Comparative Study to Evaluate the Accuracy of Differential Diagnosis Lists Generated by Gemini Advanced, Gemini, and Bard for a Case Report Series Analysis: Cross-Sectional Study.

BACKGROUND: Generative artificial intelligence (GAI) systems by Google have recently been updated from Bard to Gemini and Gemini Advanced as of December 2023. Gemini is a basic, free-to-use model after a user's login, while Gemini Advanced operates on a more advanced model requiring a fee-based subscription. These systems have the potential to enhance medical diagnostics. However, the impact of these updates on comprehensive diagnostic accuracy remains unknown. OBJECTIVE: This study aimed to compare the accuracy of the differential diagnosis lists generated by Gemini Advanced, Gemini, and Bard across comprehensive medical fields using case report series. METHODS: We identified a case report series with relevant final diagnoses published in the American Journal Case Reports from January 2022 to March 2023. After excluding nondiagnostic cases and patients aged 10 years and younger, we included the remaining case reports. After refining the case parts as case descriptions, we input the same case descriptions into Gemini Advanced, Gemini, and Bard to generate the top 10 differential diagnosis lists. In total, 2 expert physicians independently evaluated whether the final diagnosis was included in the lists and its ranking. Any discrepancies were resolved by another expert physician. Bonferroni correction was applied to adjust the P values for the number of comparisons among 3 GAI systems, setting the corrected significance level at P value <.02. RESULTS: In total, 392 case reports were included. The inclusion rates of the final diagnosis within the top 10 differential diagnosis lists were 73% (286/392) for Gemini Advanced, 76.5% (300/392) for Gemini, and 68.6% (269/392) for Bard. The top diagnoses matched the final diagnoses in 31.6% (124/392) for Gemini Advanced, 42.6% (167/392) for Gemini, and 31.4% (123/392) for Bard. Gemini demonstrated higher diagnostic accuracy than Bard both within the top 10 differential diagnosis lists (P=.02) and as the top diagnosis (P=.001). In addition, Gemini Advanced achieved significantly lower accuracy than Gemini in identifying the most probable diagnosis (P=.002). CONCLUSIONS: The results of this study suggest that Gemini outperformed Bard in diagnostic accuracy following the model update. However, Gemini Advanced requires further refinement to optimize its performance for future artificial intelligence-enhanced diagnostics. These findings should be interpreted cautiously and considered primarily for research purposes, as these GAI systems have not been adjusted for medical diagnostics nor approved for clinical use.

The ethical requirement of explainability for AI-DSS in healthcare: a systematic review of reasons.

BACKGROUND: Despite continuous performance improvements, especially in clinical contexts, a major challenge of Artificial Intelligence based Decision Support Systems (AI-DSS) remains their degree of epistemic opacity. The conditions of and the solutions for the justified use of the occasionally unexplainable technology in healthcare are an active field of research. In March 2024, the European Union agreed upon the Artificial Intelligence Act (AIA), requiring medical AI-DSS to be ad-hoc explainable or to use post-hoc explainability methods. The ethical debate does not seem to settle on this requirement yet. This systematic review aims to outline and categorize the positions and arguments in the ethical debate. METHODS: We conducted a literature search on PubMed, BASE, and Scopus for English-speaking scientific peer-reviewed publications from 2016 to 2024. The inclusion criterion was to give explicit requirements of explainability for AI-DSS in healthcare and reason for it. Non-domain-specific documents, as well as surveys, reviews, and meta-analyses were excluded. The ethical requirements for explainability outlined in the documents were qualitatively analyzed with respect to arguments for the requirement of explainability and the required level of explainability. RESULTS: The literature search resulted in 1662 documents; 44 documents were included in the review after eligibility screening of the remaining full texts. Our analysis showed that 17 records argue in favor of the requirement of explainable AI methods (xAI) or ad-hoc explainable models, providing 9 categories of arguments. The other 27 records argued against a general requirement, providing 11 categories of arguments. Also, we found that 14 works advocate the need for context-dependent levels of explainability, as opposed to 30 documents, arguing for context-independent, absolute standards. CONCLUSIONS: The systematic review of reasons shows no clear agreement on the requirement of post-hoc explainability methods or ad-hoc explainable models for AI-DSS in healthcare. The arguments found in the debate were referenced and responded to from different perspectives, demonstrating an interactive discourse. Policymakers and researchers should watch the development of the debate closely. Conversely, ethicists should be well informed by empirical and technical research, given the frequency of advancements in the field.

eConsent administered by Community Health Workers in a study using the Trials within Cohorts (TwiCs) design-Experiences from the Community-Based chronic Care Lesotho (ComBaCaL) project.

