads to information overload, a significant factor contributing to worker burnout and attrition. An algorithmic workforce, defined as a suite of algorithm-driven processes, can work alongside health professionals assisting with oversight tasks and augmenting human expertise. This selective review summarises lessons learned from our five-year experience (2018-22) of algorithmic workforce implementation research in two community mental health services in Australia covering both rural and urban populations. We retrace our implementation journey to illustrate four foundational processes: (i) algorithm design (ii) proof-of-concept validation (iii) workflow integration and (iv) optimization. By examining our previous studies, we discuss insights gained regarding intended human-centricity of services, potential algorithm-human misalignments, and unintended workload and accountability consequences for clinicians and organizations.

With rapidly evolving artificial intelligence solutions, healthcare organizations need an implementation roadmap. A "clinical trials" informed approach can promote safe and impactful implementation of artificial intelligence. This framework includes four phases: (1) Safety; (2) Efficacy; (3) Effectiveness and comparison to an existing standard; and (4) Monitoring. Combined with inter-institutional collaboration and national funding support, this approach will advance safe, usable, effective, and equitable deployments of artificial intelligence in healthcare.

Artificial intelligence (AI) and large language models (LLMs), such as OpenAI's GPT-4, are increasingly being explored for medical applications. Recently, GPT-4 gained image processing capabilities, enabling it to handle tasks such as image captioning, visual question answering, and potentially interpreting medical data. Despite promising potential in diagnostics, the effectiveness of GPT-4 in interpreting complex 12-lead electrocardiograms (ECGs) remains to be assessed.

This study utilized GPT-4 to interpret 150 12-lead ECGs from the Cardiology Research Dubrava (CaRD) registry, spanning a wide range of cardiac pathologies. The ECGs were classified into 4 categories for analysis: arrhythmias, conduction system abnormalities, acute coronary syndrome, and other. Two experiments were conducted: one where GPT-4 interpreted ECGs without clinical context, and another with added clinical scenarios. A panel of experienced cardiologists evaluated the accuracy of GPT-4's interpretations.

In this cross-sectional observational study, GPT-4 demonstrated a correct interpretation rate of 19% without clinical context and a significantly improved rate of 45% with context (P < 0.001). The addition of clinical scenarios significantly enhanced interpretative accuracy, particularly in the acute coronary syndrome category (10% vs 70%; P < 0.0.01). The "other" category showed no impact (51% vs 59%; P = 0.640), and trends toward significance were observed in the arrhythmias (9.7% vs 32%; P = 0.059) and conduction system abnormalities (4.8% vs 19%; P = 0.088) categories when given clinical context.

Although GPT-4 shows potential in aiding 12-lead ECG interpretation, its effectiveness varies significantly with clinical context. The study suggests that GPT-4 alone in its current form may not provide accurate 12-lead ECG interpretation.

Radiology, as a highly technical and information-rich medical specialty, is well suited for artificial intelligence (AI) product development, and many U.S. FDA-cleared AI medical devices are authorized for uses within the specialty. In this Clinical Perspective, we discuss the deployment of AI tools in radiology, exploring regulatory processes, the need for transparency, and other practical challenges. We further highlight the importance of rigorous validation, real-world testing, seamless workflow integration, and end user education. We emphasize the role for continuous feedback and robust monitoring processes, to guide AI tools' adaptation and help ensure sustained performance. Traditional standalone and alternative platform-based approaches to radiology AI implementation are considered. The presented strategies will help achieve successful deployment and fully realize AI's potential benefits in radiology.

The application of artificial intelligence (AI)-based clinical decision support systems (CDSS) in the healthcare domain is still limited. End-users' difficulty understanding how the outputs of opaque black AI models are generated contributes to this. It is still unknown which explanations are best presented to end users and how to design the interfaces they are presented in (explanation user interface, XUI). This article aims to provide an overview of recommendations for the user-centered design of XUIs from the scientific literature.

A scoping review was conducted to identify recommendations for the design of XUIs. Articles published between 2017 and 2022 in English or German, presenting original research or literature reviews, focusing on XUIs for end users or domain experts, which are intended for presentation in graphical user interfaces and from which recommendations could be extracted were included in the review. Articles were retrieved from Scopus, Web of Science, IEEE Explore, PubMed, ACM Digital Library, and PsychInfo. A mind map was created to organize and summarize the identified recommendations.

From the 47 included articles, 240 recommendations for the user-centered design were extracted. The organization in a mind map resulted in 64 summarized recommendations.

This review provides a synopsis of basic recommendations for the user-centered design of XUIs, focusing on the healthcare domain. During the analysis of the articles, it became clear that no specific and directly implementable design recommendations for AI-based CDSS can be given, but only basic recommendations for raising awareness about the user-centered design of XUIs.

