Citations

Background: Reliable access to postoperative guidance between clinic visits remains inconsistent, and the usefulness of large language models depends on how clinical knowledge is structured and retrieved. Knowledge graph-only (KG-only), vector retrieval-augmented generation (RAG-only), and hybrid graph-retrieval (GraphRAG) architectures have not been directly compared under controlled conditions for postoperative care. Objective: It was to compare these three pipelines on a curated postoperative question-answering benchmark using automated grounding metrics and blinded clinician review. Methods: A benchmark of 250 expert-authored questions (200 in-scope, 20 emergency-intent, 30 out-of-scope) was evaluated on an identical knowledge corpus using Gemini 2.5 Flash for generation and Gemini Embedding-001 for retrieval. Three blinded clinicians rated all 750 outputs; the primary outcome was majority-consensus accuracy across all 250 questions. GraphRAG is a multi-component pipeline that also includes query expansion, a second retrieval stage, evidence fusion, and a monotonic RAG fallback, and is therefore evaluated as a complete hybrid system rather than as graph retrieval in isolation. Results: Overall accuracy was 96.8% (95% CI: 93.8–98.4) for GraphRAG, 94.8% (95% CI: 91.3–96.9) for RAG-only, and 56.0% (95% CI: 49.8–62.0) for KG-only. Both retrieval-augmented systems outperformed KG-only (adjusted p < 0.001). The GraphRAG versus RAG-only difference was 2.0 percentage points, arose from five discordant question pairs and did not reach the Bonferroni-adjusted threshold (exact McNemar p = 0.062; adjusted p = 0.188). The monotonic RAG fallback determined 44.8% of GraphRAG outputs (112 of 250 queries). GraphRAG showed higher context recall (0.851 versus 0.796; adjusted p < 0.001), whereas RAG-only showed higher context precision. Estimated hallucination risk, derived from an LLM-as-a-judge procedure using Gemini 2.5 Flash rather than independent clinical adjudication, was 0.108, 0.132, and 0.436, respectively; the GraphRAG versus RAG-only difference was not significant (adjusted p = 0.771). Inter-rater agreement was high (Fleiss’ kappa = 0.854). Conclusions: Retrieval augmentation, rather than the choice of retrieval architecture, accounted for most of the difference observed. The incremental advantage of GraphRAG over a well-implemented RAG-only system was small, was not statistically significant after correction and reflects the complete hybrid system rather than graph augmentation alone. Emergency-intent accuracy was 75.0% for both retrieval-augmented systems; the architecture is therefore suited to supervised clinical workflows with clinician review, not autonomous patient-facing triage. Prospective validation on authentic patient messages and formal component ablations are required.

Introduction: Frontotemporal lobar degeneration (FTLD) is a devastating disease that commonly results in early onset dementia, yet its rarity and heterogeneity limit large-scale research and clinical trials. Remote monitoring via low-burden digital health tools may overcome these barriers. We evaluated clinical utility of passive smartphone monitoring in FTLD using a multi-domain mobile assessment platform. Methods: Participants were 567 adults (53% clinically normal, 21% prodromal FTLD, 26% symptomatic FTLD) who completed an in-person study visit and downloaded the ALLFTD Mobile App on their personal smartphones. The app delivered unsupervised cognitive tests and passively collected continuous data on battery percentage (proxy for smartphone use) and step count. Longitudinal follow-up included smartphone monitoring and annual in-person study visits. Primary analyses examined associations between passive smartphone features and markers of disease severity at baseline and longitudinally, and tested whether passive features added incremental value beyond app-based cognitive testing. Results: Passive smartphone features were feasible to collect and showed excellent reliability with <2 weeks of monitoring (ICCs>0.9). Features capturing smartphone use and movement demonstrated sensitivity to gold-standard clinical measures of disease severity both at baseline and longitudinally. A classification model combining passive features, the app-based cognitive composite score, and demographics detected longitudinal functional decline (AUC=0.89); passive features alone (AUC=0.84) performed comparably to a cognitive screener (AUC=0.85). Discussion: Passive smartphone monitoring is a valid, zero-burden measure of disease severity and progression in FTLD that adds modest incremental value beyond app-based cognitive testing. Findings support its use as a scalable digital endpoint in longitudinal FTLD research.

Background Cardiac rehabilitation (CR) improves peak oxygen uptake (VO 2 peak) and body composition in patients with cardiovascular disease. However, improvement in VO 2 peak is not observed in all patients following a CR program, and whether VO 2 peak non-responders also demonstrate limited changes in body composition remains unclear. This study aimed to examine the association between changes in VO 2 peak and body composition in patients undergoing a CR program. Methods Retrospective analysis of patients who completed CR between 2015 and 2022 at Mayo Clinic. The study included 143 patients who completed cardiopulmonary exercise testing and dual-energy X-ray absorptiometry pre- and post-CR. Patients were grouped by the change in relative VO 2 peak response to CR (responders= Δ&gt;0 vs non-responders=Δ ≤0 mL/kg/min). Two-way repeated ANOVA was used to compare groups, and regression models were used to examine the associations between baseline body composition and changes in body composition with VO 2 peak change. Results Among the cohort, 26% were classified as non-responders based on relative VO 2 peak measurements. Both groups exhibited a significant improvement in most body composition variables including body fat mass, body fat percentage, body lean percentage, and fat mass index (all p&lt;0.001), with no differences in the percentage of change in body composition variables between groups (p&gt;0.05). Regression models demonstrated that neither baseline body composition nor changes in body composition were significantly associated with changes in VO 2 peak. Conclusion Our data suggest that baseline and change in body composition were not significantly associated with changes in VO 2 peak in patients undergoing the CR program.

