Search

In the evolving landscape of healthcare informatics, the integration of blockchain technology with artificial intelligence (AI) offers transformative potential for secure and intelligent health data exchange. This conceptual manuscript proposes a novel scaffold for blockchain-enhanced health data intelligence (S-BEHDI), designed as a multi-layered architectural framework that facilitates seamless, secure, and intelligent interoperability among disparate health data systems. By leveraging blockchain’s immutable ledger for data provenance and AI-driven analytics for decision support, S-BEHDI addresses critical challenges in electronic health records (EHR) exchange, such as privacy breaches, data silos, and inefficient clinical workflows. The framework incorporates a unique feedback topology that dynamically adjusts intelligence layers based on governance constraints and data exchange dynamics, ensuring robust monitoring and ethical AI deployment in clinical settings. Theoretical formulas are introduced to interpret risk propagation in data exchanges, decision confidence in AI-assisted pipelines, and governance load in interoperability frameworks. Drawing from recent peer-reviewed literature, this work synthesizes advancements in clinical AI architectures, healthcare analytics infrastructures, and interoperability models to underscore the scaffold’s theoretical underpinnings. While devoid of empirical evaluations, the conceptual design highlights implications for enhanced patient-centric care, reduced monitoring burdens, and fortified data security in precision medicine applications. Ultimately, S-BEHDI represents a forward-thinking infrastructure for fostering collaborative, intelligent health data ecosystems without compromising ethical standards or system integrity.

The integration of artificial intelligence (AI) into healthcare systems has revolutionized the orchestration of personalized treatments. Yet, challenges persist in establishing causal linkages between patient data, algorithmic decisions, and clinical outcomes. This conceptual manuscript proposes the causal orchestration network for treatment intelligence (CONTI), a novel pathway model designed to facilitate seamless integration of causal inference mechanisms within AI-driven healthcare architectures. By delineating a multi-layered framework that incorporates causal pathways for data ingestion, intelligence processing, and treatment orchestration, CONTI addresses interoperability gaps in electronic health records (EHRs) and decision support pipelines. The model emphasizes governance protocols to mitigate risks such as algorithmic drift and bias propagation, ensuring ethical deployment in diverse clinical environments. Theoretical analyses explore the dynamics of causal feedback loops, highlighting their role in enhancing personalized interventions while minimizing monitoring burdens. Conceptual formulas are introduced to interpret risk propagation, decision confidence intervals, and resource allocation efficiencies. Drawing from recent literature on clinical AI architectures and healthcare analytics, this work synthesizes infrastructural insights to advance AI governance in treatment personalization. Ultimately, CONTI offers a blueprint for future AI ecosystems that prioritize causal intelligence, fostering resilient and equitable healthcare delivery without relying on empirical data or performance metrics.

The integration of large language models (LLMs) into clinical decision infrastructures represents a transformative shift in healthcare delivery, enabling enhanced reasoning, data synthesis, and adaptive support for clinicians. This conceptual manuscript proposes a novel architecture, termed the adaptive LLM-orchestrated clinical ecosystem (ALOCE), designed to seamlessly embed LLMs within existing electronic health record (EHR) systems, interoperability frameworks, and governance protocols. By delineating a multi-layered structure encompassing data ingestion, semantic processing, decision augmentation, and continuous monitoring, ALOCE addresses key challenges such as data silos, ethical AI deployment, and real-time adaptability in clinical environments. Drawing on theoretical foundations from AI governance and healthcare informatics, the architecture incorporates feedback topologies for drift detection and ethical alignment, ensuring robustness in diverse clinical workflows. Conceptual formulas are introduced to model risk propagation across layers, decision confidence thresholds, and governance load balancing, providing interpretive tools for system designers. The manuscript synthesizes recent literature on clinical AI architectures, highlighting interoperability standards like FHIR and the role of LLMs in augmenting human decision-making without empirical validation. Ultimately, this work outlines a blueprint for scalable, ethical LLM integration, fostering improved patient outcomes through intelligent infrastructure orchestration. While theoretical, the implications extend to policy, deployment strategies, and future research in AI-driven healthcare systems.