Improving access to essential health services requires the development of innovative health service delivery models and their scientific assessment in often large-scale pragmatic trials. In many low- and middle-income countries, lay Community Health Workers (CHWs) play an important role in delivering essential health services. As trusted members of their communities with basic medical training, they may also contribute to health data collection. Digital clinical decision support applications may facilitate the involvement of CHWs in service delivery and data collection. Electronic consent (eConsent) can streamline the consent process that is required if the collected data is used for the scientific purposes. Here, we describe the experiences of using eConsent in the Community-Based chronic Care Lesotho (ComBaCaL) cohort study and multiple nested pragmatic cluster-randomized trials assessing CHW-led care delivery models for type 2 diabetes and arterial hypertension using the Trials within Cohorts (TwiCs) design. More than a hundred CHWs, acting both as service providers and data collectors in remote villages of Lesotho utilize an eConsent application that is linked to a tailored clinical decision support and data collection application. The eConsent application presents simplified consent information and generates personalized consent forms that are signed electronically on a tablet and then uploaded to the database of the clinical decision support application. This significantly streamlines the consent process and allows for quality consent documentation through timely central monitoring, facilitating the CHW-led management of a large-scale population-based cohort in a remote low-resource area with continuous enrollment-currently at more than 16,000 participants.

Computed tomography referral guidelines adherence in Europe: insights from a seven-country audit.

BACKGROUND: Ensuring appropriate computed tomography (CT) utilization optimizes patient care while minimizing radiation exposure. Decision support tools show promise for standardizing appropriateness. OBJECTIVES: In the current study, we aimed to assess CT appropriateness rates using the European Society of Radiology (ESR) iGuide criteria across seven European countries. Additional objectives were to identify factors associated with appropriateness variability and examine recommended alternative exams. METHODS: As part of the European Commission-funded EU-JUST-CT project, 6734 anonymized CT referrals were audited across 125 centers in Belgium, Denmark, Estonia, Finland, Greece, Hungary, and Slovenia. In each country, two blinded radiologists independently scored each exam's appropriateness using the ESR iGuide and noted any recommended alternatives based on presented indications. Arbitration was used in case auditors disagreed. Associations between appropriateness rate and institution type, patient's age and sex, inpatient/outpatient patient status, anatomical area, and referring physician's specialty were statistically examined within each country. RESULTS: The average appropriateness rate was 75%, ranging from 58% in Greece to 86% in Denmark. Higher rates were associated with public hospitals, inpatient settings, and referrals from specialists. Variability in appropriateness existed by country and anatomical area, patient age, and gender. Common alternative exam recommendations included magnetic resonance imaging, X-ray, and ultrasound. CONCLUSION: This multi-country evaluation found that even when using a standardized imaging guideline, significant variations in CT appropriateness persist, ranging from 58% to 86% across the participating countries. The study provided valuable insights into real-world utilization patterns and identified opportunities to optimize practices and reduce clinical and demographic disparities in CT use. KEY POINTS: Question Largest multinational study (7 EU countries, 6734 CT referrals) assessed real-world CT appropriateness using ESR iGuide, enabling cross-system comparisons. Findings Significant variability in appropriateness rates across institution type, patient status, age, gender, exam area, and physician specialty, highlighted the opportunities to optimize practices based on local factors. Clinical relevance International collaboration on imaging guidelines and decision support can maximize CT benefits while optimizing radiation exposure; ongoing research is crucial for refining evidence-based guidelines globally.

Clinicians' roles and necessary levels of understanding in the use of artificial intelligence: A qualitative interview study with German medical students.

