Frontiers in Bioinformatics

Background: Early detection of pancreatic ductal adenocarcinoma (PDAC) is critical. While faecal elastase-1 (FE-1) is a standard clinical marker for pancreatic function, its diagnostic accuracy for malignancy is limited. We sought to identify plasma metabolites that enhance FE-1 performance in symptomatic "at-risk" patients. Methods: Using the DEPEND cohort (CRUK C45617/A29908), plasma metabolomics was performed on patients with resectable PDAC (n=23) and healthy volunteers (n=24). Predictive modelling included feature selection and cross-validation, with further validation in an independent external cohort. Results: Citrulline was identified as significantly depleted in PDAC patients across discovery and validation cohorts. In isolation, Citrulline achieved an AUC of 0.86 (internal) and 0.88 (external validation). Standalone FE-1 demonstrated an AUC of 0.67. However, combining Citrulline and FE-1 significantly improved diagnostic performance, achieving a combined AUC of 0.96. Stratification revealed distinct metabolomic signatures associated with poorly differentiated tumours, suggesting a link to histological grade. Conclusions: Integrating Citrulline with FE-1 testing substantially improves PDAC detection in symptomatic patients. This non-invasive panel offers high diagnostic potential, though prospective validation is required to establish clinical cut-offs for routine practice.

The therapeutic landscape for hepatocellular carcinoma (HCC) is evolving rapidly, necessitating scalable approaches to synthesize the expanding scientific literature. We characterized thematic shifts in HCC treatment and prognosis research by conducting a retrospective bibliometric analysis of influential publications from 2023 and 2024. Using the OpenAlex database, we identified the 50 most highly cited papers from each year based on eighteen-month post-publication citation counts. Large language models were deployed to extract, normalize, and classify concepts from unstructured text into canonical topics and parent themes, enabling quantitative year-over-year frequency comparisons. Analysis of these 100 papers revealed a distinct maturation in research focus. Although broad categories like general immunotherapy remained prevalent, their relative frequency declined in favor of specific dual immune checkpoint regimens, notably CTLA-4 inhibition and the durvalumab plus tremelimumab combination. Concurrently, parent themes related to radiomics, imaging, and health systems exhibited significant growth in the 2024 cohort. These findings demonstrate a thematic transition in high-impact HCC research from foundational immuno-oncology toward optimized combination therapies and precision diagnostics. Furthermore, this study highlights the utility of artificial intelligence-driven bibliometrics for objectively tracking dynamic conceptual shifts in oncology. A web interface for exploring the data is available at https://pri.pepkio.com/.

This study presents ReCo (Research Cosmos), a self-configuring and self-extending agentic research framework for the biomedical domain. ReCo is orchestrated by a large language model that interacts with native computing tools, bundled Model Context Protocol (MCP) servers, structured skills, persistent project memory, and a desktop interface. Its bundled MCP servers provide biomedical analysis capabilities while serving as implementation paradigms for integrating new computational and AI frameworks. Structured skills encode procedures for environment configuration and framework ingestion, enabling ReCo to inspect repositories, manuscripts, or local codebases; identify dependencies and execution patterns; create isolated runtime environments; design and implement MCP interfaces. Self-extension was evaluated using five heterogeneous systems: the Merlin computed tomography foundation model, MAISI-v2 medical image synthesis framework, asari liquid chromatography-mass spectrometry workflow, DosimeTron agentic radiation-dosimetry platform, and Orthanc DICOM server. ReCo successfully operationalized all five systems and completed predefined functional evaluations. Re-hosted DosimeTron outputs demonstrated near-perfect agreement with the reference pipeline across 651 organ observations (Pearson correlation and Lin concordance correlation coefficient, 0.99999; mean absolute percentage difference, 0.37%). Notably, ReCo configured Orthanc as a PACS-like coordination layer, integrated it with DosimeTron, Merlin, and TotalSegmentator, and orchestrated data retrieval, analysis, and return of valid DICOM RTSTRUCT, RTDOSE, and Structured Report. ReCo provides a unified environment for configuring, documenting, and operationalizing heterogeneous biomedical frameworks, reducing technical barriers to the adoption and integration of emerging computational and AI methods. The official open-source ReCo GitHub repository is available at: https://github.com/eltzanis/ReCo