The escalating prevalence of mental health crises necessitates innovative approaches to proactive intervention within longitudinal care ecosystems. This conceptual manuscript introduces the mental health crisis anticipation intelligence loop (MHCAIL), a theoretical architecture designed to integrate artificial intelligence (AI) for anticipating and mitigating crises in ongoing patient care pathways. By synthesizing clinical AI system architectures, healthcare analytics infrastructures, and electronic health record (EHR) intelligence ecosystems, MHCAIL establishes a closed-loop mechanism that processes multimodal data streams—such as EHR entries, wearable sensor inputs, and patient-reported outcomes—to generate anticipatory alerts. The framework emphasizes interoperability with existing decision support pipelines and AI governance protocols to ensure ethical deployment. Key components include predictive analytics layers for crisis risk stratification, adaptive feedback topologies for continuous system refinement, and monitoring interfaces to balance clinical workflow integration. Conceptual formulas model risk propagation dynamics and decision confidence thresholds, highlighting interpretive insights into resource allocation and governance burdens. While avoiding empirical evaluations, this work delineates theoretical implications for enhancing patient safety in mental health settings, fostering resilient longitudinal care systems that preemptively address vulnerabilities. Ultimately, MHCAIL advocates for a paradigm shift toward intelligence-driven anticipation, bridging gaps in current healthcare infrastructures to support timely, personalized interventions.

In an era of escalating healthcare demands, hospitals face persistent challenges in maintaining operational resilience amid fluctuating patient volumes, resource constraints, and unforeseen disruptions. This conceptual manuscript introduces a novel framework for real-time hospital capacity intelligence, designed to enhance decision-making through integrated AI-driven analytics and interoperable data ecosystems. Drawing on theoretical foundations from clinical AI architectures, healthcare analytics infrastructures, and decision support pipelines, the proposed system emphasizes seamless integration with electronic health records (EHRs), governance mechanisms for AI deployment, and dynamic monitoring to mitigate risks such as capacity overloads. The framework outlines a layered architecture that orchestrates data exchange, predictive analytics, and adaptive resource allocation, ensuring interoperability across clinical workflows. Key conceptual formulas are presented to interpret risk propagation in capacity management, decision confidence in real-time intelligence, and governance load in system operations. By synthesizing recent peer-reviewed literature on AI governance and clinical interoperability, this work highlights the potential for such frameworks to foster resilient hospital operations without relying on empirical data or model evaluations. Implications for healthcare systems include improved preparedness for surges, ethical AI integration, and scalable intelligence ecosystems. This theoretical exploration underscores the need for robust, AI-augmented infrastructures to support sustainable healthcare delivery.

The integration of artificial intelligence (AI) into telehealth networks has revolutionized remote patient monitoring, enabling real-time data analysis and decision support across distributed healthcare ecosystems. However, the governance of these AI-embedded systems remains underexplored, particularly in ensuring ethical oversight, data interoperability, and risk mitigation within networked environments. This conceptual manuscript proposes a novel governance architecture designed specifically for AI-embedded telehealth networks, emphasizing modular layers for monitoring orchestration, ethical compliance, and adaptive feedback mechanisms. Drawing on theoretical foundations from clinical AI infrastructures and healthcare analytics, the architecture introduces a unique framework termed the telehealth AI governance lattice (TAGL), which incorporates layered structures for data ingestion, AI inference governance, and network-wide monitoring. Key components include interoperability protocols to facilitate seamless data exchange among electronic health records (EHRs) and wearable devices, alongside interpretive formulas for assessing governance load and decision confidence. The manuscript synthesizes recent literature on AI system architectures in healthcare, highlighting gaps in remote monitoring governance and proposing theoretical pathways for integration into clinical workflows. By focusing on conceptual dynamics rather than empirical implementations, this work offers a blueprint for enhancing trust, scalability, and resilience in AI-driven telehealth systems. Ultimately, the TAGL framework aims to address the complexities of distributed AI governance, fostering equitable access to remote monitoring while mitigating potential biases and security vulnerabilities in networked healthcare delivery.

The rapid evolution of artificial intelligence (AI) in healthcare has paved the way for sophisticated systems aimed at enhancing early cancer detection across distributed clinical environments. This conceptual manuscript introduces the multi-center early detection orchestration network (MEDON), a novel intelligence architecture designed to integrate AI-driven analytics within multi-center screening ecosystems. MEDON conceptualizes a layered framework that facilitates seamless data interoperability, real-time decision support, and governance mechanisms to mitigate risks in federated healthcare settings. Drawing from theoretical foundations in clinical AI architectures and healthcare informatics, the architecture emphasizes modular components for intelligence orchestration, including adaptive monitoring pipelines and federated learning constructs without empirical validation. Key elements include interoperability frameworks for electronic health records (EHRs) and imaging data exchange, alongside governance models to ensure ethical deployment. The manuscript explores theoretical implications for workflow integration in screening programs, highlighting potential enhancements in detection sensitivity through conceptual risk propagation models and decision confidence formulas. By synthesizing recent literature on AI system architectures in oncology, this work proposes a blueprint for scalable, resilient intelligence ecosystems that could transform multi-center cancer screening paradigms. Ultimately, MEDON offers a theoretical pathway toward more equitable and efficient early detection strategies, addressing challenges in data silos and regulatory compliance across diverse clinical sites.