"I see, I forget, I read aloud, I remember, and when I do read purposefully by writing it, I do not forget it." This phenomenon is known as "interoception" and refers to the sensing and interpretation of internal body signals, allowing the brain to communicate with various body systems. Dysfunction in interoception is associated with cardiovascular disorders. We delve into the concept of interoception and its impact on heart failure (HF) by reviewing and exploring neural mechanisms underlying interoceptive processing. Furthermore, we review the potential of artificial intelligence (AI) in diagnosis, biomarker development, and HF treatment. In the context of HF, AI algorithms can analyze and interpret complex interoceptive data, providing valuable insights for diagnosis and treatment. These algorithms can identify patterns of disease markers that can contribute to early detection and diagnosis, enabling timely intervention and improved outcomes. These biomarkers hold significant potential in improving the precision/efficacy of HF. Additionally, AI-powered technologies offer promising avenues for treatment. By leveraging patient data, AI can personalize therapeutic interventions. AI-driven technologies such as remote monitoring devices and wearable sensors enable the monitoring of patients' health. By harnessing the power of AI, we should aim to advance the diagnosis and treatment strategies for HF. This review explores the potential of AI in diagnosing, developing biomarkers, and managing HF.

Pregnancies impacted by diabetes face the compounded challenge of strict glycemic control with mounting insulin resistance as the pregnancy progresses. New technological advances, including artificial intelligence (AI) and the Internet of Medical Things (IoMT), are revolutionizing health care delivery by providing innovative solutions for diabetes care during pregnancy. Together, AI and the IoMT are a multibillion-dollar industry that integrates advanced medical devices and sensors into a connected network that enables continuous monitoring of glucose levels. AI-driven clinical decision support systems (CDSSs) can predict glucose trends and provide tailored evidence-based treatments with real-time adjustments as insulin resistance changes with placental growth. Additionally, mobile health (mHealth) applications facilitate patient education and self-management through real-time tracking of diet, physical activity, and glucose levels. Remote monitoring capabilities are particularly beneficial for pregnant persons with diabetes as they extend quality care to underserved populations and reduce the need for frequent in-person visits. This high-resolution monitoring allows physicians and patients access to an unprecedented wealth of data to make more informed decisions based on real-time data, reducing complications for both the mother and fetus. These technologies can potentially improve maternal and fetal outcomes by enabling timely, individualized interventions based on personalized health data. While AI and IoMT offer significant promise in enhancing diabetes care for improved maternal and fetal outcomes, their implementation must address challenges such as data security, cost-effectiveness, and preserving the essential patient-provider relationship. KEY POINTS: · The IoMT expands how patients interact with their health care.. · AI has widespread application in the care of pregnancies complicated by diabetes.. · A need for validation and black-box methodologies challenges the application of AI-based tools.. · As research in AI grows, considerations for data privacy and ethical dilemmas will be required..

Artificial intelligence (AI) has become an omnipresent topic in the media. Lively discussions are being held on how AI could revolutionize the global healthcare landscape. The development of innovative AI models, including in the medical sector, is increasingly dominated by large high-tech companies. As a global technology epicenter, Silicon Valley hosts many of these technological giants which are muscling their way into healthcare provision with their advanced technologies. The annual conference of the American College of Obstetrics and Gynecology (ACOG) was held in San Francisco from 17 - 19 May 2024. ACOG celebrated its AI premier, hosting two sessions on current AI topics in gynecology at their annual conference. This paper provides an overview of the topics discussed and permits an insight into the thinking in Silicon Valley, showing how technology companies grow and fail there and examining how our American colleagues perceive increased integration of AI in gynecological and obstetric care. In addition to the classification of various, currently popular AI terms, the article also presents three areas where artificial intelligence is being used in gynecology and looks at the current developmental status in the context of existing obstacles to implementation and the current digitalization status of the German healthcare system.

Traditional medical artificial intelligence models that are approved for clinical use restrict themselves to single-modal data ( e.g ., images only), limiting their applicability in the complex, multimodal environment of medical diagnosis and treatment. Multimodal transformer models in health care can effectively process and interpret diverse data forms, such as text, images, and structured data. They have demonstrated impressive performance on standard benchmarks, like United States Medical Licensing Examination question banks, and continue to improve with scale. However, the adoption of these advanced artificial intelligence models is not without challenges. While multimodal deep learning models like transformers offer promising advancements in health care, their integration requires careful consideration of the accompanying ethical and environmental challenges.

Artificial Intelligence (AI) has made a profound impact on the medical sector, particularly in radiology. The integration of AI knowledge into medical education is essential to equip future healthcare professionals with the skills needed to effectively leverage these advancements in their practices. Despite its significance, many medical schools have yet to incorporate AI into their curricula. This study aims to assess the attitudes of medical students in Thailand toward AI and its application in radiology, with the objective of better planning for its inclusion.