BACKGROUND: Artificial intelligence-driven Clinical Decision Support Systems (AI-CDSS) are being increasingly introduced into various domains of health care for diagnostic, prognostic, therapeutic and other purposes. A significant part of the discourse on ethically appropriate conditions relate to the levels of understanding and explicability needed for ensuring responsible clinical decision-making when using AI-CDSS. Empirical evidence on stakeholders' viewpoints on these issues is scarce so far. The present study complements the empirical-ethical body of research by, on the one hand, investigating the requirements for understanding and explicability in depth with regard to the rationale behind them. On the other hand, it surveys medical students at the end of their studies as stakeholders, of whom little data is available so far, but for whom AI-CDSS will be an important part of their medical practice. METHODS: Fifteen semi-structured qualitative interviews (each lasting an average of 56 min) were conducted with German medical students to investigate their perspectives and attitudes on the use of AI-CDSS. The problem-centred interviews draw on two hypothetical case vignettes of AI-CDSS employed in nephrology and surgery. Interviewees' perceptions and convictions of their own clinical role and responsibilities in dealing with AI-CDSS were elicited as well as viewpoints on explicability as well as the necessary level of understanding and competencies needed on the clinicians' side. The qualitative data were analysed according to key principles of qualitative content analysis (Kuckartz). RESULTS: In response to the central question about the necessary understanding of AI-CDSS tools and the emergence of their outputs as well as the reasons for the requirements placed on them, two types of argumentation could be differentiated inductively from the interviewees' statements: the first type, the clinician as a systemic trustee (or "the one relying"), highlights that there needs to be empirical evidence and adequate approval processes that guarantee minimised harm and a clinical benefit from the employment of an AI-CDSS. Based on proof of these requirements, the use of an AI-CDSS would be appropriate, as according to "the one relying", clinicians should choose those measures that statistically cause the least harm. The second type, the clinician as an individual expert (or "the one controlling"), sets higher prerequisites that go beyond ensuring empirical evidence and adequate approval processes. These higher prerequisites relate to the clinician's necessary level of competence and understanding of how a specific AI-CDSS works and how to use it properly in order to evaluate its outputs and to mitigate potential risks for the individual patient. Both types are unified in their high esteem of evidence-based clinical practice and the need to communicate with the patient on the use of medical AI. However, the interviewees' different conceptions of the clinician's role and responsibilities cause them to have different requirements regarding the clinician's understanding and explicability of an AI-CDSS beyond the proof of benefit. CONCLUSIONS: The study results highlight two different types among (future) clinicians regarding their view of the necessary levels of understanding and competence. These findings should inform the debate on appropriate training programmes and professional standards (e.g. clinical practice guidelines) that enable the safe and effective clinical employment of AI-CDSS in various clinical fields. While current approaches search for appropriate minimum requirements of the necessary understanding and competence, the differences between (future) clinicians in terms of their information and understanding needs described here can lead to more differentiated approaches to solutions.

Exploring the potential cost-effectiveness of a new computerised decision support tool for identifying fetal compromise during monitored term labours: an early health economic model.

BACKGROUND: Around 60% of term labours in the UK are continuously monitored using cardiotocography (CTG) to guide clinical labour management. Interpreting the CTG trace is challenging, leading to some babies suffering adverse outcomes and others unnecessary expedited deliveries. A new data driven computerised tool combining multiple clinical risk factors with CTG data (attentive CTG) was developed to help identify term babies at risk of severe compromise during labour. This paper presents an early health economic model exploring its potential cost-effectiveness. METHODS: The model compared attentive CTG and usual care with usual care alone and simulated clinical events, healthcare costs, and infant quality-adjusted life years over 18 years. It was populated using data from a cohort of term pregnancies, the literature, and administrative datasets. Attentive CTG effectiveness was projected through improved monitoring sensitivity/specificity and potential reductions in numbers of severely compromised infants. Scenario analyses explored the impact of including litigation costs. RESULTS: Nationally, attentive CTG could potentially avoid 10,000 unnecessary alerts in labour and 2400 emergency C-section deliveries through improved specificity. A reduction of 21 intrapartum stillbirths amongst severely compromised infants was also predicted with improved sensitivity. Attentive CTG could potentially lead to cost savings and health gains with a probability of being cost-effective at £25,000 per QALY ranging from 70 to 95%. Potential exists for further cost savings if litigation costs are included. CONCLUSIONS: Attentive CTG could offer a cost-effective use of healthcare resources. Prospective patient-level studies are needed to formally evaluate its effectiveness and economic impact in routine clinical practice.

An innovative tool to prioritize the assessment of investigational COVID-19 therapeutics: A pilot project.