The rapid integration of generative artificial intelligence (AI) into clinical ecosystems has revolutionized the generation and utilization of synthetic health data, offering unprecedented opportunities for enhanced analytics, decision support, and personalized medicine while simultaneously raising critical governance concerns. This conceptual manuscript proposes a novel framework—the synthetic health orchestration and governance ecosystem (SHOGE)—designed to address the multifaceted challenges of data privacy, interoperability, ethical deployment, and continuous monitoring in generative AI-enabled environments. Drawing from theoretical models of AI system architectures and healthcare analytics infrastructures, SHOGE incorporates a layered orchestration topology that facilitates secure data exchange, real-time governance enforcement, and adaptive workflow integration. The framework emphasizes theoretical constructs such as risk propagation dynamics, decision confidence calibration, and governance load distribution, formalized through interpretive formulas to guide infrastructural design without empirical validation. By synthesizing literature on EHR intelligence ecosystems and AI monitoring systems, this work highlights operational sensitivities and human-AI interaction shifts, advocating for a balanced approach to innovation and risk mitigation. Ultimately, SHOGE provides a high-level blueprint for stakeholders to foster trustworthy generative AI applications in clinical settings, promoting equitable health outcomes and sustainable ecosystem evolution. This conceptual exploration underscores the need for proactive governance to harness synthetic health data’s potential while safeguarding patient trust and system integrity.

The integration of large language models (LLMs) into clinical healthcare systems represents a transformative shift in how data analytics, decision support, and operational infrastructure are conceptualized and deployed. This narrative review synthesizes recent advancements in LLMs within healthcare, focusing on their roles in enhancing clinical analytics, infrastructural frameworks, and oversight mechanisms while addressing inherent risk dynamics. Drawing from peer-reviewed literature, we examine how LLMs facilitate the processing of vast unstructured clinical data, such as electronic health records and patient narratives, to generate actionable insights that inform diagnostics, treatment planning, and resource allocation. Key infrastructural elements include scalable deployment pipelines that integrate LLMs with existing hospital information systems, enabling real-time analytics and predictive modeling without disrupting legacy workflows. Oversight is emphasized through regulatory frameworks that ensure ethical deployment, data privacy compliance, and bias mitigation, as LLMs amplify risks related to misinformation, algorithmic opacity, and equitable access in diverse clinical settings. Risk dynamics are explored in terms of model hallucinations, dependency on training data quality, and potential for exacerbating healthcare disparities if not properly governed. The review highlights systems-level analytics where LLMs contribute to closed-loop healthcare ecosystems, from data ingestion and inference to feedback-driven recalibration, fostering adaptive intelligence in clinical decision-making. For instance, LLMs have been adapted for tasks like text summarization, diagnostic reasoning, and patient communication, outperforming traditional methods in efficiency while requiring robust validation to maintain clinical fidelity. We underscore the need for interdisciplinary collaboration between clinicians, data scientists, and policymakers to harness LLMs' potential in optimizing healthcare delivery. By synthesizing cross-study evidence, this review proposes an original interpretive framework for LLM-enabled healthcare systems, structured around data-model-deployment-governance cycles, to guide future implementations. Ultimately, while LLMs promise enhanced analytics and infrastructural resilience, their clinical adoption demands vigilant oversight to balance innovation with patient safety and ethical integrity. This synthesis not only maps the current landscape but also identifies infrastructural gaps in scaling LLMs for equitable, high-stakes clinical environments, paving the way for more resilient healthcare analytics paradigms.

Generative artificial intelligence (GenAI) has emerged as a transformative force in healthcare systems, enabling advanced analytics, personalized interventions, and streamlined governance frameworks. This narrative review synthesizes recent literature on GenAI’s integration into healthcare infrastructures, emphasizing systems governance, safety protocols, and accountability mechanisms. We explore how GenAI enhances clinical decision-making, data analytics, and closed-loop systems while addressing ethical, regulatory, and operational challenges.At the core of healthcare systems, GenAI facilitates intelligent analytics by generating synthetic data for training models, simulating patient outcomes, and optimizing resource allocation. Governance frameworks are critical for ensuring responsible deployment, with studies highlighting the need for institutional guidelines that mitigate risks such as bias amplification and data privacy breaches. Safety considerations encompass algorithmic transparency, error detection in generative outputs, and human oversight in clinical loops. Accountability extends to lifecycle management, from model development to post-deployment monitoring, as evidenced by global initiatives and regional models like those in the GCC.The review delineates the landscape of GenAI applications in healthcare analytics, including predictive modeling for chronic disease management and real-time decision support. We propose an original systems-level framing that integrates data ingestion, inference generation, intervention deployment, and feedback recalibration under governance umbrellas. This synthesis reveals gaps in current infrastructures, such as the lack of standardized AI guardians for information overload and the challenges of scaling enterprise AI.In examining intelligent clinical decision systems, we highlight architectures that fuse GenAI with electronic health records (EHRs) for closed-loop operations, where generative models inform adaptive interventions. Ethical considerations are woven throughout, advocating for principles adapted from military contexts to healthcare. The adoption of GenAI in US hospitals underscores its potential for inpatient summaries and chronic care, yet calls for regulatory oversight to align with Helsinki declarations.Ultimately, this review positions GenAI as a cornerstone for accountable healthcare systems, urging interdisciplinary collaboration to balance innovation with safety. By synthesizing governance models, safety protocols, and accountability structures, we provide a roadmap for sustainable integration, fostering equitable health outcomes in an AI-augmented era.

