Frontiers in Computational Neuroscience

The -Bonacci weighted sequence space {ell}(,) is introduced for [&ge;] 2, 1 [&le;] < {infty}, and > 0). It is proved to be a Banach space which is isometrically isomorphic to {ell}, with an unconditional Schauder basis. The space is also reflexive and uniformly convex. Its strict inclusion chain, continuous dual, and asymptotic truncation estimate of order - / are also established. Based on this framework, Scaled &#119898-BonacciNet is proposed as a lightweight CNN that rescales Fibonacci-Net channels using the dominant &#119898-Bonacci root and employs depth wise separable convolutions. Tribonacci-Net achieved 96.09% accuracy and a 0.9581 macro F1-score on the 15-class Brain MRI dataset using 0.85 million parameters and 0.366 GFLOPs. Pentanacci-Net attained 95.87% accuracy on the 44-class dataset while reducing parameters and FLOPs by 47.10% and 44.48%, respectively. On the CKD ultrasound dataset, Hexanacci-Net achieved 83.75% accuracy and a 0.7941 macro F1-score, with corresponding reductions of 50.37% and 45.10%. These findings support dominant-root scaling as a structured approach to lightweight CNN design.

Objectives: To determine how strongly clinical referral text, rather than image content, drives the decisions of multimodal large language models (LLMs) differentiating glioma from brain abscess on MRI. Materials and methods: This retrospective single-center study included 60 contrast-enhanced T1-weighted brain MRI examinations (30 gliomas, 30 abscesses; reference-standard diagnosis). One axial slice per case was paired with five referral texts (explicitly inflammatory to explicitly tumor-suggestive) under three prompt instructions and submitted 20 times each to GPT-4o, Claude Sonnet 5, Gemini 3.5 Flash, and Qwen 3.6 (72,000 answers). The primary outcome was the text-driven span (difference in tumor-call frequency between explicit inflammatory and explicit tumor text). Two blinded radiologists classified the images without text; sensitivity, specificity, and accuracy were compared descriptively. Results: Patients (29 women) had a mean age of 55.8 +/- 16.4 years. With neutral text, models answered tumor in 45-67% of queries (specificity, 0.38-0.78), versus 52% and 60% for radiologists (specificity, 0.77 and 0.63). The text-driven span was 0.79-1.00 (p < 0.001 for all models), whereas confirmed gliomas and abscesses differed by only 0.01-0.10 in three models and 0.25 in Gemini 3.5 Flash. Subtle wording produced at least 83% of the span in three models. An instruction to ignore the text reduced the span to 0.48-0.70 in three models and to 0.04 in Gemini 3.5 Flash, partly through reversion to a tumor default. Model accuracy was 0.52-0.73 versus 0.78 and 0.73 for radiologists. Conclusion: Multimodal LLMs deferred to referral text over imaging evidence in differentiating glioma from brain abscess, and prompt instructions did not reliably correct this.

Computed tomography (CT) interpretation requires recognition of the anatomical level of each slice. Classification may facilitate preprocessing and AI-assisted fracture diagnosis. We developed deep learning models to classify craniofacial CT slices into six anatomical categories. A total of 11,769 images were collected from a multicenter database involving 10 Japanese university hospitals and the public Head-Neck-Radiomics-HN1 dataset. Patients were divided into training (70%) and evaluation (30%) sets. Five CNN architectures were trained from scratch, and hard- and soft-voting ensembles were evaluated. Individual-model accuracy ranged from 94.72% to 96.09%; the hard-voting ensemble achieved 96.46% accuracy (Cohen's k = 0.954). Although performance differed among individual models (Cochran's Q = 99.29, p < 0.001), the ensemble did not significantly outperform EfficientNetV2B1 (McNemar's p = 0.163). Errors occurred mainly between adjacent anatomical regions. Deep learning can classify craniofacial CT slices accurately and may support automated dataset organization and AI-assisted fracture diagnosis. External validation is needed.