Background: Changes in the extracellular matrix (ECM) are a recognised feature of aggressive prostate cancer, but they are not exploited in clinical decision-making. We aimed to develop automated quantitative ECM parameters to facilitate risk stratification for localised prostate cancer. Methods: 378 quantitative ECM parameters were derived from picrosirius red-stained diagnostic prostate biopsies in a cohort of 422 patients, matched 1:1 for recurrence, recruited to the CHHiP (Conventional or Hypofractionated High Dose Intensity Modulated Radiotherapy in Prostate Cancer) trial of radiotherapy fractionation for localised prostate cancer. These ECM parameters comprehensively described fibre architecture, gaps and ECM texture. Machine learning models at the level of both individual image tiles and patients defined how ECM parameters related to tumour versus normal prostate, Gleason grade group and recurrence. Shapley analysis was used to interpret ECM feature importance and develop signatures associated with recurrence. Results: Specific ECM patterns identified tumour versus normal prostate, Gleason pattern 4 versus 3 and recurrence. ECM patterns associated with recurrence were enriched in Gleason 4+3 patients, versus Gleason 3+4 patients. Shapley analysis revealed that biopsies from patients with recurrence had smaller more elongated gaps between fibres, with finer grained ECM texture and lower ECM homogeneity than less recurrent regions. Interpretation: Quantitative automated analysis of ECM architecture can inform probability of prostate cancer recurrence after radiotherapy; Features relating to ECM gap size and texture are of particular relevance.

Background: Increased public access to data from disparate sources provides opportunities to study and validate predictive and subphenotype models in heterogeneous disease conditions using aggregated individual patient data. Robust, explicit, and transparent harmonization of data elements is critical to ensure interpretability, reproducibility, and generalizability of secondary and retrospective analyses. Methods & Results: We designed and implemented ADAPT (Aggregating Data to Accelerate Personalized Therapy), a scalable framework using multiple software packages (R, SQL, BigQuery) that enables rapid, explicit harmonization of structured data elements from randomized trials and observational studies using a standard spreadsheet interface. User-specified criteria are applied to primary study data to produce harmonized longitudinal datasets comprised of demographics, medical history, quantitative observations, repeated measures, and clinical outcomes. We demonstrate this functionality using 26 clinical studies found in the National Heart, Lung, and Blood Institute BioLINCC resource. We illustrate the scalability of ADAPT to the order of billions of datapoints using administrative clinical data in a cloud-computing platform. We also present examples of collaborators using ADAPT for independent harmonization tasks for secondary analyses and democratization of publicly available data. Conclusion: ADAPT is a disease-agnostic, extensible, and scalable platform to support robust, transparent harmonization of structured research data using interfaces accessible to a variety of researchers regardless of programming ability. It extends FAIR principles beyond research data to also represent harmonization analyses by improving Findability of harmonization decisions, Accessibility of methods to other stakeholders, Interoperability with independent analyses and datasets, and Reusability through efficient implementation in a variety of analysis environments.