Background: As the COVID-19 pandemic unfolded, hundreds of investigational COVID-19 therapeutics emerged. Maintaining situational awareness of this extensive and rapidly evolving therapeutic landscape represented an unprecedented challenge for the Public Health Agency of Canada, as it worked to promote and protect the health of Canadians. A tool to triage and prioritize the assessment of these therapeutics was needed. Methods: The objective was to develop and conduct an initial validation of a tool to identify investigational COVID-19 therapeutics for further review based on an efficient preliminary assessment, using a systematic and reliable process that would be practical to validate, implement and update. Phase 1 of this pilot project consisted of a literature search to identify existing COVID-19 therapeutic assessment prioritization tools, development of the Rapid Scoring Tool (RST) and initial validation of the tool. Results: No tools designed to rank investigational COVID-19 therapeutics for the purpose of prioritizing their assessment were identified. However, a few publications provided criteria to consider and therapeutic ranking methods, which helped shape the development of the RST. The RST included eight criteria and several descriptors ("characteristics"). A universal characteristic scoring scale from -10 to 10 was developed. The sum of all the characteristic scores yielded an overall benefit score for each therapeutic. The RST appropriately ranked therapeutics using a systematic, reliable and practical approach. Conclusion: Phase 1 was successfully completed. The RST presents several distinct aspects compared with other tools, including its scoring scale and method, and capacity to factor in incomplete or pending information. It is anticipated that the framework used for the RST will lend itself to use in other dynamic situations involving many interventions.

Development and validation of a sepsis risk index supporting early identification of ICU-acquired sepsis: An observational study.

BACKGROUND: Sepsis is a threat to global health, and domestically is the major cause of in-hospital mortality. Due to increases in inpatient morbidity and mortality resulting from sepsis, healthcare providers (HCPs) would accrue significant benefits from identifying the syndrome early and treating it promptly and effectively. Prompt and effective detection, diagnosis, and treatment of sepsis requires frequent monitoring and assessment of patient vital signs and other relevant data present in the electronic health record. METHODS: This study explored the development of machine learning-based models to generate a novel sepsis risk index (SRI) which is an intuitive 0-100 marker that reflects the risk of a patient acquiring sepsis or septic shock and assists in timely diagnosis. Machine learning models were developed and validated using openly accessible critical care databases. The model was developed using a single database (from one institution) and validated on a separate database consisting of patient data collected across multiple ICUs. RESULTS: The developed model achieved an area under the receiver operating characteristic curve of 0.82 and 0.84 for the diagnosis of sepsis and septic shock, respectively, with a sensitivity and specificity of 79.1% [75.1, 82.7] and 73.3% [72.8, 73.8] for a sepsis diagnosis and 83.8% [80.8, 86.5] and 73.3% [72.8, 73.8] for a septic shock diagnosis. CONCLUSION: The SRI provides critical care HCPs with an intuitive quantitative measure related to the risk of a patient having or acquiring a life-threatening infection. Evaluation of the SRI over time may provide HCPs the ability to initiate protective interventions (e.g. targeted antibiotic therapy).

Decision Support Techniques for determination of the causal relationship between health problems of workers and their work activities.

Reflecting on the complexity and impacts of determination of the causal relationship between health problems of workers and the exercise of their work activities, there is a need to learn about scientific articles that expose techniques to determine this type of causal relationship. There is also a need to reveal whether any article exposes multicriteria decision analysis technique. The aim is to quantify the techniques used to determine the causal relationship between health problems of workers and the exercise of their work activities. Bibliometric analysis was performed, searching for articles in Portuguese, Spanish and English. An advanced search was performed on the website of the ministerial journals portal and then on the Gale Academic OneFile, SciVerse Scopus, Scientific Electronic Library Online (SciELO) and PubMed Central collections. In summary, 38 articles were selected from portal, 50 from Gale Academic OneFile, 20 from SciVerse Scopus, 37 from SciELO and 5 from PubMed Central, totaling 150 articles of interest for analysis of their contents. Among these 150 articles, 33.33% addressed the causal relationship between illness and work, 3.33% described some process related to occupational diagnostic investigation and 0.66%, which represents only one article, exhibited a technique to determine this type of causal relationship: the probability of causality in neoplastic diseases. No article described multicriteria decision analysis method as a technique for determine this type of causal relationship. Therefore, there is a need to carry out and disseminate scientific research on methods to help determine a causal relationship between illness and work.