Between February and June 2022, we conducted a survey in two Thai medical schools: Chiang Mai University in Northern Thailand and Prince of Songkla University in Southern Thailand. We employed 5-point Likert scale questions (ranging from strongly agree to strongly disagree) to evaluate students' opinions on three main aspects: (1) their understanding of AI, (2) the inclusion of AI in their medical education, and (3) the potential impact of AI on medicine and radiology.

Our findings revealed that merely 31% of medical students perceived to have a basic understanding of AI. Nevertheless, nearly all students (93.6%) recognized the value of AI training for their careers and strongly advocated for its inclusion in the medical school curriculum. Furthermore, those students who had a better understanding of AI were more likely to believe that AI would revolutionize the field of radiology (p = 0.02), making it more captivating and impactful (p = 0.04).

Our study highlights a noticeable gap in the understanding of AI among medical students in Thailand and its practical applications in healthcare. However, the overwhelming consensus among these students is their readiness to embrace the incorporation of AI training into their medical education. This enthusiasm holds the promise of enhancing AI adoption, ultimately leading to an improvement in the standard of healthcare services in Thailand, aligning with the country's healthcare vision.

The increasing torrents of health AI innovations hold promise for facilitating the delivery of patient-centered care. Yet the enablement and adoption of AI innovations in the healthcare and life science industries can be challenging with the rising concerns of AI risks and the potential harms to health equity. This paper describes Ethicara, a system that enables health AI risk assessment for responsible AI model development. Ethicara works by orchestrating a collection of self-analytics services that detect and mitigate bias and increase model transparency from harmonized data models. For the lack of risk controls currently in the health AI development and deployment process, the self-analytics tools enhanced by Ethicara are expected to provide repeatable and measurable controls to operationalize voluntary risk management frameworks and guidelines (e.g., NIST RMF, FDA GMLP) and regulatory requirements emerging from the upcoming AI regulations (e.g., EU AI Act, US Blueprint for an AI Bill of Rights). In addition, Ethicara provides plug-ins via which analytics results are incorporated into healthcare applications. This paper provides an overview of Ethicara's architecture, pipeline, and technical components and showcases the system's capability to facilitate responsible AI use, and exemplifies the types of AI risk controls it enables in the healthcare and life science industry.

Large language models (LLM) show potential for decision support in breast cancer care. Their use in clinical care is currently prohibited by lack of control over sources used for decision-making, explainability of the decision-making process and health data security issues. Recent development of Small Language Models (SLM) is discussed to address these challenges. This preclinical proof-of-concept study tailors an open-source SLM to the German breast cancer guideline (BC-SLM) to evaluate initial clinical accuracy and technical functionality in a preclinical simulation.

A multidisciplinary tumor board (MTB) is used as the gold-standard to assess the initial clinical accuracy in terms of concordance of the BC-SLM with MTB and comparing it to two publicly available LLM, ChatGPT3.5 and 4. The study includes 20 fictional patient profiles and recommendations for 5 treatment modalities, resulting in 100 binary treatment recommendations (recommended or not recommended). Statistical evaluation includes concordance with MTB in % including Cohen's Kappa statistic (κ). Technical functionality is assessed qualitatively in terms of local hosting, adherence to the guideline and information retrieval.

The overall concordance amounts to 86% for BC-SLM (κ = 0.721, p < 0.001), 90% for ChatGPT4 (κ = 0.820, p < 0.001) and 83% for ChatGPT3.5 (κ = 0.661, p < 0.001). Specific concordance for each treatment modality ranges from 65 to 100% for BC-SLM, 85-100% for ChatGPT4, and 55-95% for ChatGPT3.5. The BC-SLM is locally functional, adheres to the standards of the German breast cancer guideline and provides referenced sections for its decision-making.

The tailored BC-SLM shows initial clinical accuracy and technical functionality, with concordance to the MTB that is comparable to publicly-available LLMs like ChatGPT4 and 3.5. This serves as a proof-of-concept for adapting a SLM to an oncological disease and its guideline to address prevailing issues with LLM by ensuring decision transparency, explainability, source control, and data security, which represents a necessary step towards clinical validation and safe use of language models in clinical oncology.

Artificial intelligence (AI) and machine learning (ML) research within medicine has exponentially increased over the last decade, with studies showcasing the potential of AI/ML algorithms to improve clinical practice and outcomes. Ongoing research and efforts to develop AI-based models have expanded to aid in the identification of inborn errors of immunity (IEI). The use of larger electronic health record data sets, coupled with advances in phenotyping precision and enhancements in ML techniques, has the potential to significantly improve the early recognition of IEI, thereby increasing access to equitable care. In this review, we provide a comprehensive examination of AI/ML for IEI, covering the spectrum from data preprocessing for AI/ML analysis to current applications within immunology, and address the challenges associated with implementing clinical decision support systems to refine the diagnosis and management of IEI.