Summary It has become standard practice to visualize regional signals from genomewide association studies GWAS using LocusZoom plots Similarly GWAS signals are compared to regionally matched quantitative trait loci QTLs ie varianttogene regulation data using LocusCompare plots to aid assessment of candidate traitrelated genes Despite broad usage these tools annotate variants by linkage disequilibrium LD to a single lead or index variant This singleindex representation has limitations for visualizing complex loci that contain multiple independent signals We present LocusBlend an interactive web application for multiindex LDblended visualization of genomic loci LocusBlend supports one or two genomic association summarystatistic datasets and one to three index variants multiindex LocusZoom colorblended plots and matching LocusCompare visualizations Applications to Alzheimers disease GWAS and QTL signals illustrate LocusBlend enables visualization and separation of independent signals despite shared LD and high genomic complexity Overall LocusBlend is aimed at supporting researchers handle the continuously expanding complexity of human genomics findings Availability and Implementation LocusBlend is freely available at httpslocusblendwustledu Publication ready plots are generated in 1min Source code documentation example datasets input templates and reproducibility instructions are available at httpsgithubcomBelloyLabLocusBlend LocusBlend is implemented in Python using Streamlit Plotly and PLINK Supplementary Information Supplementary data are available online

Background. Translating a clinical guideline decision algorithm into a computational graph requires judgment, and unconstrained coding yields divergent graphs; any complexity measure computed from such a graph inherits that variation, so its reproducibility must be demonstrated rather than assumed. Objective. To develop, and prospectively test, an empirical method for making graph extraction reproducible, using the Clinical Guideline Complexity Index (CGCI) and four guideline algorithms as a case study. Methods. We built a Graph Representation Specification (an ontology, a motif catalogue, disambiguation conventions, decomposition rules, a deterministic validator, and a scoring engine) and refined it by error-driven grammar induction: measure inter-coder disagreement, localize its dominant class, induce a single grammar rule, and prospectively test whether that rule improves agreement in the anticipated class. Reproducibility was quantified with a pre-specified, topology-based endpoint (Decision Topology Agreement) rather than edge agreement, which is oversensitive to representational choices that do not affect the score. Two trained coders independently coded the diabetes, dyslipidemia, heart-failure, and hypertension algorithms. Results. A rule induced from the diabetes comorbidity panel (assessment topology) generated a pre-specified prediction that heart-failure figures, sharing the same motif, would converge; on a fresh, independently coded pair they did, with an absolute CGCI difference of approximately one. Decision topology reproduced closely (decision-order agreement at or near 1.00 for three of four guidelines), while breadth counting was rule-sensitive: an explicit modifier-counting rule reduced the largest disagreement from 27 to 4 tokens. Residual disagreement was bounded and localizable to specific, nameable representational choices. Conclusions. Graph-extraction reproducibility can be systematically improved through iterative grammar refinement, and a prospectively derived rule can be confirmed to improve agreement. These results establish the measurement foundation (reliability, not construct validity) for a companion study interpreting CGCI as cognitive load, and the method may apply wherever graphs are extracted from structured source artifacts.

While extensive cellular heterogeneity in renal cell carcinomas (RCC) is linked to diverse clinical outcomes, our understanding of this diversity is limited to those driven by clonal patterns or activity of canonical pathways. Here, we present a compendium of over 85,000 single-cell gene expression profiles from primary and metastatic tumors as well as patient-derived models across four RCC subtypes, including the rare clear cell papillary renal cell tumors, which we show are often misclassified and for which we identify CASP14 as a highly sensitive and specific biomarker. We dissect malignant cell variation within and across tumors using a generative modeling framework that accounts for clonal and copy number-driven expression shifts, defining 59 gene expression programs that deconstruct canonical pathways into functional submodules with divergent activity patterns, distinct regulators, and differential association with clinical outcomes. Despite the canonical view that VHL-deficient clear cell RCC exists in a constitutive pseudohypoxic state, we show strong intra-tumor variability of a hypoxia inducible factor 2 (HIF2)-driven program linked to poor outcome. We also identify early, spatially organized activation of a complete epithelial-to-mesenchymal transition (EMT) program, loss of epithelial identity, and upregulation of protein translation programs as key characteristics of metastatic progression. Finally, a metastatic signature capturing cellular de-differentiation and translational activity identifies primary tumors associated with adverse clinical outcomes. Together, this resource establishes a framework for dissecting malignant cell heterogeneity, refines RCC subtype classification, and defines transcriptional programs underlying metastasis progression.

