The “pneumotaxic center” in the Kölliker-Fuse and medial parabrachial nuclei of dorsolateral pons (dl-pons) plays an important role in respiratory phase switching, modulation of respiratory reflex, and rhythmogenesis. Recent electrophysiological and neural tracing data implicate additional pneumotaxic nuclei in (and a broader role for) the dl-pons in integrating respiratory and nonrespiratory information. Here, we examined the cytoarchitecture of the greater pneumotaxic center and its integrating function by using combined extracellular recording and juxtacellular labeling of unit respiratory rhythmic neurons in dl-pons in urethane-anesthetized, vagotomized, paralyzed, and servo-ventilated adult Sprague Dawley rats. Perievent histogram analysis identified four major types of neuronal discharge patterns: inspiratory, expiratory (with three subdivisions), inspiratory–expiratory, and expiratory–inspiratory phase spanning, sometimes with mild tonic background activity. Most recorded neurons were localized in the Kölliker-Fuse and medial parabrachial nuclei, but some were also found in lateral parabrachial nucleus, intertrigeminal nucleus, principal trigeminal sensory nucleus, and supratrigeminal nucleus. The majority of labeled neurons had large and spatially extended dendritic trees that spanned several of these dl-pons subnuclei, often with terminal dendrites ending in the ventral spinocerebellar tract. The distal sections of the primary and higher-order dendrites exhibited rich varicosities, sometimes with dendritic spines. Axons of some labeled neurons were traced all the way to the ventrolateral pons (vl-pons). These findings extend and generalize the classical definition of the pneumotaxic center to include extensive somatic–axonal–dendritic integration of complex descending and ascending respiratory information as well as nociceptive and possibly musculoskeletal and trigeminal information in multiple dl-pons and vl-pons structures in the rat.

Background: Artificial Intelligence (AI)-driven Decision Support Systems (DSSs) promise improvements in diagnostic accuracy and clinical workflow efficiency, but their adoption is hindered by inadequate confidence calibration, limited transparency, and poor alignment with real-world decision processes, which limit clinician trust and lead to high override rates. Methods: We developed and validated a dynamic scoring framework to enhance trust in AI-generated diagnoses by integrating AI confidence scores, semantic similarity measures, and transparency weighting into the override decision process using 6689 cardiovascular cases from the MIMIC-III dataset. Override thresholds were calibrated and validated across varying transparency and confidence levels, with override rate as the primary acceptance measure. Results: The implementation of this framework reduced the override rate to 33.29%, with high-confidence predictions (90–99%) overridden at a rate of only 1.7%, and low-confidence predictions (70–79%) at a rate of 99.3%. Minimal transparency diagnoses had a 73.9% override rate compared to 49.3% for moderate transparency. Statistical analyses confirmed significant associations between confidence, transparency, and override rates (p < 0.001). Conclusions: These findings suggest that enhanced transparency and confidence calibration can substantially reduce override rates and promote clinician acceptance of AI diagnostics. Future work should focus on clinical validation to optimize patient safety, diagnostic accuracy, and efficiency.

Current clinical care relies heavily on complex, rule-based systems for tasks like diagnosis and treatment. However, these systems can be cumbersome and require constant updates. This study explores the potential of the large language model (LLM), LLaMA 2, to address these limitations. We tested LLaMA 2′s performance in interpreting complex clinical process models, such as Mayo Clinic Care Pathway Models (CPMs), and providing accurate clinical recommendations. LLM was trained on encoded pathways versions using DOT language, embedding them with SentenceTransformer, and then presented with hypothetical patient cases. We compared the token-level accuracy between LLM output and the ground truth by measuring both node and edge accuracy. LLaMA 2 accurately retrieved the diagnosis, suggested further evaluation, and delivered appropriate management steps, all based on the pathways. The average node accuracy across the different pathways was 0.91 (SD ± 0.045), while the average edge accuracy was 0.92 (SD ± 0.122). This study highlights the potential of LLMs for healthcare information retrieval, especially when relevant data are provided. Future research should focus on improving these models’ interpretability and their integration into existing clinical workflows.