Evaluation of artificial intelligence (AI) tools in clinical trials remains the gold standard for translation into clinical settings. However, design factors associated with successful trial completion and the common reasons for trial failure are unknown.

This study aims to compare trial design factors of complete and incomplete clinical trials testing AI tools. We conducted a case-control study of complete (n=485) and incomplete (n=51) clinical trials that evaluated AI as an intervention of ClinicalTrials.gov.

Trial design factors, including area of clinical application, intended use population, and intended role of AI, were extracted. Trials that did not evaluate AI as an intervention and active trials were excluded. The assessed trial design factors related to AI interventions included the domain of clinical application related to organ systems; intended use population for patients or health care providers; and the role of AI for different applications in patient-facing clinical workflows, such as diagnosis, screening, and treatment. In addition, we also assessed general trial design factors including study type, allocation, intervention model, masking, age, sex, funder, continent, length of time, sample size, number of enrollment sites, and study start year. The main outcome was the completion of the clinical trial. Odds ratio (OR) and 95% CI values were calculated for all trial design factors using propensity-matched, multivariable logistic regression.

We queried ClinicalTrials.gov on December 23, 2023, using AI keywords to identify complete and incomplete trials testing AI technologies as a primary intervention, yielding 485 complete and 51 incomplete trials for inclusion in this study. Our nested propensity-matched, case-control results suggest that trials conducted in Europe were significantly associated with trial completion when compared with North American trials (OR 2.85, 95% CI 1.14-7.10; P=.03), and the trial sample size was positively associated with trial completion (OR 1.00, 95% CI 1.00-1.00; P=.02).

Our case-control study is one of the first to identify trial design factors associated with completion of AI trials and catalog study-reported reasons for AI trial failure. We observed that trial design factors positively associated with trial completion include trials conducted in Europe and sample size. Given the promising clinical use of AI tools in health care, our results suggest that future translational research should prioritize addressing the design factors of AI clinical trials associated with trial incompletion and common reasons for study failure.

Quality assessment of diagnostic accuracy studies (QUADAS), and more recently QUADAS-2, were developed to aid the evaluation of methodological quality within primary diagnostic accuracy studies. However, its current form, QUADAS-2 does not address the unique considerations raised by artificial intelligence (AI)-centered diagnostic systems. The rapid progression of the AI diagnostics field mandates suitable quality assessment tools to determine the risk of bias and applicability, and subsequently evaluate translational potential for clinical practice.

We aim to develop an AI-specific QUADAS (QUADAS-AI) tool that addresses the specific challenges associated with the appraisal of AI diagnostic accuracy studies. This paper describes the processes and methods that will be used to develop QUADAS-AI.

The development of QUADAS-AI can be distilled into 3 broad stages. Stage 1-a project organization phase had been undertaken, during which a project team and a steering committee were established. The steering committee consists of a panel of international experts representing diverse stakeholder groups. Following this, the scope of the project was finalized. Stage 2-an item generation process will be completed following (1) a mapping review, (2) a meta-research study, (3) a scoping survey of international experts, and (4) a patient and public involvement and engagement exercise. Candidate items will then be put forward to the international Delphi panel to achieve consensus for inclusion in the revised tool. A modified Delphi consensus methodology involving multiple online rounds and a final consensus meeting will be carried out to refine the tool, following which the initial QUADAS-AI tool will be drafted. A piloting phase will be carried out to identify components that are considered to be either ambiguous or missing. Stage 3-once the steering committee has finalized the QUADAS-AI tool, specific dissemination strategies will be aimed toward academic, policy, regulatory, industry, and public stakeholders, respectively.

As of July 2024, the project organization phase, as well as the mapping review and meta-research study, have been completed. We aim to complete the item generation, including the Delphi consensus, and finalize the tool by the end of 2024. Therefore, QUADAS-AI will be able to provide a consensus-derived platform upon which stakeholders may systematically appraise the methodological quality associated with AI diagnostic accuracy studies by the beginning of 2025.

AI-driven systems comprise an increasingly significant proportion of research in clinical diagnostics. Through this process, QUADAS-AI will aid the evaluation of studies in this domain in order to identify bias and applicability concerns. As such, QUADAS-AI may form a key part of clinical, governmental, and regulatory evaluation frameworks for AI diagnostic systems globally.

DERR1-10.2196/58202.