Abstract Purpose To evaluate whether two automated measures of nasality, Nasal Cognate Distinctiveness (NCD) and nasal posterior probability (NPP), provide clinically meaningful markers of hypernasality in speakers with motor speech disorders (MSDs) and to determine whether hypernasality-related information extends beyond stop consonants, which are typically targeted in cognate-based approaches. Method Sentence-level speech recordings were obtained from 561 patients, including 374 neurologically typical controls and 187 individuals with MSDs. NCD was computed for voiceless stops (/p/, /t/, /k/) using likelihood ratios between stops and their nasal cognates. NPP was extracted using a phonological feature-based neural network model and analyzed for stops, consonants, and vowels. Hypernasality severity was rated by speech-language pathologists. Group differences and associations with hypernasality severity were evaluated using linear mixed-effects models and Spearman correlations, and receiver operating characteristic (ROC) analyses were used to evaluate discrimination of absent versus present hypernasality. An additional 20 controls with repeat recordings were used to evaluate test-retest reliability and minimal detectable change (MDC). Results NCD generally decreased and NPP increased with increasing hypernasality severity, with phoneme-specific severity effects observed for NCD. Across stops, NCD and NPP were strongly inversely correlated and showed comparable discrimination of absent versus present hypernasality (AUCs = 0.796 and 0.730, respectively). Consonant and vowel NPP measures were both positively associated with hypernasality severity and demonstrated identical classification performance (AUC = 0.853). No significant differences in ROC performance were observed between NCD and NPP or between consonant and vowel NPP measures. Test-retest change exceeded the MDC threshold in 2 of 20 controls for NCD and 1 of 20 for NPP. Conclusions Both NCD and NPP provide objective markers of hypernasality in speakers with MSDs. The measures capture related aspects of nasalization and demonstrate comparable ability to identify clinically meaningful hypernasality. NPP captures hypernasality-related acoustic information across both consonants and vowels, with similar relationships to severity and classification performance across sound classes. Test-retest findings provide preliminary MDC benchmarks for interpreting longitudinal change.

Abstract Gallbladder (GB) cancer is an aggressive biliary malignancy with a silent clinical course, leading to delayed diagnosis and poor outcomes. Early detection and accurate staging are pivotal for guiding management, yet diagnosis remains challenging due to overlapping imaging features with benign pathologies. This review synthesizes the current state-of-the-art imaging modalities, including ultrasound (USG), multiphase computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET)/CT, and PET/MRI, highlighting their roles in diagnosis, staging, and treatment planning. A systematic USG-based triaging helps in management decision. Algorithmic approach to GB masses and incidental GB polyps is also covered with future directives on artificial intelligence and radiomics.

Abstract Pancreatic ductal adenocarcinoma (PDAC), the most common subtype of pancreatic cancer, is a deadly disease with a complex tumor microenvironment (TME). Cytotoxic combination chemotherapy treatments are the mainstay of PDAC therapy, but patients rapidly develop chemoresistance, highlighting the need to elucidate how chemotherapy alters the TME and whether these changes drive chemoresistance. Here, we examined matched pre- and post-treatment tissue specimens using single-cell RNA-sequencing and found near-universal enrichment after treatment of axonal guidance genes in cancer-associated fibroblasts (CAFs). These neural CAFs were enriched near sites of perineural invasion, coinciding with regions of increased tumor cell proliferation, and were enriched in tumor areas distant from nerves after chemotherapy. Metastatic recurrent lesions had the highest prevalence of neural CAFs versus primary tumors and untreated metastases. Neural CAFs showed elevated non-canonical WNT mediators and axonal guidance genes, which complemented matching cognate binding partners in tumor epithelial cells, suggesting a role in tumor-stroma crosstalk. Additionally, in vitro studies revealed enrichment of axonal guidance genes in patient-derived CAF lines upon exposure to chemotherapy or tumor cells. Altogether, these findings implicate fibroblast-derived axonal-guidance genes in promoting PDAC invasion and point to a promising therapeutic target for this disease.

Cancer immunotherapy is a transformative technology allowing for the elimination of malignant cells through immune system stimulation; however, the large-scale clinical application of adoptive cellular therapy technologies is limited due to difficulties in ex vivo immune cell engineering, high costs, and complicated manufacturing process. Recently, in vivo immune cell engineering was discovered as an emerging approach that allows for the genetic reprogramming of immune cells in patients, thus circumventing many problems inherent to traditional cellular therapy technologies. Breakthroughs in targeted drug delivery systems such as cell-penetrating peptides, antibody- and receptor-mediated delivery approaches, lipid nanoparticles (LNPs), polymeric drug carriers, and biomimetic delivery systems allow for the more effective selective delivery of messenger RNA (mRNA), clustered regularly interspaced short palindromic repeats (CRISPR)/Cas gene editing systems, proteins, and other types of cargo to the target immune cells. In vivo immune cell engineering allows for the efficient generation of chimeric antigen receptor T-cells (CAR-T), CAR- natural killer (NK) cells, engineered macrophages, dendritic cells, and other immune cell types; moreover, transient and permanent immune reprogramming is achievable via vivo technologies. The present review discusses the biological barriers limiting immune cell-targeted delivery, including biodistribution, cellular uptake, intracellular transport, and endosomal escape of delivered drugs, and reviews the latest advancements in targeted delivery technologies addressing these obstacles. We also address the latest applications of in vivo immune cell engineering in hematological malignancies and solid tumors, including mRNA therapeutics, CRISPR-based genome editing, and advanced programmable immunotherapy technologies. Furthermore, we discuss unresolved issues related to the specificity of targeting, efficiency of drug delivery, safety of treatment, manufacturing processes, and regulation of in vivo immune engineering technologies and outline future opportunities arising from the integration of nanomedicine, synthetic biology, genome engineering, and artificial intelligence-assisted materials science.