Background: Anatomically accurate illustrations are imperative in medical education, serving as crucial tools to facilitate comprehension of complex anatomical structures. While traditional illustration methods involving human artists remain the gold standard, the rapid advancement of Generative Artificial Intelligence (GAI) models presents a new opportunity to automate and accelerate this process. This study evaluated the potential of GAI models to produce craniofacial anatomy illustrations for educational purposes. Methods: Four GAI models, including Midjourney v6.0, DALL-E 3, Gemini Ultra 1.0, and Stable Diffusion 2.0 were used to generate 736 images across multiple views of surface anatomy, bones, muscles, blood vessels, and nerves of the cranium in both oil painting and realistic photograph styles. Four reviewers evaluated the images for anatomical detail, aesthetic quality, usability, and cost-effectiveness. Inter-rater reliability analysis assessed evaluation consistency. Results: Midjourney v6.0 scored highest for aesthetic quality and cost-effectiveness, and DALL-E 3 performed best for anatomical detail and usability. The inter-rater reliability analysis demonstrated a high level of agreement among reviewers (ICC = 0.858, 95% CI). However, all models showed significant flaws in depicting crucial anatomical details such as foramina, suture lines, muscular origins/insertions, and neurovascular structures. These limitations were further characterized by abstract depictions, mixing of layers, shadowing, abnormal muscle arrangements, and labeling errors. Conclusions: These findings highlight GAI’s potential for rapidly creating craniofacial anatomy illustrations but also its current limitations due to inadequate training data and incomplete understanding of complex anatomy. Refining these models through precise training data and expert feedback is vital. Ethical considerations, such as potential biases, copyright challenges, and the risks of propagating inaccurate information, must also be carefully navigated. Further refinement of GAI models and ethical safeguards are essential for safe use.

The integration of artificial intelligence (AI) into clinical medicine presents a persistent paradox: diagnostic models routinely demonstrate benchmark superiority over human experts, yet bedside adoption remains fragile, and clinician trust is low. Conventional forecasting approaches-projecting model performance along optimistic trend lines-are epistemologically insufficient because they cannot account for the nonlinear sociotechnical transitions that separate technical capability from institutional trust. This Viewpoint applies backcasting, a normative futures methodology with a 4-decade evidence base in energy policy and public governance, to the specific challenge of clinician adoption of AI diagnostics, with the aim of identifying the structural interventions required to achieve durable trust by 2040. Consistent with the tradition of single-expert normative foresight analysis, we applied backcasting as a structured reasoning framework using a STEEP (social, technological, economic, environmental, and political) analysis. Sources from PubMed, IEEE Xplore, Google Scholar, and policy repositories (the US Food and Drug Administration, World Health Organization, Organisation for Economic Co-Operation and Development, and European Commission) published between 2010 and 2025 were reviewed; barriers and enablers were coded across STEEP dimensions to identify pivot points representing convergent, time-bound structural changes. Working backward from a defined 2040 vision state-a health care ecosystem with risk-stratified clinician trust thresholds, semantic transparency of AI outputs, integrated AI governance, and futures literacy in medical education-we identified three temporal pivot points: (1) the 2030 standardization of dual-process AI architectures, in which large language models are verified in real time by locally deployed small language models, producing a calibrated confidence score; (2) the 2035 institutionalization of agentic AI orchestration governed by a formally designated chief AI officer; and (3) the 2040 integration of futures literacy and human-AI teaming competencies into standard medical curricula. The AI trust gap is an institutional design problem, not a technical inevitability. Backcasting reframes the central question from "when will AI be ready for medicine?" to "what must we build to make medicine ready for AI?" The 3 pivot points identified here-verifiable AI by 2030, agentic governance by 2035, and futures literacy by 2040-are structural commitments that clinicians, health system leaders, and policymakers can begin building today.

Large language models (LLMs) are increasingly deployed within multi-step agentic workflows in healthcare, where sequential decisions shape downstream clinical actions. However, existing evaluations primarily assess single-turn accuracy and fail to capture how stochastic variability propagates across multi-step trajectories. In this work, we introduce Variability-Aware Agentic AI Trust (VA-AAT), a framework for modeling and measuring trajectory-level uncertainty in agentic clinical systems. We formalize variability propagation through trajectory divergence and derive conservative bounds that characterize how per-step stochasticity may accumulate under simplifying assumptions. Building on this formulation, we define repeatability-based metrics that quantify consistency at both step and trajectory levels, and we propose a risk-informed trust framework linking repeatability to graded oversight strategies. We empirically evaluate VA-AAT on simulated and real-world clinical workflows, demonstrating that small per-step variations can lead to substantial divergence in multi-step trajectories. Variability was strongly influenced by model selection and decoding temperature, and we identified instruction-following failures (“unknown actions”) as an additional source of instability. Importantly, we show that repeatability provides a useful but incomplete indicator of system reliability: highly repeatable systems may still produce consistently incorrect outputs, while less repeatable systems may reflect underlying clinical ambiguity. Accordingly, repeatability should be interpreted as a diagnostic signal of variability-induced risk rather than a guarantee of safety. These findings highlight the need for trajectory-level evaluation and variability-aware governance in clinical AI. The proposed framework provides a foundation for assessing stochastic reliability in agentic systems and supports the design of risk-sensitive deployment strategies in healthcare settings.