The integration of artificial intelligence (AI) into healthcare systems and analytics represents a transformative shift toward more efficient, personalized, and predictive clinical practices. However, this evolution necessitates robust governance frameworks to ensure transparency, mitigate biases, and enable continuous lifecycle monitoring. This narrative review synthesizes recent literature on AI governance in healthcare, focusing on systems-level infrastructure and clinical analytics. Drawing from peer-reviewed publications, we examine how AI tools enhance healthcare delivery through data-driven insights while addressing ethical, regulatory, and operational challenges.Central to AI governance is transparency, which involves making algorithmic processes interpretable to clinicians and stakeholders. Studies highlight the need for explainable AI models in clinical decision-making, where opaque “black-box” systems can undermine trust and accountability. For instance, frameworks for implementing machine learning in healthcare emphasize ethical considerations, such as disclosing model limitations and decision rationales to prevent misinformed clinical actions. Bias mitigation emerges as a critical pillar, with research demonstrating how algorithmic biases in electronic health records can perpetuate health disparities, particularly among underrepresented populations. Strategies include proactive monitoring of algorithms for equity, incorporating diverse datasets during development, and post-deployment audits to detect and correct biases.Lifecycle monitoring models ensure sustained performance and safety of AI systems over time. This encompasses ongoing evaluation, recalibration, and governance structures that adapt to evolving clinical environments. Nationwide initiatives propose AI assurance laboratories to standardize monitoring, while institutional guidelines advocate for step-by-step implementation to avoid “AI winters” caused by unaddressed failures. In analytics contexts, large AI models facilitate health informatics by processing vast datasets for predictive analytics, yet they require governance to handle challenges like data privacy and model drift.The review structures its synthesis around healthcare systems’ end-to-end loops: from data ingestion to intelligent decision support and closed-loop interventions. It integrates cross-study analyses to propose original interpretive models for governance, emphasizing human-AI collaboration in clinical workflows. Key findings underscore the importance of multidisciplinary approaches, combining technical, ethical, and regulatory perspectives to foster responsible AI adoption. Ultimately, effective governance not only enhances patient outcomes but also builds public trust in AI-driven healthcare. This synthesis highlights gaps in current practices and advocates for integrative monitoring systems to realize AI’s full potential in equitable healthcare delivery.

Surgical site infections (SSIs) affect 2–20% of surgical procedures and are a major source of postoperative morbidity, prolonged hospitalization, readmission, mortality, and healthcare costs, making prevention a key priority. Existing prediction tools such as the NNIS index and SENIC score depend on a limited set of clinical variables including wound class, ASA status, and operative duration, while failing to capture complex interactions among patients, surgeons, and comorbidities. To address this limitation, we propose a graph attention network (GAT) framework that represents each surgical case as a heterogeneous graph composed of patient, surgeon, and comorbidity nodes, with intraoperative variables included as features and attention mechanisms used to learn the most influential relationships. This approach models relational dependencies such as the interaction between surgeon experience, patient conditions, and comorbidity combinations, enabling more accurate and context-aware SSI risk prediction to support personalized preventive interventions.

The integration of artificial intelligence into clinical decision support systems offers improved diagnostic accuracy and efficiency, but the opacity of many machine learning models raises concerns about trust, accountability, and regulatory compliance. Explainable artificial intelligence (XAI) has been proposed to address this by making model predictions interpretable to clinicians; however, its true clinical value remains uncertain, and evaluation has not kept pace with methodological development. This systematic review aimed to identify XAI methods used in clinical decision support systems, assess how they are evaluated with clinicians, and determine whether explanations improve diagnostic accuracy, trust, mental models, and efficiency. Following PRISMA guidelines, we searched PubMed, Web of Science, IEEE Xplore, ACM Digital Library, and Scopus for studies published between 2017 and 2024. Eligible studies included original research evaluating XAI in clinical decision support systems with clinician participants and reporting quantitative or qualitative outcomes. Risk of bias was assessed using adapted QUADAS-2 and ROBIS tools, and findings were synthesized narratively with subgroup analyses. From 2,847 records, 68 studies were included. The most common XAI methods were SHAP-based feature attribution (38%), saliency or heatmap methods (29%), concept-based approaches such as TCAV (15%), and counterfactual or example-based explanations (12%). Radiology was the dominant field (54%), followed by dermatology (18%) and pathology (12%). Evaluation approaches were highly inconsistent, with few validated instruments and most studies relying on Likert-scale trust measures or qualitative feedback. Only 16% of studies showed improved diagnostic accuracy with explanations, 67% showed no significant effect, and 17% reported reduced accuracy due to over-reliance or misinterpretation. Although 82% of studies reported increased clinician trust, trust rarely correlated with actual diagnostic performance. Overall, while XAI methods are widely studied in clinical decision support, their evaluation is inconsistent and their benefits are limited. Explanations tend to increase clinician trust without reliably improving diagnostic accuracy, and may sometimes worsen performance, highlighting a trust–accuracy gap that poses important safety concerns for clinical deployment.