In the high-stakes realm of critical care, where daily decisions are crucial and clear communication is paramount, comprehending the rationale behind Artificial Intelligence (AI)-driven decisions appears essential. While AI has the potential to improve decision-making, its complexity can hinder comprehension and adherence to its recommendations. "Explainable AI" (XAI) aims to bridge this gap, enhancing confidence among patients and doctors. It also helps to meet regulatory transparency requirements, offers actionable insights, and promotes fairness and safety. Yet, defining explainability and standardising assessments are ongoing challenges and balancing performance and explainability can be needed, even if XAI is a growing field.

Machine learning (ML) and prediction modelling have become increasingly influential in healthcare, providing critical insights and supporting clinical decisions, particularly in the age of big data. This paper serves as an introductory guide for health researchers and readers interested in prediction modelling and explores how these technologies support clinical decisions, particularly with big data, and covers all aspects of the development, assessment and reporting of a model using ML. The paper starts with the importance of prediction modelling for precision medicine. It outlines different types of prediction and machine learning approaches, including supervised, unsupervised and semi-supervised learning, and provides an overview of popular algorithms for various outcomes and settings. It also introduces key theoretical ML concepts. The importance of data quality, preprocessing and unbiased model performance evaluation is highlighted. Concepts of apparent, internal and external validation will be introduced along with metrics for discrimination and calibration for different types of outcomes. Additionally, the paper addresses model interpretation, fairness and implementation in clinical practice. Finally, the paper provides recommendations for reporting and identifies common pitfalls in prediction modelling and machine learning. The aim of the paper is to help readers understand and critically evaluate research papers that present ML models and to serve as a first guide for developing, assessing and implementing their own.

Computer-aided detection (CAD) can help identify people with active tuberculosis left undetected. However, few studies have compared the performance of commercially available CAD products for screening in high tuberculosis and high HIV settings, and there is poor understanding of threshold selection across products in different populations. We aimed to compare CAD products' performance, with further analyses on subgroup performance and threshold selection.

We evaluated 12 CAD products on a case-control sample of participants from a South African tuberculosis prevalence survey. Only those with microbiological test results were eligible. The primary outcome was comparing products' accuracy using the area under the receiver operating characteristic curve (AUC) against microbiological evidence. Threshold analyses were performed based on pre-defined criteria and across all thresholds. We conducted subgroup analyses including age, gender, HIV status, previous tuberculosis history, symptoms presence, and current smoking status.

Of the 774 people included, 516 were bacteriologically negative and 258 were bacteriologically positive. Diverse accuracy was noted: Lunit and Nexus had AUCs near 0·9, followed by qXR, JF CXR-2, InferRead, Xvision, and ChestEye (AUCs 0·8-0·9). XrayAME, RADIFY, and TiSepX-TB had AUC under 0·8. Thresholds varied notably across these products and different versions of the same products. Certain products (Lunit, Nexus, JF CXR-2, and qXR) maintained high sensitivity (>90%) across a wide threshold range while reducing the number of individuals requiring confirmatory diagnostic testing. All products generally performed worst in older individuals, people with previous tuberculosis, and people with HIV. Variations in thresholds, sensitivity, and specificity existed across groups and settings.

Several previously unevaluated products performed similarly to those evaluated by WHO. Thresholds differed across products and demographic subgroups. The rapid emergence of products and versions necessitates a global strategy to validate new versions and software to support CAD product and threshold selections.

Government of Canada.

Background: Aging is a multifaceted process that may lead to an increased risk of developing cancer. Artificial intelligence (AI) applications in clinical cancer research may optimize cancer treatments, improve patient care, and minimize risks, prompting AI to receive high levels of attention in clinical medicine. This systematic review aims to synthesize current articles about the effectiveness of artificial intelligence in cancer treatments for older adults. Methods: We conducted a systematic review by searching CINAHL, PsycINFO, and MEDLINE via EBSCO. We also conducted forward and backward hand searching for a comprehensive search. Eligible studies included a study population of older adults (60 and older) with cancer, used AI technology to treat cancer, and were published in a peer-reviewed journal in English. This study was registered on PROSPERO (CRD42024529270). Results: This systematic review identified seven articles focusing on lung, breast, and gastrointestinal cancers. They were predominantly conducted in the USA (42.9%), with others from India, China, and Germany. The measures of overall and progression-free survival, local control, and treatment plan concordance suggested that AI interventions were equally or less effective than standard care in treating older adult cancer patients. Conclusions: Despite promising initial findings, the utility of AI technologies in cancer treatment for older adults remains in its early stages, as further developments are necessary to enhance accuracy, consistency, and reliability for broader clinical use.