Ao refletir sobre a complexidade e os impactos do estabelecimento do nexo causal entre o agravo à saúde dos trabalhadores e o exercício de suas atividades laborais, surge a necessidade de conhecer artigos científicos que expõem técnicas para estabelecer esse tipo de nexo. Surge também a necessidade de revelar se algum artigo expõe auxílio multicritério à decisão. O objetivo foi quantificar as técnicas utilizadas no estabelecimento do nexo causal entre o agravo à saúde dos trabalhadores e o exercício de suas atividades laborais. Foi realizada uma análise bibliométrica, buscando artigos em português, espanhol e inglês. Realizou-se uma busca avançada no site do portal ministerial de periódicos e, em seguida, nas coleções Gale Academic OneFile, SciVerse Scopus, Scientific Electronic Library Online (SciELO) e PubMed Central. Em síntese, foram selecionados 38 artigos do portal ministerial de periódicos, 50 da Gale Academic OneFile, 20 da SciVerse Scopus, 37 do SciELO e 5 da PubMed Central, totalizando 150 artigos para análise de conteúdo. Entre esses 150 artigos, 33,33% abordavam o nexo causal entre doença e trabalho, 3,33% descreviam algum processo relacionado à investigação diagnóstica ocupacional, e 0,66%, o que representa apenas um artigo, exibia uma técnica para se estabelecer esse tipo de nexo causal: a probabilidade de causalidade em doenças neoplásicas. Nenhum artigo descrevia o auxílio multicritério à decisão como técnica para estabelecer esse tipo de nexo causal. Portanto, nota-se a necessidade da realização e divulgação de pesquisas científicas sobre métodos de auxílio ao estabelecimento de nexo causal entre doença e trabalho.

Artificial intelligence based real-time prediction of imminent heart failure hospitalisation in patients undergoing non-invasive telemedicine.

Background: Remote patient management may improve prognosis in heart failure. Daily review of transmitted data for early recognition of patients at risk requires substantial resources that represent a major barrier to wide implementation. An automated analysis of incoming data for detection of risk for imminent events would allow focusing on patients requiring prompt medical intervention. Methods: We analysed data of the Telemedical Interventional Management in Heart Failure II (TIM-HF2) randomized trial that were collected during quarterly in-patient visits and daily transmissions from non-invasive monitoring devices. By application of machine learning, we developed and internally validated a risk score for heart failure hospitalisation within seven days following data transmission as estimate of short-term patient risk for adverse heart failure events. Score performance was assessed by the area under the receiver-operating characteristic (ROCAUC) and compared with a conventional algorithm, a heuristic rule set originally applied in the randomized trial. Results: The machine learning model significantly outperformed the conventional algorithm (ROCAUC 0.855 vs. 0.727, p < 0.001). On average, the machine learning risk score increased continuously in the three weeks preceding heart failure hospitalisations, indicating potential for early detection of risk. In a simulated one-year scenario, daily review of only the one third of patients with the highest machine learning risk score would have led to detection of 95% of HF hospitalisations occurring within the following seven days. Conclusions: A machine learning model allowed automated analysis of incoming remote monitoring data and reliable identification of patients at risk of heart failure hospitalisation requiring immediate medical intervention. This approach may significantly reduce the need for manual data review.

An Artificial Intelligence Algorithm for Detection of Severe Aortic Stenosis: A Clinical Cohort Study.

Background: Identifying individuals with severe aortic stenosis (AS) at high risk of mortality remains challenging using current clinical imaging methods. Objectives: The purpose of this study was to evaluate an artificial intelligence decision support algorithm (AI-DSA) to augment the detection of severe AS within a well-resourced health care setting. Methods: Agnostic to clinical information, an AI-DSA trained to identify echocardiographic phenotype associated with an aortic valve area (AVA)<1 cm2 using minimal input data (excluding left ventricular outflow tract measures) was applied to routine transthoracic echocardiograms (TTE) reports from 31,141 U.S. Medicare beneficiaries at an academic medical center (2003-2017). Results: Performance of AI-DSA to detect the phenotype associated with an AVA<1 cm2 was excellent (sensitivity 82.2%, specificity 98.1%, negative predictive value 9.2%, c-statistic = 0.986). In addition to identifying clinical severe AS cases, AI-DSA identified an additional 1,034 (3.3%) individuals with guideline-defined moderate AS but with a similar clinical and TTE phenotype to those with severe AS with low rates of aortic valve replacement (6.6%). Five-year mortality was 75.9% in those with known severe AS, 73.5% in those with a similar phenotype to severe AS, and 44.6% in those without severe AS. The AI-DSA continued to perform well to identify severe AS among those with a depressed left ventricular ejection fraction. Overall rates of aortic valve replacement remained low, even in those with an AVA<1 cm2 (21.9%). Conclusions: Without relying on left ventricular outflow tract measurements, an AI-DSA used echocardiographic reports to reliably identify the phenotype of severe AS. These results suggest possible utility for this AI-DSA to enhance detection of severe AS individuals at risk for adverse outcomes.