Federated learning (FL) is promoted as a privacy-preserving method for training machine learning models across healthcare institutions without sharing patient data, with growing use in medical imaging, electronic health records, and rare disease research. This critical review examines FL studies from 2017–2024, focusing on privacy guarantees, statistical heterogeneity, communication efficiency, and real-world clinical deployment. A structured search of PubMed, IEEE Xplore, arXiv, and Google Scholar was conducted using relevant FL and healthcare terms, including studies addressing privacy, heterogeneity, communication, or deployment. Reported privacy guarantees are often overstated, with most studies relying on FedAvg without differential privacy. Statistical heterogeneity in non-IID settings remains largely unresolved. Fewer than 5% of studies report real-world deployment, typically at very small scale. A significant gap exists between FL research and clinical application. Current methods fall short of healthcare-grade privacy and real-world constraints, limiting readiness for high-stakes clinical use.

Postoperative complications including SSI (2–20%), VTE (1–5%), and respiratory failure (1–8%) significantly increase morbidity, mortality, length of stay, and readmissions. This systematic review assessed machine learning models predicting these outcomes, their performance, external validation, and clinical deployment. A PRISMA-based search (2017–2024) identified 32 eligible studies. Models such as random forest and XGBoost showed AUROC ranges of 0.70–0.85 for SSI, 0.75–0.90 for VTE (outperforming Caprini scores), and 0.75–0.88 for respiratory failure. However, fewer than 20% of studies included external validation and less than 5% reported clinical deployment. Overall, while machine learning models show strong retrospective performance, limited validation and minimal real-world implementation remain major barriers to clinical translation.

Suicidality and depression are major global health burdens, with over 700,000 suicide deaths annually and ~280 million people affected by major depressive disorder. Early risk prediction could support prevention, but traditional methods show limited accuracy. This PRISMA-compliant systematic review evaluated machine learning models for predicting suicidality and depression across electronic health records, social media, and wearable sensor data, focusing on performance, unimodal vs multimodal approaches, and ethical reporting. Searches of PubMed, PsycINFO, IEEE Xplore, arXiv, and ACM Digital Library identified eligible studies. EHR-based models showed AUROC 0.70–0.85 for suicide attempt prediction, social media models 0.70–0.80 for suicidal ideation, and wearable sensor models lower performance (0.65–0.75). Multimodal approaches improved performance by 5–10% over unimodal models. However, fewer than 20% of studies reported ethical considerations such as privacy, bias, or deployment safeguards. Overall, machine learning shows moderate-to-good predictive performance, with multimodal models performing best, but ethical reporting remains critically insufficient for clinical translation.

Clinicians often need rapid, evidence-based answers that integrate patient-specific electronic health records (EHRs) with clinical guidelines, but existing decision support tools are limited in real-time personalization. While large language models (LLMs) offer strong medical reasoning, they are prone to hallucinations and lack direct access to local EHR data, making them unsafe for standalone clinical use; meanwhile, traditional retrieval systems cannot synthesize coherent, context-aware responses. This paper proposes a retrieval-augmented generation (RAG) framework that combines dual-source retrieval from both institutional EHRs and clinical guideline databases. The system includes an EHR indexer, a guideline repository, a semantic retriever, an LLM-based generator, and a safety filter for hallucination mitigation. By grounding outputs in retrieved patient data and evidence-based recommendations, the model improves factual reliability, explainability, and clinical trustworthiness. Overall, the framework enables safe, real-time clinical question answering by integrating LLM reasoning with verified medical sources, with future validation planned on public EHR and guideline datasets.