Longitudinal clinical cohorts contain heterogeneous progression patterns that are obscured by coarse diagnostic labels. Discovering these latent progression phenotypes requires models that distinguish patients not only by their outcomes, but also by how their clinical trajectories evolve. This paper proposes TRACE, an interpretable temporal phenotyping method that enables the learning of outcome-informed progression phenotypes from longitudinal trajectories. TRACE combines a trajectory- and outcome-aware contrastive objective with a probabilistic predictive phenotype layer, encouraging phenotypes to capture progression structure while remaining predictive of diagnostic state. A dual-attention mechanism further supports interpretation by identifying the visits and clinical variables most associated with each phenotype assignment. TRACE is evaluated on the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort for cognitive disease modeling, showing improved performance in both diagnostic prediction and phenotype discovery compared with other temporal phenotyping baselines. Finally, by allowing the number of phenotypes to exceed the diagnostic labels, TRACE enables discovering additional progression patterns that are not captured by labels defined a priori.

Purpose: To evaluate whether visually assessed high-quality prostate MRI, as defined by the Prostate Imaging Quality (PI-QUAL) scoring system, is sufficient to ensure consistent AI model performance, or whether hidden acquisition-related technical effects introduce variability that cannot be detected through visual inspection. Methods: This retrospective study included 568 bpMRI scans from seven institutions, classified using PI-QUAL v2 into high-quality (scores 2-3) and low-quality (score 1) groups. Quantitative image quality features were extracted using MRQy from T2w and ADC sequences. Three experiments were conducted on high-quality scans: (1) unsupervised clustering to identify hidden technical batch effects; (2) V-Net models (M) for prostate capsule segmentation on T2w using single-cluster and balanced multi-cluster training strategies; and (3) VoxelMorph-based deformable registration (M2) to align ADC maps to T2w images. Performance metrics included DSC, precision, recall, volume ratio, HD95, ASSD, and TRE. Kruskal-Wallis tests assessed cross-cluster variability, Two One-Sided Tests (TOST) evaluated performance equivalence, and bootstrap resampling quantified training efficiency. Results: Clustering identified four clusters among 277 high-quality bpMRI scans. Single-cluster models showed variable performance (DSC=0.41-0.85) with significant batch effects (p<0.001), even with 140 training samples. The balanced multi-cluster model (n=60) eliminated batch effects (p=0.075), achieved equivalent performance (p<0.001), and demonstrated 2.23-fold greater training efficiency despite using 2.3-fold fewer samples. Registration performance also varied significantly across clusters (TRE=0.86-5.47 mm, p<0.001). Conclusion: Acquisition-related batch effects significantly impact AI generalizability in prostate MRI, even among high-quality scans. Quantitative clustering enables detection of technical variations and supports diversity-aware training strategies that achieve equivalent performance with 2.3-fold fewer samples while eliminating batch effects.

Introduction: This study aimed to characterize the neurochemical profile of Alzheimer s disease (AD) continuum stages using proton magnetic resonance spectroscopy (1H-MRS) data for brain metabolites and macromolecules. We used Orthogonal Projection to Latent Structures Discriminant Analysis (OPLS-DA) to classify diagnostic groups and assess feature importance. Methods: Our dataset consisted of 7 T short-echo time 1H-MRS spectra of the posterior cingulate cortex/precuneus of individuals diagnosed with AD (n = 27) or Mild Cognitive Impairment (MCI, n = 27) and Healthy Controls (HC, n = 31). Classification used 100-round Monte Carlo cross-validation based on metabolite concentrations and macromolecular peak intensities. Variable importance in projection (VIP) was assessed across classification models to identify the most relevant neurochemical features. The resulting metabolite rankings were compared with previous literature to evaluate their biological plausibility. Results: Classification was strongest for AD versus HC (AUC=0.88) and lower for comparisons involving MCI. Metabolites contributing most strongly to classification included myo-inositol, ascorbate, and choline-containing compounds, consistent with previous studies on AD-related neurochemical alterations. In addition, specific macromolecular components were identified as relevant features, suggesting a potential contribution of MM signals to the characterization of AD-related neurochemistry. Discussion: These findings support the use of short-echo-time 1H-MRS as a complementary tool for characterizing neurochemical changes in the AD continuum, with potential applications in disease characterization and monitoring, warranting validation in larger cohorts.