INTRODUCTION: The biomarker-based amyloid/ tau/ neurodegeneration (A/T/N) framework has become a popular staging method for Alzheimer's disease (AD) research. Previous studies use the framework either as a rule-based or data-driven approach but typically sacrifice either adaptivity or interpretability. METHODS: We present an interpretable, hybrid method, called Neurosymodal Data Fusion, for predicting incident AD in the ADNI dataset. Specifically, we encode the A/T/N framework as a logic program, where the input biomarker features are extracted by one or more neural networks. RESULTS: Our pipeline predicted four-year incident AD with a sensitivity of up to 0.84. Additionally, our models learned scores for each A/T/N profile, denoting relative importances to model predictions. These scores also indicated that empirically-derived cut-off values for the A and T criteria might be uninformative for the ADNI data. DISCUSSION: Our pipeline provides a novel way to use the A/T/N framework that could potentially improve early AD screening years before clinical manifestations.

Chronic kidney disease and heart failure disproportionately burden populations of African ancestry, yet Mendelian randomisation (MR) studies of the causal relationship between kidney function and heart failure subtypes have been conducted exclusively in European ancestry populations. We performed a forward two-sample MR analysis to evaluate the causal effect of genetically predicted estimated glomerular filtration rate (eGFR) on heart failure with preserved ejection fraction (HFpEF) and heart failure with reduced ejection fraction (HFrEF) in individuals of African ancestry. Genetic instruments were selected from an African ancestry eGFR genome-wide association study (N = 67,943) at genome-wide significance, with linkage disequilibrium clumping using an African ancestry reference panel. Heart failure subtype summary statistics were obtained from the Million Veteran Program (HFpEF: 5,379 cases / 113,041 controls; HFrEF: 9,104 cases / 109,632 controls). Six independent SNPs (F-statistics 30.5 &#8211 107.3; R&#178 = 0.62%) were retained as instruments. The primary inverse-variance weighted analysis provided no evidence of a causal effect of eGFR on HFpEF (OR 0.92, 95% CI 0.80 &#8211 1.06, p = 0.248) or HFrEF (OR 0.98, 95% CI 0.78 &#8211 1.23, p = 0.878). Sensitivity analyses were directionally consistent. There was no evidence of heterogeneity or directional pleiotropy. Minimum detectable effects at 80% power were OR 1.28 for HFpEF and OR 1.22 for HFrEF. These null findings should be interpreted as inconclusive given current power constraints; larger ancestry-matched studies are needed.

Background with objective: Detecting EGFR mutations is critical for treating lung adenocarcinoma with highly effective targeted therapies. However, standard genetic testing is expensive, complex, and often unavailable in resource-limited settings like Bangladesh. Because elevated serum CEA has been linked to these genetic alterations, it could serve as an accessible screening tool. This study aims to evaluate the association between serum CEA levels and EGFR mutation status to determine if routine CEA testing can reliably predict these mutations and guide treatment. Methodology: In this cross-sectional analytical study, we recruited 58 patients with histologically confirmed treatment naive lung adenocarcinoma. The presence of EGFR mutations in the ctDNA was determined via ARMS (Amplification Refractory Mutation System) PCR. Patient data was statistically analyzed to assess the diagnostic correlation between serum CEA levels and the presence of EGFR mutations. Result: The overall EGFR mutation rate was 43.1% with exon 19 deletion (48%) and exon 21 mutations (44%) were the predominant types. Median serum CEA levels were significantly higher in patients with EGFR mutations compared to wild-type cases (14.6 ng/ml vs 2.8 ng/ml, p<0.001). A multivariate analysis revealed a 14% increased likelihood of an EGFR mutation for 1 ng/ml rise in serum CEA. Furthermore, serum CEA showed strong diagnostic accuracy for ctDNA samples at a 6.39 ng/ml cut-off (AUC 0.82, sensitivity 68.0%, specificity 84.8%). Conclusion: Serum CEA is a valuable, cost-effective, and non-invasive biomarker demonstrating significantly higher levels and strong diagnostic accuracy in EGFR-mutated lung adenocarcinoma compared to wild-type cases.