Hospital discharge summaries are critical for care transitions, directly impacting readmission prevention and medication reconciliation, yet physicians spend 15-30 minutes per patient drafting these documents, contributing substantially to documentation burden and professional burnout. Manual summarization of daily progress notes and laboratory results is repetitive, time-consuming, and error-prone, as clinicians must sift through lengthy unstructured notes across multiple hospital days while identifying salient events and trends. We propose a large language model with parameter-efficient fine-tuning for automated discharge summary generation that processes chronologically ordered daily progress notes alongside time-series laboratory results to produce structured discharge documentation. The framework consists of a base LLM augmented with LoRA adapters, a progress note encoder for section segmentation, a laboratory result integrator that computes trend indicators, and a summary generator that produces sectioned discharge output. Parameter-efficient fine-tuning enables domain adaptation to clinical text with minimal computational resources, preserving patient-specific information while reducing hallucination through retrieval of key factual details from the input notes. This framework offers a practical pathway to reduced documentation burden and improved discharge quality, with potential for widespread deployment across health systems given the modest computational requirements of PEFT approaches.

Type 1 diabetes mellitus requires exogenous insulin and accurate glucose forecasting is critical for closed-loop artificial pancreas systems. While continuous glucose monitoring provides real-time data, purely data-driven recurrent neural networks may produce physiologically implausible predictions, and purely mechanistic models cannot fully capture individual variability in insulin sensitivity, meal absorption, or exercise response. This framework proposes a physics-guided recurrent neural network that integrates insulin delivery records, carbohydrate intake, and physical activity data. It combines a mechanistic glucose–insulin compartmental model with a residual LSTM network that learns patient-specific deviations, supported by a physics-based loss function enforcing physiological constraints such as non-negativity and realistic glucose dynamics. By merging physiological modeling with deep learning, the system preserves biological plausibility while adapting to individual patient patterns. Incorporating multimodal wearable and device data enables more accurate, longer-horizon glucose predictions, supporting safer and more proactive insulin dosing in closed-loop diabetes management.

Rare diseases collectively affect over 300 million people globally, yet individual conditions are often missed due to low clinician familiarity and non-specific presenting symptoms that mimic common disorders. Supervised machine learning requires large numbers of labeled examples for training, but rare diseases have too few diagnosed cases to develop condition-specific predictive models using traditional approaches. We propose a multimodal foundation model pretrained on 10 million de-identified electronic health records (EHRs) combining clinical notes and laboratory values for zero-shot rare disease diagnosis without requiring labeled training examples. The framework comprises four components: a clinical note encoder based on a large language model, a laboratory value encoder using a time-series transformer, a multimodal fusion module with cross-attention, and a zero-shot classifier that compares patient embeddings to disease descriptions. Pretraining on large-scale EHR data enables the model to learn general medical knowledge and disease patterns, allowing diagnosis of rare conditions by recognizing manifestations even when no labeled examples of that specific disease were used for training.

Pandemic surges can rapidly overwhelm hospital capacity, where shortages of beds and nurse fatigue contribute directly to increased excess mortality, making coordinated decision-making across emergency departments, intensive care units, and general wards essential yet difficult to achieve under centralized control systems. Centralized approaches to bed allocation and nurse staffing optimization are limited because each hospital unit holds critical local information—such as real-time patient acuity, staff availability, and infection control status—that cannot be easily shared due to privacy constraints and communication delays during crisis conditions. To address these challenges, we propose a federated multi-agent reinforcement learning framework that enables coordinated decision-making for bed distribution and nurse staffing across hospital units without requiring centralization of sensitive clinical or workforce data. The system consists of local reinforcement learning agents deployed in each unit that participate in federated aggregation, a coordination mechanism that aligns inter-unit policies, and a surge detection module that dynamically switches operational strategies during pandemic escalation periods. This distributed architecture maintains data privacy while supporting adaptive, system-wide coordination under surge conditions, overcoming the limitations of both centralized optimization models and rule-based heuristic approaches.

Long COVID (post-acute sequelae of SARS-CoV-2 infection, PASC) affects roughly 10–30% of COVID-19 survivors and is marked by persistent symptoms such as fatigue, cognitive dysfunction (“brain fog”), shortness of breath, loss of smell, and post-exertional malaise that can last for months or years, while its underlying biological mechanisms and validated diagnostic biomarkers remain unclear. The condition is highly heterogeneous, with patients showing different recovery patterns and no clearly defined clinical subtypes, and the scarcity of labeled datasets further limits the use of supervised machine learning methods for phenotyping. To address this, we propose a self-supervised contrastive multi-view learning framework that integrates three temporal data modalities—pre-infection electronic health records, acute-phase clinical and biomarker data (e.g., CRP, ferritin, D-dimer, lymphocyte counts), and post-acute symptom trajectories—using separate encoders and a shared latent space aligned through contrastive learning without requiring phenotype labels, followed by unsupervised clustering to identify potential subtypes. By exploiting the natural temporal linkage within each patient and contrasts across patients, this approach enables data-driven discovery of long COVID phenotypes, supports early prediction of subgroup membership, and may ultimately inform personalized treatment strategies, clinical trial design, and improved understanding of disease mechanisms.