In this paper, we analyse the different advances in artificial intelligence (AI) approaches in multiple sclerosis (MS). AI applications in MS range across investigation of disease pathogenesis, diagnosis, treatment, and prognosis. A subset of AI, Machine learning (ML) models analyse various data sources, including magnetic resonance imaging (MRI), genetic, and clinical data, to distinguish MS from other conditions, predict disease progression, and personalize treatment strategies. Additionally, AI models have been extensively applied to lesion segmentation, identification of biomarkers, and prediction of outcomes, disease monitoring, and management. Despite the big promises of AI solutions, model interpretability and transparency remain critical for gaining clinician and patient trust in these methods. The future of AI in MS holds potential for open data initiatives that could feed ML models and increasing generalizability, the implementation of federated learning solutions for training the models addressing data sharing issues, and generative AI approaches to address challenges in model interpretability, and transparency. In conclusion, AI presents an opportunity to advance our understanding and management of MS. AI promises to aid clinicians in MS diagnosis and prognosis improving patient outcomes and quality of life, however ensuring the interpretability and transparency of AI-generated results is going to be key for facilitating the integration of AI into clinical practice.

Background This study aims to evaluate the performance of ChatGPT in the medical specialization exam (MSE) that medical graduates take when choosing their postgraduate specialization and to reveal how artificial intelligence-supported education can increase the quality and academic success of medical education. The research aims to explore the potential applications and advantages of artificial intelligence in medical education and examine ways in which this technology can contribute to student learning and exam preparation. Methodology A total of 240 MSE questions were posed to ChatGPT, 120 of which were basic medical sciences questions and 120 were clinical medical sciences questions. A total of 18,481 people participated in the exam. The performance of medical school graduates was compared with ChatGPT-3.5 in terms of answering these questions correctly. The average score for ChatGPT-3.5 was calculated by averaging the minimum and maximum scores. Calculations were done using the R.4.0.2 environment. Results The general average score of graduates was a minimum of 7.51 in basic sciences and a maximum of 81.46, while in clinical sciences, the average was a minimum of 12.51 and a maximum of 80.78. ChatGPT, on the other hand, had an average of at least 60.00 in basic sciences and a maximum of 72.00, with an average of at least 66.25 and a maximum of 77.00 in clinical sciences. The rate of correct answers in basic medical sciences for graduates was 43.03%, while for ChatGPT was 60.00%. In clinical medical sciences, the rate of correct answers for graduates was 53.29%, while for ChatGPT was 64.16%. ChatGPT performed best with a 91.66% correct answer rate in Obstetrics and Gynecology and an 86.36% correct answer rate in Medical Microbiology. The least successful area for ChatGPT was Anatomy, with a 28.00% correct answer rate, a subfield of basic medical sciences. Graduates outperformed ChatGPT in the Anatomy and Physiology subfields. Significant differences were found in all comparisons between ChatGPT and graduates. Conclusions This study shows that artificial intelligence models such as ChatGPT can provide significant advantages to graduates, as they score higher than medical school graduates. In terms of these benefits, recommended applications include interactive support, private lessons, learning material production, personalized learning plans, self-assessment, motivation boosting, and 24/7 access, among a variety of benefits. As a result, artificial intelligence-supported education can play an important role in improving the quality of medical education and increasing student success.

Given the current state of medical artificial intelligence (AI) and perceptions towards it, collaborative systems are becoming the preferred choice for clinical workflows. This work aims to address expert interaction with medical AI support systems to gain insight towards how these systems can be better designed with the user in mind. As eye tracking metrics have been shown to be robust indicators of usability, we employ them for evaluating the usability and user interaction with medical AI support systems. We use expert gaze to assess experts' interaction with an AI software for caries detection in bitewing x-ray images. We compared standard viewing of bitewing images without AI support versus viewing where AI support could be freely toggled on and off. We found that experts turned the AI on for roughly 25% of the total inspection task, and generally turned it on halfway through the course of the inspection. Gaze behavior showed that when supported by AI, more attention was dedicated to user interface elements related to the AI support, with more frequent transitions from the image itself to these elements. When considering that expert visual strategy is already optimized for fast and effective image inspection, such interruptions in attention can lead to increased time needed for the overall assessment. Gaze analysis provided valuable insights into an AI's usability for medical image inspection. Further analyses of these tools and how to delineate metrical measures of usability should be developed.

Experts are currently investigating the potential applications of the metaverse in healthcare. The metaverse, a groundbreaking concept that arose in the early 21st century through the fusion of virtual reality and augmented reality technologies, holds promise for transforming healthcare delivery. Alongside its implementation, the issue of digital professionalism in healthcare must be addressed. Digital professionalism refers to the knowledge and skills required by healthcare specialists to navigate digital technologies effectively and ethically. This study aims to identify the core principles of digital professionalism for the use of metaverse in healthcare.