Background: Early-onset colorectal cancer (EOCRC) is increasing globally, particularly among Hispanic/Latino (H/L) populations, yet the contribution of tumor-colonizing microbiota to age-associated colorectal cancer (CRC) biology remains poorly understood. Most microbiome studies have focused on fecal communities or non-Hispanic populations, leaving the intratumoral microbial landscape of H/L patients largely unexplored. Methods: We performed an exploratory characterization of tumor-colonizing microbiota using whole-exome sequencing (WES) data from four primary colorectal tumors obtained from H/L patients treated at City of Hope, including two EOCRC (<50 years) and two late-onset colorectal cancer (LOCRC; [&ge;]50 years) cases. Following removal of host-derived sequences, microbial taxonomic profiling was conducted at the family, genus, and species levels, and microbial metabolic pathways were inferred. Clinical and pathological data were integrated to evaluate age-associated differences in microbial composition and predicted function. Results: Family-, genus-, and species-level analyses consistently demonstrated greater microbial diversity in LOCRC than EOCRC. LOCRC contained more than twice the number of unique bacterial families, nearly three times as many unique genera, and more than twice as many unique bacterial species. A conserved core microbiota, including Fusobacteriaceae, Prevotellaceae, Fusobacterium, and Prevotella, was identified across both age groups, whereas LOCRC was enriched in CRC-associated taxa including Fusobacterium nucleatum, Bacteroides fragilis, Parvimonas micra, Porphyromonas asaccharolytica, and Dialister pneumosintes. Species-level analyses revealed only a single shared bacterial species between EOCRC and LOCRC, indicating progressive microbial divergence with increasing taxonomic resolution. In contrast, functional profiling identified 11 predicted microbial metabolic pathways, of which nine were shared between age groups, two were unique to EOCRC, and none were exclusive to LOCRC. Core metabolic pathways involved in energy metabolism, amino acid biosynthesis, phospholipid metabolism, and central carbon metabolism exhibited comparable abundance across both groups, demonstrating substantial functional conservation despite pronounced taxonomic differences. Conclusions: Tumor-colonizing microbiota differ markedly between EOCRC and LOCRC in H/L patients, with late-onset tumors exhibiting substantially greater microbial richness and taxonomic complexity. Despite these compositional differences, microbial metabolic functions remain largely conserved, supporting the concept of functional redundancy within the colorectal tumor microenvironment (TME). Although exploratory, this proof-of-concept study provides one of the first characterizations of intratumoral microbiota in H/L EOCRC and establishes a foundation for larger multi-omics investigations aimed at identifying microbiome-based biomarkers and therapeutic targets for precision oncology.

The Cancer Genome Atlas (TCGA) holds clinical data for over 11,000 patients across 33 cancer types, but access is hard because of complex file structures, heterogeneous formats, and the need for programming. We present an agentic system for natural language querying and statistical analysis of TCGA clinical data. The system uses a large language model as an autonomous ReAct agent that selects from eight computational tools, including data extraction, descriptive statistics, Kaplan-Meier survival analysis with log-rank tests, hypothesis testing, and verification against the curated TCGA Pan-Cancer Clinical Data Resource (CDR). The agent reasons about intermediate results, adapts its approach, and returns clinically contextualized responses with source attribution and auditable traces. We introduce TCGA-Agent-Bench, 440 queries across five difficulty tiers with ground truth from the independently curated TCGA-CDR, evaluated with dual metrics of numerical accuracy and clinical completeness. The system achieves 93.4% overall accuracy (100% single-patient lookups, 99.1% cohort statistics, 92.8% comparative analyses), outperforming a fixed rule-based pipeline (87.1%), a single-pass LLM (81.8%), and retrieval-augmented generation (66.9% on a subset). Most of the benchmark is answerable from the CDR alone, so we locate the extraction layer's value in fields the CDR lacks (drug treatments, TNM components, biomarkers, biospecimen metadata): on 26 queries targeting these, the full system answers 100% versus 3.8% for CDR-only. Ablations show the reasoning loop is most impactful (+9.1% accuracy, +22.0 completeness points). A tool-based agentic architecture enables accurate, auditable analysis of clinical repositories, with value driven by tool design and recovered fields rather than model scale.