Background: Ultra-low field (ULF) magnetic resonance imaging (MRI) offers accessible point-of-care neuroimaging, yet its capability to support multivariate network analyses remains understudied. Methods: We systematically evaluated the feasibility and reliability of Morphometric Similarity Networks (MSN) and Morphometric Inverse Divergence (MIND) networks derived from paired high-field (HF, 3T/1.5T) and ULF (64mT) scans in a primary cohort(N=23) and an independent replication cohort (N=19). We tested the impact of super-resolution processing, cortical parcellation granularity (Desikan-Killiany vs. Destrieux), and mesh resolution (fsnative vs. fsaverage5) on edge-level cross-scanner agreement (Lin's concordance correlation coefficient, CCC) and subject-level identifiability. Results: MIND networks demonstrated higher edge-level HF-ULF agreement (mean CCC = 0.95) and ULF reproducibility (mean CCC = 0.98) compared to MSNs (mean CCC = 0.86-0.87 and 0.91-0.94, respectively). In contrast, MSNs yielded higher subject-level differential identifiability across field strengths due to a broader distinction between within- and between-subject similarity. Reconstructing MIND networks on native high-resolution meshes (fsnative) maximised cross-scanner agreement, whereas coarser meshes and finer cortical parcellations enhanced individual network identifiability. Conclusions: Structural similarity networks can be reliably constructed from ULF MRI data. Optimal analytic choices depend on study objectives: MIND on native meshes maximises cross-scanner stability, whereas MSNs excel at preserving individual uniqueness, paving the way for globally scalable connectomic research.

Purpose: Posterior pharyngeal wall (PPW) thickening is a common structural manifestation of treatment-related internal lymphedema in patients with head and neck cancer and can be evaluated using lateral videofluoroscopic images acquired during modified barium swallow (MBS) studies. However, no standardized method currently exists for quantitative measurement. Methods: This two-phase methodological study first conducted a scoping review to characterize existing approaches to PPW thickness measurement and identify the principal factors contributing to variation in PPW thickness measurement. Eight studies were identified with variation observed across these measurement factors. An empirical reliability study was then performed using 107 MBS studies from 93 Head and Neck Cancer patients. Two independent raters annotated 10 anatomical landmarks on pre-swallow and post-swallow resting frames, and a computer-assisted algorithm quantitatively generated seven candidate PPW thickness measurements based on different vertebral reference points and measurement directions. Reliability was assessed intraclass correlation coefficients (ICCs). Results: Inter-rater reliability was highest for the measurement originating from the midpoint of C3 and extending perpendicular to the anterior C3 surface (ICC, 0.78; 95% CI, 0.72-0.83). Among the normalization references, C2-C4 vertebral length demonstrated greater reliability than C3 vertebral length, with an ICC of 0.88 (95% CI, 0.85-0.91). No significant differences were observed between pre-swallow and post-swallow PPW thickness measurements across the evaluated approaches. Conclusions: These findings provide a methodological foundation for more standardized and reliable quantification of PPW thickness, with potential to expand the role of MBS from the functional evaluation of swallowing to include quantitative characterization of pharyngeal structural abnormalities.

Purpose To evaluate the external transportability of source-derived probability calibration for chest radiograph deep learning classifiers under distribution shift and determine its dependence on sample size and disease prevalence. Materials and Methods This retrospective secondary computational analysis of public, de-identified radiographic datasets involved no participant contact or new data collection. Cohort sizes were prespecified by availability without prospective power calculation: 106,686 NIH ChestX-ray14 images (29,323 patients; 1992-2015) for development and internal testing, 20,000 CheXpert images (20,000 patients; October 2002-July 2017), and 14,998 VinBigData images (2018-2020) for external testing. ResNet-18, DenseNet-121, and DeiT-Tiny classifiers for cardiomegaly and pleural effusion were trained across four patient fractions and three seeds (36 fits). Calibrators were fitted exclusively on NIH data. Twelve primary external Brier-score contrasts (temperature-scaled versus raw probabilities) were evaluated using paired cluster-bootstrapping with Holm adjustment. Results Temperature scaling reduced Brier scores in nine of 12 primary contrasts but worsened pleural effusion calibration on VinBigData across all architectures (Brier increases: ResNet-18, 0.001460 [95% CI: 0.001347, 0.001568]; DenseNet-121, 0.001193 [95% CI: 0.001109, 0.001274]; DeiT-Tiny, 0.000498 [95% CI: 0.000417, 0.000577]; all adjusted P=.006). Standardizing class weights to NIH prevalence reversed all CheXpert improvements. Although full versus 10% training improved external AUROC across all comparisons, DeiT-Tiny CheXpert pleural effusion Brier score worsened by 0.007039 despite a 0.032466 AUROC gain. At 80% selective image retention, calibrated ensembles retained only 12.3%-17.2% of VinBigData positive pleural effusion examinations. Conclusion Source-fitted calibration is not uniformly transportable across health systems, and discrimination gains do not guarantee calibrated probabilities. External evaluation must simultaneously assess probability reliability, prevalence shifts, and retained-case composition.