Integrating tabular data through image conversion for enhanced diagnosis: A novel intelligent decision support system for stratifying obstructive sleep apnoea patients using convolutional neural networks.

Objective: High-dimensional databases make it difficult to apply traditional learning algorithms to biomedical applications. Recent developments in computer technology have introduced deep learning (DL) as a potential solution to these difficulties. This study presents a novel intelligent decision support system based on a novel interpretation of data formalisation from tabular data in DL techniques. Once defined, it is used to diagnose the severity of obstructive sleep apnoea, distinguishing between moderate to severe and mild/no cases. Methods: The study uses a complete database extract from electronic health records of 2472 patients, including anthropometric data, habits, medications, comorbidities, and patient-reported symptoms. The novelty of this methodology lies in the initial processing of the patients' data, which is formalised into images. These images are then used as input to train a convolutional neural network (CNN), which acts as the inference engine of the system. Results: The initial tests of the system were performed on a set of 247 samples from the Pulmonary Department of the Álvaro Cunqueiro Hospital in Vigo (Galicia, Spain), with an AUC value of ≈ 0.8. Conclusions: This study demonstrates the benefits of an intelligent decision support system based on a novel data formalisation approach that allows the use of advanced DL techniques starting from tabular data. In this way, the ability of CNNs to recognise complex patterns using visual elements such as gradients and contrasts can be exploited. This approach effectively addresses the challenges of analysing large amounts of tabular data and reduces common problems such as bias and variance, resulting in improved diagnostic accuracy.

Development and Assessment of a Point-of-Care Application (Genomic Medicine Guidance) for Heritable Thoracic Aortic Disease.

Background: Genetic testing can determine familial and personal risks for heritable thoracic aortic aneurysms and dissections (TAD). The 2022 American College of Cardiology/American Heart Association guidelines for TAD recommend management decisions based on the specific gene mutation. However, many clinicians lack sufficient comfort or insight to integrate genetic information into clinical practice. Objective: We therefore developed the Genomic Medicine Guidance (GMG) application, an interactive point-of-care tool to inform clinicians and patients about TAD diagnosis, treatment, and surveillance. GMG is a REDCap-based application that combines publicly available genetic data and clinical recommendations based on the TAD guidelines into one translational education tool. Methods: TAD genetic information in GMG was sourced from the Montalcino Aortic Consortium, a worldwide collaboration of TAD centers of excellence, and the National Institutes of Health genetic repositories ClinVar and ClinGen. Results: The application streamlines data on the 13 most frequently mutated TAD genes with 2286 unique pathogenic mutations that cause TAD so that users receive comprehensive recommendations for diagnostic testing, imaging, surveillance, medical therapy, and preventative surgical repair, as well as guidance for exercise safety and management during pregnancy. The application output can be displayed in a clinician view or exported as an informative pamphlet in a patient-friendly format. Conclusions: The overall goal of the GMG application is to make genomic medicine more accessible to clinicians and patients while serving as a unifying platform for research. We anticipate that these features will be catalysts for collaborative projects aiming to understand the spectrum of genetic variants contributing to TAD.

Investigating Data Diversity and Model Robustness of AI Applications in Palliative Care and Hospice: Protocol for Scoping Review.

BACKGROUND: Artificial intelligence (AI) has become a pivotal element in health care, leading to significant advancements across various medical domains, including palliative care and hospice services. These services focus on improving the quality of life for patients with life-limiting illnesses, and AI's ability to process complex datasets can enhance decision-making and personalize care in these sensitive settings. However, incorporating AI into palliative and hospice care requires careful examination to ensure it reflects the multifaceted nature of these settings. OBJECTIVE: This scoping review aims to systematically map the landscape of AI in palliative care and hospice settings, focusing on the data diversity and model robustness. The goal is to understand AI's role, its clinical integration, and the transparency of its development, ultimately providing a foundation for developing AI applications that adhere to established ethical guidelines and principles. METHODS: Our scoping review involves six stages: (1) identifying the research question; (2) identifying relevant studies; (3) study selection; (4) charting the data; (5) collating, summarizing, and reporting the results; and (6) consulting with stakeholders. Searches were conducted across databases including MEDLINE through PubMed, Embase.com, IEEE Xplore, ClinicalTrials.gov, and Web of Science Core Collection, covering studies from the inception of each database up to November 1, 2023. We used a comprehensive set of search terms to capture relevant studies, and non-English records were excluded if their abstracts were not in English. Data extraction will follow a systematic approach, and stakeholder consultations will refine the findings. RESULTS: The electronic database searches conducted in November 2023 resulted in 4614 studies. After removing duplicates, 330 studies were selected for full-text review to determine their eligibility based on predefined criteria. The extracted data will be organized into a table to aid in crafting a narrative summary. The review is expected to be completed by May 2025. CONCLUSIONS: This scoping review will advance the understanding of AI in palliative care and hospice, focusing on data diversity and model robustness. It will identify gaps and guide future research, contributing to the development of ethically responsible and effective AI applications in these settings. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/56353.