This study utilized a qualitative design and collected data through semi-structured online interviews with 20 medical information and health informatics specialists from various countries (USA, UK, Sweden, Netherlands, Poland, Romania, Italy, Iran). Data analysis was conducted using the open coding method, wherein concepts (codes) related to the themes of digital professionalism for the metaverse in healthcare were assigned to the data. The analysis was performed using the MAXQDA software (VER BI GmbH, Berlin, Germany).

The study revealed ten fundamental principles of digital professionalism for the metaverse in healthcare: Privacy and Security, Informed Consent, Trust and Integrity, Accessibility and Inclusion, Professional Boundaries, Evidence-Based Practice, Continuous Education and Training, Collaboration and Interoperability, Feedback and Improvement, and Regulatory Compliance.

As the metaverse continues to expand and integrate itself into various industries, including healthcare, it becomes vital to establish principles of digital professionalism to ensure ethical and responsible practices. Healthcare professionals can uphold these principles to maintain ethical standards, safeguard patient privacy, and deliver effective care within the metaverse.

The integration of artificial intelligence (AI) into healthcare practice has improved patient management and care. Many clinical laboratory specialties have already integrated AI in diagnostic specialties such as radiology and pathology, where it can assist in image analysis, diagnosis, and clinical reporting. As AI technologies continue to advance, it is crucial for biomedical science students to receive comprehensive education and training in AI concepts and applications and to understand the ethical consequences for such development. This review focus on the importance of integrating AI into biomedical science curricula and proposes strategies to enhance curricula for different specialties to prepare future healthcare workers. Improving the curriculum can be achieved by introducing specific subjects related to AI such as informatics, data sciences, and digital health. However, there are many challenges to enhancing the curriculum with AI. In this narrative review, we discuss these challenges and suggest mitigation strategies.

Digital technology has emerged as a transformative tool in dental implantation, profoundly enhancing accuracy and effectiveness across multiple facets, such as diagnosis, preoperative treatment planning, surgical procedures, and restoration delivery. The multiple integration of radiographic data and intraoral data, sometimes with facial scan data or electronic facebow through virtual planning software, enables comprehensive 3-dimensional visualization of the hard and soft tissue and the position of future restoration, resulting in heightened diagnostic precision. In virtual surgery design, the incorporation of both prosthetic arrangement and individual anatomical details enables the virtual execution of critical procedures (e.g., implant placement, extended applications, etc.) through analysis of cross-sectional images and the reconstruction of 3-dimensional surface models. After verification, the utilization of digital technology including templates, navigation, combined techniques, and implant robots achieved seamless transfer of the virtual treatment plan to the actual surgical sites, ultimately leading to enhanced surgical outcomes with highly improved accuracy. In restoration delivery, digital techniques for impression, shade matching, and prosthesis fabrication have advanced, enabling seamless digital data conversion and efficient communication among clinicians and technicians. Compared with clinical medicine, artificial intelligence (AI) technology in dental implantology primarily focuses on diagnosis and prediction. AI-supported preoperative planning and surgery remain in developmental phases, impeded by the complexity of clinical cases and ethical considerations, thereby constraining widespread adoption.

Digital medicine is already well established in respiratory medicine through remote monitoring digital devices which are used in the day-to-day care of patients with asthma, COPD and sleep disorders. Image recognition software, deployed in thoracic radiology for many applications including lung cancer screening, is another application of digital medicine. Used as clinical decision support, this software will soon become part of day-to-day practice once concerns regarding generalisability have been addressed. Embodied in the electronic health record, digital medicine also plays a substantial role in the day-to-day clinical practice of respiratory medicine. Given the considerable work the electronic health record demands from clinicians, the next tangible impact of digital medicine may be artificial intelligence that aids administration, makes record keeping easier and facilitates better digital communication with patients. Future promises of digital medicine are based on their potential to analyse and characterise the large amounts of digital clinical data that are collected in routine care. Offering the potential to predict outcomes and personalise therapy, there is much to be excited by in this new epoch of innovation. However, these digital tools are by no means a silver bullet. It remains uncertain whether, let alone when, the promises of better models of personalisation and prediction will translate into clinically meaningful and cost-effective products for clinicians.

Given that mental health problems in adolescence may have lifelong impacts, the role of primary care physicians (PCPs) in identifying and managing these issues is important. Artificial Intelligence (AI) may offer solutions to the current challenges involved in mental health care. We therefore explored PCPs' challenges in addressing adolescents' mental health, along with their attitudes towards using AI to assist them in their tasks.

We used purposeful sampling to recruit PCPs for a virtual Focus Group (FG). The virtual FG lasted 75 minutes and was moderated by two facilitators. A life transcription was produced by an online meeting software. Transcribed data was cleaned, followed by a priori and inductive coding and thematic analysis.