Purpose: Proton density (PD) and macromolecular proton fraction (MPF) are fundamental MR parameters characterizing water and macromolecular content in tissues. Both parameters can be mapped using a time-efficient MRI protocol, though PD maps require additional corrections for the reception profile (RP) non-uniformity and global scaling. Existing PD mapping methods suffer from uncertainties related to PD normalization to reference tissue. This study aimed to develop a data-driven RP correction method for PD mapping based on MPF as an internal scale. Theory and Methods: One-parameter model linking PD and MPF was derived under assumption of proportionality between total macromolecular volume and proton concentration. This model with a uniform parameter value across brain tissues generates a surrogate PD map, which is used to estimate RP from an uncorrected PD map. The proposed algorithm was tested against a common PD mapping method utilizing bias-field correction and normalization to cerebrospinal fluid. Brain MRI data were acquired at 3T from seven healthy individuals. The protocol included 3D multi-echo T1-, PD-, and magnetization-transfer-weighted gradient-echo sequences. Parametric maps were reconstructed using the single-point MPF mapping algorithm with data-driven B1 correction. Results: MPF-based and bias-field corrections yielded similar PD measurements in white and gray matter with no significant bias and remarkably small variability (coefficients of variation <0.6%). Conclusions: MPF-based RP correction enables computationally simple and robust quantitative PD mapping at no additional cost if MPF maps are available. The main advantage of the proposed method is automatic elimination of global scaling that obviates the need for PD normalization.

Objectives To evaluate discrepancies between radiologists and referring physicians in the interpretation of diagnostic probabilities conveyed by common Japanese expressions used in radiology reports. Materials and Methods Fifteen expressions commonly used to convey diagnostic probabilities in radiology reports were selected by five attending radiologists. An anonymous online survey was distributed to 26 radiologists and 33 referring physicians at a single institution. Participants assigned a diagnostic probability (from 0% to 100% in 10% increments) to each expression. Differences in median probabilities between radiologists and referring physicians were analyzed using the Mann-Whitney U test. Variability in interpretation was assessed using interquartile range (IQR), with IQR [&ge;] 30% defined as substantial variability. Results All radiologists (26/26) and 97% of referring physicians (32/33) completed the survey. Seven of the 15 expressions showed statistically significant differences in interpretation between radiologists and referring physicians: "c/w" (80% vs. 60%; p = 0.031), the Japanese equivalents of "present" (100% vs. 100%; p = 0.024), "suspect of" (80% vs. 70%; p = 0.007), "possibly" (50% vs. 70%; p < 0.001), "unlikely" (10% vs. 10%; p = 0.022), "cannot be ruled out" (25% vs. 40%; p < 0.001), and "not inconsistent with" (80% vs. 90%; p = 0.008). Substantial variability in interpretation (IQR [&ge;] 30%) was observed for four expressions: "cannot be ruled out" (IQR, 10%-40%), "s/o" (IQR, 60%-90%), "c/w" (IQR, 50%-90%), and "not inconsistent with" (IQR, 70%-100%). Conclusion This study revealed that nearly half of the Japanese expressions commonly used to convey diagnostic probabilities were interpreted differently by radiologists and referring physicians. These discrepancies highlight the need for a standardized lexicon to improve clarity and consistency in radiology reporting.