Transparency improves the accuracy of automation use, but automation confidence information does not.

Increased automation transparency can improve the accuracy of automation use but can lead to increased bias towards agreeing with advice. Information about the automation's confidence in its advice may also increase the predictability of automation errors. We examined the effects of providing automation transparency, automation confidence information, and their potential interacting effect on the accuracy of automation use and other outcomes. An uninhabited vehicle (UV) management task was completed where participants selected the optimal UV to complete missions. Low or high automation transparency was provided, and participants agreed/disagreed with automated advice on each mission. We manipulated between participants whether automated advice was accompanied by confidence information. This information indicated on each trial whether automation was "somewhat" or "highly" confident in its advice. Higher transparency improved the accuracy of automation use, led to faster decisions, lower perceived workload, and increased trust and perceived usability. Providing participant automation confidence information, as compared with not, did not have an overall impact on any outcome variable and did not interact with transparency. Despite no benefit, participants who were provided confidence information did use it. For trials where lower compared to higher confidence information was presented, hit rates decreased, correct rejection rates increased, decision times slowed, and perceived workload increased, all suggestive of decreased reliance on automated advice. Such trial-by-trial shifts in automation use bias and other outcomes were not moderated by transparency. These findings can potentially inform the design of automated decision-support systems that are more understandable by humans in order to optimise human-automation interaction.

Aliado - A design concept of AI for decision support in oncological liver surgery.

BACKGROUND: The interest in artificial intelligence (AI) is increasing. Systematic reviews suggest that there are many machine learning algorithms in surgery, however, only a minority of the studies integrate AI applications in clinical workflows. Our objective was to design and evaluate a concept to use different kinds of AI for decision support in oncological liver surgery along the treatment path. METHODS: In an exploratory co-creation between design experts, surgeons, and data scientists, pain points along the treatment path were identified. Potential designs for AI-assisted solutions were developed and iteratively refined. Finally, an evaluation of the design concept was performed with n = 20 surgeons to get feedback on the different functionalities and evaluate the usability with the System Usability Scale (SUS). Participating surgeons had a mean of 14.0 ± 5.0 years of experience after graduation. RESULTS: The design concept was named "Aliado". Five different scenarios were identified where AI could support surgeons. Mean score of SUS was 68.2 ( ± 13.6 SD). The highest valued functionalities were "individualized prediction of survival, short-term mortality and morbidity", and "individualized recommendation of surgical strategy". CONCLUSION: Aliado is a design prototype that shows how AI could be integrated into the clinical workflow. Even without a fleshed out user interface, the SUS already yielded borderline good results. Expert surgeons rated the functionalities favorably, and most of them expressed their willingness to work with a similar application in the future. Thus, Aliado can serve as a surgical vision of how an ideal AI-based assistance could look like.

Construction and Validation of an Assistant Decision-Making Model for Platelet Transfusion Refractoriness in Patients with Acute Myeloid Leukemia.

Background: Platelet transfusion refractoriness (PTR) is a complication of multiple transfusions in patients with hematological malignancies. PTR may induce a series of adverse events, such as delaying the treatment of the primary disease and life-threatening bleeding. Early prediction of PTR holds promise in facilitating prompt adjustments to treatment strategies by clinicians. Methods: We collected the clinical data of 250 patients with acute myeloid leukemia (AML). Subsequently, the patients were randomly divided into a training cohort and a validation cohort at a ratio of 7:3. The least absolute shrinkage and selection operator (LASSO) and multivariate logistic-regression methods were used to select characteristic variables. Assessment of the model was conducted through the receiver operating characteristic (ROC), calibration curve and decision curve analysis (DCA). Results: Out of 250 patients with AML, 95 individuals (38.0%) experienced PTR. Among those with positive platelet associated antibodies (PAAs), the incidence of PTR was 66.7% (30/45), while among patients positive for human leukocyte antigen(HLA)-I antibodies, the PTR incidence was 56.5% (48/85). The final predictive model incorporated risk factors such as KIT mutations, splenomegaly, the number of HLA-I antibodies, and positive PAAs. A prediction nomogram model was constructed based on these four risk factors. The LASSO-logistic regression model demonstrated excellent discrimination, calibration, and clinical decision value. Conclusion: The LASSO-logistic regression model in the study can better predict the risk of PTR. The study includes both PAAs and HLA antibodies, expanding the field of work that has not been involved in the previous prediction model of PTR.