We reached out to 35 potential participants via email. Seven agreed to participate, and ultimately four took part in the FG. PCPs perceived that AI systems have the potential to be cost-effective, credible, and useful in collecting large amounts of patients' data, and relatively credible. They envisioned AI assisting with tasks such as diagnoses and establishing treatment plans. However, they feared that reliance on AI might result in a loss of clinical competency. PCPs wanted AI systems to be user-friendly, and they were willing to assist in achieving this goal if it was within their scope of practice and they were compensated for their contribution. They stressed a need for regulatory bodies to deal with medicolegal and ethical aspects of AI and clear guidelines to reduce or eliminate the potential of patient harm.

This study provides the groundwork for assessing PCPs' perceptions of AI systems' features and characteristics, potential applications, possible negative aspects, and requirements for using them. A future study of adolescents' perspectives on integrating AI into mental healthcare might contribute a fuller understanding of the potential of AI for this population.

With the increasing prevalence of autism spectrum disorders (ASD), the importance of early screening and diagnosis has been subject to considerable discussion. Given the subtle differences between ASD children and typically developing children during the early stages of development, it is imperative to investigate the utilization of automatic recognition methods powered by artificial intelligence. We aim to summarize the research work on this topic and sort out the markers that can be used for identification.

We searched the papers published in the Web of Science, PubMed, Scopus, Medline, SpringerLink, Wiley Online Library, and EBSCO databases from 1st January 2013 to 13th November 2023, and 43 articles were included.

These articles mainly divided recognition markers into five categories: gaze behaviors, facial expressions, motor movements, voice features, and task performance. Based on the above markers, the accuracy of artificial intelligence screening ranged from 62.13 to 100%, the sensitivity ranged from 69.67 to 100%, the specificity ranged from 54 to 100%.

Therefore, artificial intelligence recognition holds promise as a tool for identifying children with ASD. However, it still needs to continually enhance the screening model and improve accuracy through multimodal screening, thereby facilitating timely intervention and treatment.

Psychological distress is a major contributor to human physiology and pathophysiology, and it has been linked to several conditions, such as auto-immune diseases, metabolic syndrome, sleep disorders, and suicidal thoughts and inclination. Therefore, early detection and management of chronic stress is crucial for the prevention of several diseases. Artificial intelligence (AI) and Machine Learning (ML) have promoted a paradigm shift in several areas of biomedicine including diagnosis, monitoring, and prognosis of disease. Here, our review aims to present some of the AI and ML applications for solving biomedical issues related to psychological stress. We provide several lines of evidence from previous studies highlighting that AI and ML have been able to predict stress and detect the brain normal states vs. abnormal states (notably, in post-traumatic stress disorder (PTSD)) with accuracy around 90%. Of note, AI/ML-driven technology applied to identify ubiquitously present stress exposure may not reach its full potential, unless future analytics focus on detecting prolonged distress through such technology instead of merely assessing stress exposure. Moving forward, we propose that a new subcategory of AI methods called Swarm Intelligence (SI) can be used towards detecting stress and PTSD. SI involves ensemble learning techniques to efficiently solve a complex problem, such as stress detection, and it offers particular strength in clinical settings, such as privacy preservation. We posit that AI and ML approaches will be beneficial for the medical and patient community when applied to predict and assess stress levels. Last, we encourage additional research to bring AI and ML into the standard clinical practice for diagnostics in the not-too-distant future.

Using the reaction history in logistic regression and machine learning (ML) models to predict penicillin allergy has been reported based on non-US data.

We developed ML positive penicillin allergy testing prediction models from multisite US data.

Retrospective data from 4 US-based hospitals were grouped into 4 datasets: enriched training (1:3 case-control matched cohort), enriched testing, nonenriched internal testing, and nonenriched external testing. ML algorithms were used for model development. We determined area under the curve (AUC) and applied the Shapley Additive exPlanations (SHAP) framework to interpret risk drivers.

Of 4777 patients (mean age 60 [standard deviation: 17] years; 68% women, 91% White, and 86% non-Hispanic) evaluated for penicillin allergy labels, 513 (11%) had positive penicillin allergy testing. Model input variables were frequently missing: immediate or delayed onset (71%), signs or symptoms (13%), and treatment (31%). The gradient-boosted model was the strongest model with an AUC of 0.67 (95% confidence interval [CI]: 0.57-0.77), which improved to 0.87 (95% CI: 0.73-1) when only cases with complete data were used. Top SHAP drivers for positive testing were reactions within the last year and reactions requiring medical attention; female sex and reaction of hives/urticaria were also positive drivers.

An ML prediction model for positive penicillin allergy skin testing using US-based retrospective data did not achieve performance strong enough for acceptance and adoption. The optimal ML prediction model for positive penicillin allergy testing was driven by time since reaction, seek medical attention, female sex, and hives/urticaria.