We present FootNet, a 453-image multi-view smartphone foot dataset for binary foot segmentation, with expertannotated masks across six anatomical views (dorsal, medial, and plantar, both left and right). We benchmark four segmentation models under a controlled protocol: U-Net with a MobileNetV2 encoder achieves the best performance (IoU 0.9268, Dice 0.9608, 95 % CI [0.9209, 0.9320]); DeepLabV3 with MobileNetV3-Large scores IoU 0.8984 (Dice 0.9449); UNet++ with MobileNetV2 scores IoU 0.8913 (Dice 0.9391); and SAM ViT-B with oracle boundingbox prompt scores IoU 0.9219 on the matched 191-image subset. Bonferroni-corrected Wilcoxon signed-rank tests (k = 6 comparisons) show U-Net significantly outperforms DeepLab (p < 0.001, r = 0.638) and SAM ViT-B with oracle boundingbox (p = 0.005, r = 0.202); UNet++ does not significantly differ from DeepLab (p = 0.062). Connected-component postprocessing yields negligible benefit (mean {triangleup}IoU = +0.0003, 12 of 453 images improved). The extended dataset is available upon request

Automated interpretation of neuroimaging studies requires simultaneous assessment of multiple imaging evidence variables, each tied to distinct anatomical structures. Vision-language models (VLMs) offer a unified framework for multi-task analysis, but adapting pre-trained VLMs remains challenging. Full fine-tuning is computationally prohibitive, and joint multi-task training requires simultaneous access to all task data, which is often infeasible in clinical settings. Although model merging enables multi-task composition without joint re-training, existing methods focus on post-hoc algorithms with limited extension to VLMs and minimal application to neuroimaging. Here, we present GRadient-guided Adapter Merging (GRAM), a layer-selective low-rank adaptation (LoRA)-based fine-tuning and merging framework for multi-task neuroimaging visual question-answering (VQA). GRAM uses a gradient ratio that contrasts class-specific gradients to identify task-discriminative layers, and applies subspace-constrained projected gradient descent to restrict LoRA updates to directions consistent with the geometry of the pre-trained model. We leveraged a structured VQA benchmark, developed from the National Alzheimer's Coordinating Center (NACC) dataset, that pairs multi-sequence brain MRI studies with question-answer pairs across clinically relevant imaging evidence variables. Experiments on the VQA benchmark showed that GRAM outperformed or matched all-layer LoRA fine-tuning and a standard merging baseline while reducing inter-task interference during merging, and approached or surpassed the performance of joint multi-task training without joint re-training.