Signs and symptoms in hospitalized children and adolescents' triggering nurses' awareness - A qualitative study across Denmark and Norway.

BACKGROUND: Assessment of signs and symptoms in hospitalized children presents unique challenges due to the children's age-related differences, such as vital signs and the broad range of medical conditions that affects children. Early detection of clinical changes in children is crucial to prevent deterioration, and while standardized tools exist, there is a growing recognition of the need to consider subjective factors based on experienced nurses' knowledge and intuition. OBJECTIVE: To explore which signs and symptoms, apart from vital signs, that trigger nurses' concern regarding deterioration of hospitalized children and adolescents. DESIGN: This study used a descriptive qualitative design. SETTINGS: The study was conducted at three pediatric departments in Denmark and a nursing department of a university in Norway, offering post graduate education programs for health care professions working with children and adolescents throughout Norway. PARTICIPANTS: A total sample of 29 registered nurses with varying levels of experience participated. METHOD: Four focus group interviews were used to collect data and analyzed with inductive content analysis approach. RESULTS: Nurses' knowledge about children's clinical conditions is influenced by the nurses experience, their use of senses like touching the child with their hands, and the use of various approaches. Information from parents about the child's normal behavior are considered valuable. These sources of information, often difficult to verbalize, might be referred to as intuition or "gut feeling" and often guides the nurses' actions when vital signs appear normal, and nurses rely on their senses to assess the child's condition. Specific indicators triggering concern include changes in respiration, circulation, level of consciousness, and facial expressions. Challenges arise from nighttime assessments, interactions with parents, the presence of electronic devices, and children's ability to compensate. Clinical experience is a significant factor in nurses' ability to recognize changes in in the child's condition. CONCLUSION: This study highlights the multifaceted nature of nurses' assessments of clinical conditions in hospitalized children. Nurses draw on their experiences, intuition, and interactions with parents to complement vital signs-based assessments. Their intuition, or "gut feeling" serves as a valuable tool when vital signs do not fully capture the child's clinical status. Specific signs and symptoms that trigger nurses' concern, along with the challenges they face, contribute to a comprehensive understanding of the complexity of assessing children's clinical conditions. These findings, emphasize the role of nurses in early recognition of clinical deterioration in hospitalized children and the need for assessments that go beyond vital signs. TWEETABLE ABSTRACT: Both objective assessments and intuitive clinical judgment play an important role in identifying potential deterioration in pediatric patients.

Delays in bipolar depression treatment in primary care vs. integrated behavioral health and specialty care.

INTRODUCTION: While bipolar disorder is not uncommon in primary care, collaborative care models for bipolar depression treatment are underdeveloped. Our aim was to compare initial pharmacological treatment patterns for an episode of bipolar depression in different care models, namely primary care (PC), integrated behavioral health (IBH), and mood specialty clinic (SC). METHODS: A retrospective study of adults diagnosed with bipolar disorder who received outpatient care in 2020 was completed. Depressive episodes were captured based on DSM-5 criteria, ICD codes, or de novo emergent symptom burden (PHQ-9 ≥ 10). Pharmacological strategies were classified as 1) continuation of current regimen, 2) dose increase or 3) augmentation 4) switch to monotherapy or 5) a combination of more than two different strategies. Logistic regression was applied. RESULTS: A total of 217 encounters (PC = 32, IBH = 53, SC = 132) representing 186 unique patients were identified. PC was significantly more likely to continue the current regimen, while combination strategies were significantly more likely recommended in IBH and SC. Mood stabilizers were significantly more utilized in IBH and SC. There were no significant group differences in antidepressant use. LIMITATIONS: Retrospective study design at a single site. CONCLUSIONS: This study provides evidence of delays in depression care in bipolar disorder. This is the first study to compare treatment recommendations for bipolar depression in different clinical settings. Future studies are encouraged to better understand this gap and to guide future clinical practice, regardless of care model, emphasizing the potential benefits of decision support tools and collaborative care models tailored for bipolar depression.