Background: Conventional performance metrics do not indicate whether artificial intelligence (AI)-platforms and radiologists are right or wrong on the same examinations, nor whether their non-agreement is more often beneficial or harmful. Objective: To develop, illustrate, and apply directional metrics quantifying non-agreement between radiologists and AI-platforms against a common reference standard, and to adapt a fragility-based framework for quantifying the statistical robustness of these metrics. Methods: We defined Positive Non-Agreement (PNA, AI-platform correct, radiologist incorrect), Negative Non-Agreement (NNA, radiologist correct, AI-platform incorrect), and Net Non-Agreement (NENA, their difference). Metrics were illustrated in three hypothetical scenarios of 1,000 examinations each and applied to three previously published AI-radiologist studies on pulmonary embolism and mammography. Bootstrap confidence intervals were calculated, and a reclassification fragility index (FI) quantified the minimum case reclassification needed to remove statistical support for NENA's direction. Results: In illustration, NENA was positive in two scenarios and negative in one. In the three published studies, NENA ranged from -4.18% to +10.74% and decreased consistently with increasing radiologist experience (junior residents, +6.27%; consultants, +1.45%), a gradient driven primarily by declining PNA with modest increase in NNA. The reclassification FI was defined for all three illustrative scenarios (range, 0.10%-1.70%) and for six of the seven comparisons from these studies (range, 0.34%-4.31%), but not for the weakest comparison, whose confidence interval for NENA included zero. Conclusions: PNA, NNA, and NENA provide a directional, case-level complement to conventional performance metrics for AI-radiologist non-agreement. The accompanying reclassification FI offers a statistically grounded, generalizable method for quantifying their robustness.

Background: Self-supervised learning may improve segmentation when labelled data are limited, but its benefit may depend on model and initialisation strategy. This study evaluated in-domain masked-reconstruction pretraining across limited annotation budgets. Methods: The International Skin Imaging Collaboration (ISIC) 2016 development set was partitioned into 765 images for self-supervised pretraining, 94 for fine-tuning and 41 for validation. Five nested annotation budgets (12-94 images) and five seeds were evaluated on the 379-image test set. A custom U-Net compared random and in-domain self-supervised initialisation; a ResNet-34 U-Net compared random, ImageNet, in-domain self-supervised and sequential ImageNet-to-self-supervised initialisation. Models trained with 94 labelled images were evaluated without adaptation on ISIC 2017. A separate experiment examined restoration-based pretraining. Results: In the custom U-Net, self-supervised pretraining achieved higher mean Dice than random initialisation at all five budgets, including 0.664 versus 0.615 with 12 labels and 0.811 versus 0.768 with 94 labels. Seed-level 95% confidence intervals for paired differences excluded zero at 47 and 94 labels but not at the other budgets. With 47 labels, the self-supervised model achieved mean Dice of 0.789, compared with 0.768 for random initialisation with 94 labels. In ResNet-34, ImageNet initialisation achieved higher mean Dice than in-domain self-supervised pretraining at all five budgets, and adding self-supervised pretraining after ImageNet reduced mean Dice at four of five budgets. All six 94-label configurations achieved lower mean Dice on ISIC 2017 than ISIC 2016. In the restoration experiment, mean Dice was 0.686 with random initialisation and 0.638-0.642 following pretraining. Conclusions: In-domain masked-reconstruction pretraining improved mean Dice relative to random initialisation in the custom U-Net, but not consistently in the ResNet-34 U-Net, where ImageNet initialisation performed better at every budget. Adding self-supervised pretraining after ImageNet provided no consistent gain. These findings are specific to the dermoscopic lesion-segmentation experiments and do not establish melanoma-specific diagnostic or clinical performance.

Background: Cervical spinal cord (CSC) compression and myelopathy are frequent, severe complications in mucopolysaccharidosis (MPS) types I and VI. Early detection is critical to prevent irreversible neurological deficits and preserve quality of life. This prospective study aimed to identify quantitative neuroimaging biomarkers of structural CSC damage and myelopathy in MPS I and VI. Methods: Structural and diffusion magnetic resonance imaging (MRI) of the CSC was analyzed in 17 MPS I patients, 3 MPS VI patients (19{+/-}7 years), and 19 case-matched healthy controls (20{+/-}7 years). CSC macrostructure (cross-sectional area, diameters) and microstructure (fractional anisotropy [FA], radial [RD], mean, axial diffusivities) were evaluated within and beyond the maximal compression areas. Results: MPS patients demonstrated reduced CSC cross-sectional area and anterior-posterior diameter at upper cervical levels, alongside widespread microstructural alterations (lower FA, higher RD) throughout the CSC versus controls (pFDR<0.05). Restricted foramen magnum space was associated with microstructural damage at lower cervical levels without overt macrostructural compression. Morphological and microstructural metrics were sensitive to CSC compression and myelopathy severity (pFDR<0.05). Lower CSC integrity correlated with more advanced growth retardation (p<0.001). Among patients receiving hematopoietic stem cell transplantation, earlier treatment correlated with preserved upper CSC morphology (p<0.05). Distinct neuroimaging phenotypes were observed across MPS genotypes (pFDR<0.05). Conclusions: CSC MRI provides candidate neuroimaging biomarkers for early detection and characterization of myelopathy in MPS I and VI. Structural damage extends beyond primary maximal compression areas, highlighting these markers for clinical monitoring and future therapeutic trials.