Background: In Japan, acute inpatient care is divided into approximately 335 secondary medical care areas, which serve as the basic units for planning healthcare delivery systems under the 8th National Health Care Plan. While comparisons between regions and facilities typically rely on a single risk-adjusted metric, this approach confuses differences in patient demographics with differences in the actual infrastructure of intensive care units (ICUs). This paper presents a framework - MedZone Embedder - for deriving data-driven indicators of regional structural vulnerability by mapping secondary medical care areas onto a learned similarity space, together with its working implementation. The paper sets out the concept, the method, a proof of concept, and an explicit staged validation program, rather than national empirical results. Methods: Each area is represented by a feature vector consisting of aggregated values of intensive care provision indicators derived directly from the Japan Intensive Care Patient Database (JIPAD) - specifically, risk-adjusted mortality rates (standardized mortality ratios and an in-hospital composite indicator), technical efficiency, length of stay, readmission rates, case severity, and case composition - with the within-area variance of these indicators also taken into account. No hierarchical processing by facility type is performed. A contrastive autoencoder (multilayer perceptron encoder 32 -> 16 -> 8, symmetric decoder) is trained by self-supervised learning, using an objective function that combines reconstruction and normalized temperature cross-entropy (NT-Xent) on noise-augmented views. The resulting 8-dimensional embedding supports area searches based on cosine similarity and anomaly scoring in the embedding space (using isolation forest, Mahalanobis distance, or k-nearest-neighbor density), which is normalized to a vulnerability score ranging from 0 to 1. If deep learning libraries are unavailable, or if the number of areas is small, an alternative method using deterministic principal component analysis is employed. Results: This method was implemented and deployed within an operational ICU decision support system on a managed cloud platform. The proof of concept (PoC) is structured around five secondary medical care areas within Kyoto Prefecture and runs entirely on synthetic facility-level aggregate data constructed to follow the JIPAD indicator schema; no registry data were accessed. It generated: an aggregate provision profile for each area; an area embedding space equipped with a similar-area search function; and a vulnerability ranking that identifies areas with low patient numbers and low diversity that exhibit overall poor outcomes. At this scale, the contrastive autoencoder falls back to principal component projection. The deep learning pathway has been implemented and unit testing has been completed; training and evaluation on actual registry data are pending data-use approval and the expansion of data integration. Validation is staged: Stage 2 will train the contrastive pathway over JIPAD-covered areas to assess construct validity against public structural indicators (ICU/HCU beds, population, accessibility), and Stage 3 will extend coverage to all areas via National Database (NDB) linkage. Conclusion: MedZone Embedder reframes regional comparison from single-indicator ranking to structural representation: which areas are alike, and which are structural outliers. The contribution of this paper is the framework - the proposal that the intensive care provision structure of Japanese secondary medical care areas can be learned from a national outcomes registry and read through the lens of what we call institutional debt - together with a deployed implementation and a pre-specified validation program. To our knowledge, this is a candidate first application of contrastive representation learning to Japanese secondary medical care areas.

Patients and caregivers seek informational and emotional support throughout medical care, especially when interpreting unfamiliar laboratory test results. Although resources such as patient portals and online health communities (OHCs) help address questions, gaps remain. The emergence of large language models (LLMs) offers the potential to be a complementary source of support to assist patients and caregivers in understanding and using their test results. The objective of our study is to empirically compare LLM responses to patients online questions containing their laboratory test results to responses written by peers in an OHC. We compared the 519 peer replies to 122 laboratory test-related posts from an OHC to 488 responses generated from four LLMs using mixed computational and qualitative methods. LLMs frequently provided clear explanations of medical terminology and structured interpretations of numeric results but were longer and less readable. Peers offered more personalized, context-specific emotional support. Overall, LLMs have the potential to complement peer responses in OHCs, but require greater emotional depth, reasoning transparency, and alignment with community norms.

Objective To develop and validate a risk model for predicting postoperative bleeding in patients with thyroid cancer. Methods A total of 2800 consecutive patients diagnosed with thyroid cancer in the Department of Thyroid and Breast Surgery of the Affiliated Hospital of Xuzhou Medical University between January 2020 and December 2023 were retrospectively analyzed. Patients were categorized into two groups based on postoperative bleeding occurrence: bleeding and non-bleeding groups. Univariate and multivariate logistic regression analyses were utilized to screen independent risk factors. Meanwhile, risk prediction models were developed and nomogram . Subgroup analysis was performed to identify independent risk factors. The predictive effects of the models were assessed using the Hosmer-Lemeshow test and receiver operating characteristic (ROC) curves. Results Of the 2800 recruited patients, 50 had postoperative bleeding, with an incidence rate of 1.7%. Multivariate logistic regression analysis showed that age, hypertension, total thyroidectomy, tumor size [&ge;]4 cm, and operation time [&ge;]90 min were the risk factors for postoperative bleeding in thyroid cancer patients (P<0.05). A risk prediction model was established based on the above factors, and the area under the ROC curve was 0.881, with a sensitivity of 94.0%, a specificity of 67.3%, and an accuracy of 74.0%. Decision curve analysis revealed that the model had good predictive ability. Conclusions The constructed risk prediction model has good predictive power and can provide a reference for healthcare professionals to predict the risk of bleeding in patients after thyroid cancer surgery.