Deep learning models for arrhythmia detection require large, balanced datasets to achieve clinically acceptable performance. Rare arrhythmias such as ventricular tachycardia, ventricular fibrillation, and complete heart block are severely under-represented in public ECG repositories, leading to classifiers that perform well on normal sinus rhythm but fail catastrophically on minority classes. Traditional data augmentation techniques including scaling, noise addition, and time warping cannot generate new arrhythmia morphological patterns. Real-world collection of rare arrhythmia events is impractical due to low prevalence, ethical constraints, and the need for expert annotation. We present a diffusion-based generative framework that synthesizes realistic ECG signals with controlled arrhythmia patterns. The architecture comprises a conditional denoising diffusion probabilistic model trained on a small set of labeled arrhythmia examples, enabling unlimited generation of specific arrhythmia types including atrial fibrillation, ventricular tachycardia, and premature ventricular contractions. The framework includes three core components: (1) an ECG diffusion model with a 1D U-Net denoising architecture, (2) a condition encoder that accepts arrhythmia class labels and optional morphological parameters, and (3) a downstream classifier training pipeline that leverages synthetic data to correct class imbalance. This approach generates unlimited realistic arrhythmia examples with preserved morphological features including QRS duration, QT interval, and RR interval dynamics. The generative process inherently resists membership inference attacks, providing a privacy-preserving alternative to sharing real patient ECGs. The proposed framework offers a viable pathway toward balanced, privacy-preserving ECG datasets for arrhythmia detection, requiring only a small seed set of labeled rare arrhythmia examples to generate clinically useful synthetic data.

Chronic postsurgical pain (CPSP) affects 10–50% of surgical patients and is a major contributor to long-term opioid use and reduced quality of life. Current predictive models treat patients independently and fail to capture how risk evolves over time or how postoperative opioid trajectories influence divergence in outcomes. We propose a dynamic graph neural network (GNN) framework in which patients are modeled as nodes and similarity-based edges evolve over time based on opioid prescription patterns, pain scores, and preoperative psychological factors. The model includes (1) a patient graph with static preoperative features, (2) a temporal edge update mechanism, (3) a GNN message-passing layer that aggregates information from dynamically connected patients, and (4) a prediction head estimating CPSP risk at 3, 6, and 12 months. By modeling changing patient relationships after surgery, the framework captures how similar patients may diverge or converge depending on postoperative management, enabling more accurate and personalized CPSP risk prediction using longitudinal electronic health record data.

Prostate cancer metastasis to bone and lymph nodes marks a critical transition to incurable disease, with five-year survival dropping dramatically compared to localized disease. Early identification of patients at high risk of metastasis enables timely intensification of treatment, including androgen deprivation therapy, salvage radiation, or systemic therapies. Current deep survival models that integrate serial PSA measurements and genomic risk scores achieve high predictive accuracy for time-to-metastasis but operate as black boxes, providing no explanation for why a particular patient is predicted to have early or late metastasis. Clinicians cannot trust or act upon predictions without understanding which PSA features or genomic markers drive the risk assessment. We present an explainable deep survival framework that combines a deep survival model for time-to-metastasis prediction with Integrated Gradients attribution, a method that distributes the model's hazard prediction among input features. The framework produces patient-specific explanations showing how each serial PSA value and each genomic score component contributes to the predicted metastasis hazard. The framework consists of three core components: (1) a deep survival model (DeepSurv architecture) with a PSA time-series encoder and genomic risk encoder, (2) Integrated Gradients attribution computed over the hazard function, and (3) visualization tools for individual and population-level interpretations. Integrated Gradients attributes the predicted hazard to individual PSA measurements across time and specific genomic markers, enabling clinicians to distinguish between risk driven by rapid PSA kinetics versus high genomic risk scores. This interpretability transforms a black-box survival prediction into an actionable clinical decision support tool.

Hypertension affects over 1.4 billion adults worldwide, and antihypertensive dose titration is a common but complex clinical decision. Although electronic health records contain longitudinal data on medication adjustments and blood pressure outcomes, determining optimal individualized dosing remains challenging due to confounding in observational data, where patients receiving higher doses often have worse baseline health. We propose a transformer-based model with causal attention masking to estimate counterfactual blood pressure outcomes under alternative dose regimens. The architecture ensures temporal validity by preventing information leakage from future events and encodes medication dose changes in a continuous representation. It includes a dose encoder, outcome predictor, and counterfactual contrastive loss to distinguish between competing treatment paths. This framework learns patient-specific dose–response relationships and enables personalized predictions for antihypertensive adjustments. While it supports individualized treatment planning from observational EHR data, prospective validation is still required before clinical deployment.