Schizophrenia spectrum disorders (SSD) have been linked to immune dysregulation and altered intestinal barrier function, although findings across studies remain inconsistent. The present study investigated circulating markers of intestinal permeability and immune activation in individuals with early-stage SSD and healthy controls and examined their associations with clinical outcomes. Serum concentrations of intestinal fatty acid-binding protein (iFABP), zonulin, soluble CD14 (sCD14), bactericidal/permeability-increasing protein (BPI), and lipopolysaccharide-binding protein (LBP) were measured in 36 patients with SSD at baseline and one-year follow-up and in 36 matched healthy controls. Linear mixed-effects models adjusted for age, sex, and body mass index (BMI) were used to evaluate group differences and longitudinal changes. No biomarker differences between patients and controls, and no longitudinal changes within patients, remained statistically significant after correction for multiple testing. Similarly, biomarker concentrations were not associated with symptom severity, global functioning, or neutrophil-to-lymphocyte ratio (NLR). Exploratory analyses identified a robust positive association between BMI and zonulin concentrations and additional associations among iFABP, LBP, and sCD14. Exploratory mixed-effects models identified a negative non-significant association between iFABP and sCD14 after adjustment for age, sex, and BMI. Overall, the findings do not support a generalized alteration of gut permeability or immune activation markers in SSD. Instead, they suggest that variation in gut-related biomarkers may be more strongly influenced by metabolic and individual factors than by diagnosis or current symptom severity.

Background: The relationship between recent viral infection and subsequent susceptibility to, or protection against, other viruses remains unclear, particularly across ages and symptom status. Methods: A prospective cohort study enrolled participants aged 6 months to 49 years in Washington and Oregon, USA, 2022-2024. Weekly nasal swabs were tested for SARS-CoV-2, respiratory syncytial virus (RSV), and influenza A/B; additional swabs were collected during symptomatic illness and were also tested for parainfluenza (PIV) 1-4, endemic human coronaviruses (hCOV), adenovirus, human metapneumovirus, human rhinovirus (hRV), and enteroviruses. Proportional hazards regression models were used to characterize associations between incident symptomatic infection (illness) and history of recent illness due to other viruses (within 1-7, 8-14, 15-30, 31-60, 61-90, 91-180 days), adjusted for geographic region, age, sex, daycare/school/work attendance, household size, and number of children in the household. For SARS-CoV-2, influenza, and RSV, these analyses were repeated to study incidence of infection (including asymptomatic cases). Findings: Of 3,560 participants, 2,969 (83%) had [&ge;]1 illness over 4,299 person-years. Adjusted hazard ratios (aHRs) in the short term ([&le;]2 weeks) after an illness were predominantly below 1, indicating that recent viral illness was often associated with lower risk of subsequent illness with other viruses. Across age groups, recent hRV illness was associated with lower risk of hCOV, PIV, and RSV illness (aHRs 0.07-0.27). SARS-CoV-2 illness was associated with a higher short-term risk of adenovirus and PIV illness (aHRs 3.20-3.33). In children aged 6 months-4 years, recent influenza illness was associated with higher short-term risk of SARS-CoV-2 illness. Including asymptomatic infections generally attenuated these associations. Interpretation: Some viral illnesses may alter short-term susceptibility to subsequent viral illnesses. Interactions may differ by age and symptom status. Funding: US Centers for Disease Control and Prevention; Washington State Department of Health.