Importance Pharmacogenomic (PGx) guidelines can improve medication efficacy and reduce toxicity, but their application in pediatric intensive care units (PICUs) remains largely unexplored. Objective To determine the frequency of medications with established PGx guidelines administered in the PICU and assess the capacity of exome sequencing to capture PGx phenotypes for these medications. Design Retrospective cohort study integrating electronic medical record and exome sequencing data. Setting Morgan Stanley Children's Hospital of NewYork-Presbyterian, a single center tertiary care children's hospital. Participants A total of 4,939 children admitted to the PICU (2020 - 2024), and 192 children admitted to the PICU who underwent exome sequencing for research purposes (2015 - 2023). Exposure Critical illness requiring PICU admission. Main Outcomes and Measures Frequencies of administration of medications with established PGx guidelines in the PICU and the proportion of individuals with exome sequencing with identifiable PGx phenotypes. Results Among 4,939 PICU patients, 37.2% (n=1,837) received at least one medication with established PGx guidelines and 14.4% (n=712) received two or more such medications. Twenty PGx genes were implicated; CYP2C9 was most common (17.3%, n=853). An estimated 8.2% of patients received medications for which PGx-guided recommendations would have altered clinical management. Among 192 patients who underwent exome sequencing, at least one metabolizer phenotype was identified in 62% (n=119). Conclusions and Relevance Many critically ill children receive medications with established PGx guidelines. This study highlights an opportunity for more personalized medicine for critically ill children admitted to a tertiary care hospital and assesses the strengths and weaknesses of exome sequencing to uncover pertinent PGx phenotypes.

Background: Cases requiring 13 or more tissue sections in Mohs micrographic surgery (MMS) demand extended operative time, additional resources, and often specialised closure techniques. Pre-operative identification of such cases would improve surgical scheduling, resource allocation, and patient counselling. We aimed to develop and validate a machine learning prediction tool using pre-operative clinical features to identify cases likely to require13 sections. Objectives: To develop and validate machine learning models for predicting which Mohs procedures will require 13 sections, using pre-operative clinical features, and to identify key predictive factors. Methods: We analysed 408 consecutive Mohs procedures with 16 pre-operative clinical variables. Thirty machine learning algorithms were evaluated, including ensemble methods (Stacking, Voting), gradient boosting (XGBoost, LightGBM, CatBoost), neural networks (3-7 layers), support vector machines, and traditional classifiers. Model performance was assessed using 5-fold stratified cross-validation and independent test set evaluation. Feature importance was determined using SHAP (SHapley Additive exPlanations) analysis. Results: The stacking ensemble achieved the highest cross-validation AUC of 0.891 (95% CI: 0.849-0.934) and test AUC of 0.884. Tumour area (cm2), calculated using the ellipse formula to approximate clinical tumour morphology, emerged as the strongest predictor (SHAP importance: 0.141), followed by tumour size dimensions (0.086 and 0.068), aggressive histopathology (0.046), and recurrence status (0.035). Wide neural network architectures (5-layer) outperformed deeper configurations (7-layer). The model demonstrated 70.7% high-confidence predictions with uncertainty <15%. Conclusions: Machine learning models using pre-operative clinical features can accurately predict which Mohs procedures will require 13 or more sections. The stacking ensemble approach provides robust predictions suitable for clinical decision support. External validation in multi-centre cohorts with diverse patient populations and practice patterns is warranted to assess model generalisability.