Healthcare billing fraud imposes major financial losses globally, costing public and private payers hundreds of billions annually. It exploits fragmented healthcare payment systems where multiple insurers process overlapping patient populations without coordination, creating blind spots that enable sophisticated cross-payer fraud schemes. Individual payers cannot detect patterns such as duplicate billing across Medicare and commercial insurers because current detection models operate within isolated organizational and regulatory boundaries. Strict privacy laws like HIPAA and GDPR further prevent sharing patient-level claims data, limiting centralized analytics. To address this, a federated anomaly detection framework is proposed in which autoencoders are trained locally at each payer without exchanging raw data. Each institution learns normal billing patterns through reconstruction-based unsupervised learning and identifies anomalies via reconstruction error. A central server aggregates encoder parameters using FedAvg, optionally with differential privacy, to build a globally informed model while preserving data locality. The resulting system enables detection of cross-payer fraud patterns, such as double billing and unbundling, that single-payer systems miss, while transmitting only model parameters through secure channels. This approach provides a privacy-preserving, scalable solution for multi-payer healthcare fraud detection under strict regulatory constraints.

Atrial fibrillation affects over 30 million people worldwide and requires long-term anticoagulation, with warfarin still widely used due to its efficacy and reversibility, but its narrow therapeutic window (INR 2.0–3.0) makes dosing particularly challenging, especially in high bleeding-risk patients where both under- and over-anticoagulation can lead to serious complications. Conventional dosing approaches rely on population-based nomograms and clinician judgment, failing to capture individual variability driven by genetics, diet, comorbidities, and drug interactions. To address this limitation, this article proposes a conceptual framework that integrates deep reinforcement learning with a safety-shield mechanism for personalized warfarin dosing. The system uses a deep Q-network trained on historical patient trajectories within an offline Markov Decision Process to recommend dose adjustments based on INR history and clinical risk factors, while a deterministic rule-based safety layer blocks unsafe actions, such as dose increases when INR exceeds 3.5 or extreme adjustments requiring clinician review. Conservative offline reinforcement learning further reduces the risk of unsafe policy extrapolation by limiting overestimation of out-of-distribution actions. Together, this hybrid architecture aims to improve time in therapeutic range while minimizing bleeding risk, providing a structured and clinically constrained approach for safer, individualized anticoagulation management in high-risk atrial fibrillation patients.

Rare pediatric cancers are difficult to treat due to their very low incidence, which limits drug development and makes experimental screening of therapies slow, costly, and dependent on scarce tumor samples. Traditional supervised machine learning approaches are also constrained by the lack of labeled drug–response data, while rich but unlabeled protein–protein interaction networks remain underutilized. We propose a self-supervised graph representation learning framework that integrates protein interaction networks with patient gene expression data to support drug repurposing. The model builds a heterogeneous graph of drugs, genes, diseases, and proteins, and uses a graph neural network trained with self-supervised objectives such as contrastive learning and masked prediction to learn molecular representations without labeled data. It is then fine-tuned on small pediatric cancer datasets. The framework enables prediction of candidate drug therapies by combining learned biological network representations with disease-specific expression profiles. This approach reduces reliance on large labeled datasets and allows adaptation to rare cancer contexts, offering a scalable strategy for computational drug repurposing in pediatric oncology.

Skilled nursing facilities (SNFs) in the U.S. serve over 1.5 million residents and experience continuous census volatility driven by admissions, discharges, and mortality, impacting staffing, bed availability, and care quality. Existing forecasting methods rarely capture these dynamics together, leading to reactive and inefficient operational decisions. A need exists for accurate, multi-horizon, and data-integrated forecasting systems. Traditional models like ARIMA and LSTM are limited in SNF census forecasting because they produce single-point estimates, fail to model uncertainty, and cannot effectively integrate heterogeneous data such as facility characteristics, temporal utilization patterns, and external factors like COVID-19 prevalence. They also lack interpretability, reducing their usefulness for decision-making. This study introduces an attention-based Temporal Fusion Transformer (TFT) for multi-horizon SNF census forecasting (1, 7, 14, and 30 days). It integrates admissions, discharges, and COVID-19 prevalence through dedicated encoders and applies variable selection networks, LSTM layers, and multi-head attention to capture temporal dependencies and feature importance. The model outputs quantile forecasts (10th, 50th, 90th percentiles) to quantify uncertainty. The TFT enhances interpretability by identifying which past events and features most influence predictions at each horizon, enabling administrators to understand how admissions trends, discharge patterns, and COVID-19 surges affect census dynamics. The proposed framework enables proactive SNF capacity planning by combining multi-source data with interpretable, uncertainty-aware forecasting, supporting a shift from reactive staffing to anticipatory resource allocation and improved operational efficiency.