BACKGROUND: The chromosome 9p21.3 locus is among the strongest genetic risk loci for cardiovascular disease (CVD), yet the regulatory mechanisms underlying this association remain poorly understood. Previous vascular disease studies focused predominantly on the role of this locus in smooth muscle cell (SMC) dysfunction, whereas its potential role in endothelial cells (ECs) remains less well characterized. Here, we sought to identify candidate regulatory variants at 9p21.3, their distal target genes, and effector transcripts that contribute to CVD-related phenotypes in ECs. METHODS: Promoter capture Hi-C was used to map chromatin interactions at 9p21.3 in primary human ascending aortic SMCs (n=6), supplementing the previously obtained ECs (n=16) dataset. To support the regulatory relevance of these interactions, expression quantitative trait locus analysis was performed in 136 vascular biopsies. Variant effects were prioritized in silico using AlphaGenome. The regulatory activity of the prioritized region was validated using CRISPR activation. RESULTS: Chromatin interaction mapping identified a complex regulatory architecture at the 9p21.3 locus involving 19 putative regulatory elements and 15 genes across ECs and SMCs. MIR31HG showed the most recurrent distal promoter connectivity with the 9p21.3 region. Consistent with the chromatin interaction map, expression of MIR31HG transcripts was associated with genetic variation at the 9p21.3 locus. In ECs, the lncRNA MIR31HG, but not miR-31, was responsive to hemodynamic flow. Depletion of MIR31HG in ECs was associated with endothelial activation. Epigenetic tracks and deep learning predictions highlighted an active enhancer element at 9p21.3, which was subsequently validated by CRISPR activation and increased MIR31HG expression, confirming its capacity to regulate MIR31HG over a large genomic distance of approximately 538 kb. CONCLUSIONS: Altogether, these data implicate lncRNA MIR31HG as a likely effector of the 9p21.3 CVD locus, linking the cis-regulatory element and heritable genetic variation in the respective region to endothelial activation through long-range enhancer regulation.

Background: Artificial intelligence (AI) may improve electrocardiographic differentiation of Takotsubo syndrome from acute anterior ST-segment elevation myocardial infarction (STEMI), but its impact on human diagnostic performance, the added value of visual explanations, and differences by reader background and experience remain uncertain. Methods: Using 12-lead electrocardiogram (ECG) images from two hospitals, we developed and externally validated an ensemble of three deep learning models after self-supervised pretraining and fine-tuning. The diagnostic task used ECGs recorded within 14 days after symptom onset, regardless of reperfusion status. We conducted a four-phase reader study. Phases 1-3 compared unaided interpretation, a reference sheet, and additional AI assistance. In Phase 4, readers evaluated a new ECG set after randomization to reference-sheet support alone, with AI predictions, or with AI predictions plus gradient-weighted class activation mapping (Grad-CAM) heatmaps. Diagnostic correctness was analyzed using generalized linear mixed-effects models. Exploratory analyses examined differences by cardiologist status and experience. Results: Development and external-validation cohorts, including healthy controls, comprised 832 and 95 participants, respectively. For Takotsubo-STEMI discrimination, external-validation accuracy was 0.914, with areas under the receiver operating characteristic and precision-recall curves of 0.970 and 0.984. Seventy healthcare providers completed the reader study. The estimated probability of correct diagnosis increased from 0.632 without assistance to 0.864 with AI assistance (odds ratio [OR], 3.71; P<0.001). In randomized Phase 4, AI assistance improved accuracy vs reference-sheet support alone (OR, 4.28; 95% CI, 3.08-5.95; P<0.001), and Grad-CAM provided additional benefit (OR, 1.53; 95% CI, 1.04-2.25; P=0.033). Exploratory analyses showed greater improvement among less experienced readers; with AI assistance, non-cardiologists outperformed cardiologists. AI-assisted readers remained less accurate overall than the model alone. Conclusions: An ECG image-based ensemble differentiated Takotsubo syndrome from anterior STEMI and improved reader accuracy. Visual explanations provided additional benefit, while experience-related differences highlight the importance of effective human-AI interaction.

Background. In people with type 2 diabetes (T2D), postprandial glucose weighs on glycaemic control, and carbohydrate counting is difficult. Patients can describe their meal but rarely quantify it. Objective. To assess (1) whether a small open large language model (LLM), fine-tuned on public data, estimates the carbohydrate content of a meal described in words, and (2) what this carbohydrate adds to the prediction of the postprandial glucose rise in two public cohorts. Methods. Step 1: the Qwen3-4B LLM was fine-tuned with QLoRA on Nutrition5k dishes and compared with an automatically matched USDA food composition table and with Qwen3-4B and Qwen3-8B used without training, on test dishes (internal control) and on 100 real meals of ShanghaiT2DM patients, whose reference carbohydrate was computed from the recorded quantities. Step 2: the maximum glucose rise between 0 and 120 minutes ({Delta}Gmax) was predicted by personalised gradient boosting (M2), compared with a spline linear regression and with the Chronos-2 foundation model, using only information available at meal start and person-level cross-validation, in ShanghaiT2DM (people with T2D recruited in two Shanghai hospitals; carbohydrate estimated by the LLM) and CGMacros (free-living adults in Santa Barbara, California; carbohydrate entered in the dataset, no text). Step 3: the model was transferred between datasets. 95% confidence intervals (CI) were obtained by cluster BCa bootstrap (2,000 resamples). Results. On the 100 real meals, the carbohydrate mean absolute error (MAE) of the fine-tuned model was 12.2 [9.9, 15.2] g, almost half that of the table (22.6 g; paired difference -10.5 [-16.8, -6.0] g) and nearly a third of that of the same model without fine-tuning (34.3 g), with a bias of -10.3 g; on test dishes it was 2.2 [2.0, 2.5] g. Personalised gradient boosting (M2) reached a correlation between predicted and observed rises of r = 0.63 [0.55, 0.69] in CGMacros (1,286 meals, 45 adults) and 0.49 [0.42, 0.54] in ShanghaiT2DM (2,339 meals, 99 people with T2D), about twice the linear reference (r = 0.32 and 0.23); for a rise above 50 mg/dL, its area under the ROC curve (AUROC) was 0.80 and 0.74, versus 0.64 and 0.68 with pre-meal glucose alone. Adding carbohydrate to pre-meal glucose changed r by 0.095 [0.068, 0.128] in CGMacros and by 0.004 [-0.010, 0.017] in ShanghaiT2DM; difference between datasets: 0.091 [0.060, 0.126]. Applied from one dataset to the other, the model lost correlation ({Delta}r = -0.11 [-0.17, -0.05] towards ShanghaiT2DM and -0.12 [-0.19, -0.04] towards CGMacros). The full "text [-&gt;] LLM [-&gt;] glucose" chain was evaluated in ShanghaiT2DM only. Conclusion. An open 4-billion-parameter LLM, fine-tuned with QLoRA on public data and usable offline, almost halves the error of carbohydrate counting for real meals described in words compared with an automatically matched table, with an underestimation of large meals. Personalised gradient boosting, based on pre-meal glucose and the person's usual response, doubles the correlation of a linear reference for predicting the glucose rise in both cohorts. Carbohydrate improved this prediction in CGMacros but not in ShanghaiT2DM; the two datasets differ in many respects, and the cause of this gap is not established.

Background: Epidermal Growth Factor Receptor (EGFR) tyrosine kinase inhibitors (TKIs) are widely used for EGFR-mutated non-small cell lung cancer (NSCLC) but have distinct toxicity profiles across generations. Aims: To investigate relationships between EGFR-TKI physicochemical, pharmacokinetic and pharmacological properties and drug-specific adverse drug reactions (ADRs) using national and global pharmacovigilance data. Methods: ADRs for erlotinib, afatinib and osimertinib were extracted from the UK MHRA Yellow Card scheme and WHO VigiAccess. Disproportionality analysis used reporting odds ratios (RORs). Signals were defined as ROR >2, lower 95% confidence interval >1 and >3 cases. Drug properties were curated from ChEMBL, emc, SmPC and EMA sources. Results: VigiAccess contained 88,662 individual case safety reports (163,619 ADRs) and Yellow Card contained 752 reports (1,814 ADRs). Four System Organ Class (SOC) signals were identified in VigiAccess and three in Yellow Card, with five additional Yellow Card Preferred Term signals. Eye disorders showed the strongest consistent association with erlotinib (VigiAccess ROR 4.33; Yellow Card ROR 4.18). Osimertinib was associated with pneumonitis (ROR 7.67) and pulmonary embolism (ROR 3.94). Afatinib showed psychiatric disorder signals in both databases (ROR 3.2-3.3) and diarrhoea (ROR 2.79). Neoplasm-related signals likely reflected confounding by indication. Conclusions: EGFR-TKIs demonstrate generation-specific pharmacovigilance profiles. Integration with drug properties supports mechanistic plausibility for several signals, while the atypical afatinib psychiatric association warrants further investigation. Findings are hypothesis-generating and require confirmation in controlled observational or mechanistic studies.