Citations

Abstract The prevalence of steatotic liver disease (SLD) is rising globally, presenting a significant public health challenge. While non-contrast computed tomography (CT) scans hold promise for the opportunistic detection of SLD, their full potential has not been systematically explored. Clinical indicators, such as serum data, are crucial for monitoring the progression of SLD. We introduce a novel Multimodal AI for Opportunistic Steatotic Liver Disease Screening (MAOSS), which integrates demographic data, serum tests, conventional CT biomarkers, and deep learning-derived CT image features to enhance model performance and explainability. MAOSS employs an innovative learning strategy, trained on a large-scale dataset (n = 1,783) encompassing both histopathologically-confirmed SLD and radiologically-reported steatosis cases. The model underwent extensive retrospective validation using steatosis grades derived from both histology and MRI-PDFF across internal and external datasets (n = 913), as well as real-world patient data (n = 18,504). It accurately detected mild to severe steatosis (≥ S1) with AUCs ranging from 0.903 to 0.929. Notably, MAOSS significantly improved the overall AUC for radiologists (n = 11) from 0.709 to 0.798 (p < 0.001). Furthermore, MAOSS was applied to optimize the standard clinical pathway for a simulated primary risk screening of SLD patients (n = 1,192), resulting in the identification of 31% more patients at risk of steatohepatitis and liver fibrosis progression, highlighting its potential to enhance SLD detection and management.

Migraine with aura(MwA) is associated with an increased risk of stroke and adverse vascular outcomes compared to those with migraine without aura (MwoA). AI-ECG prediction models developed at our institution can evaluate the probability of atrial fibrillation (AF) and estimate a patient’s age based on a normal sinus rhythm (NSR) ECG. Delta age, AI-ECG estimated age minus chronological age, indicates endothelial dysfunction and is associated with cardiovascular mortality. We aim to assess how AI-ECG prediction model outputs, specifically the AF probability and delta age, are associated with adverse vascular outcomes in patients with migraine. Adult patients diagnosed with MwA and MwoA from 2000-2020 with at least one digital, standard 12-lead ECG were identified. The first ECG that showed NSR is used as the index ECG. Patients with any adverse vascular outcomes before the index ECG were excluded. Adverse vascular outcomes used as endpoints include acute ischemic stroke, acute myocardial infarction, deep vein thrombosis/pulmonary embolism, AF, and carotid artery dissection. A total of 31301 patients (13783 MwA, 17518 MwoA) were included in our analysis. The mean age at the time of the index ECG was 44.3 (14.5), and the average follow-up time was 71.6 (69.8) months. Based on the index ECG, the mean AF probability was significantly higher in MwA compared to MwoA (2.39 vs 2.06, p< 0.001), while the delta age was lower in MwA compared to MwoA (3.7 vs 4.2, p <.001). The prevalences of all vascular outcomes are summarized in Table 1. Multivariate Cox regression models showed that a higher AF prediction model output (every 10% increase), as well as a higher delta age (every 10-year increase), are associated with an increased risk of composite vascular outcomes (HR1.20, 95%CI 1.17-1.23, p<.001, and HR1.19, 95%CI 1.14-1.23, p < .001) [Table 2]. The Kaplan-Meier curve of adverse vascular outcome-free survival by AF prediction model output, using 1% as the cutoff (HR 1.49, 95% CI 1.41, 1.57, p<.001), and by delta age, using 2-year as the cutoff (HR 1.14, 95% CI 1.08, 1.21, p <. 001) are presented in Figure 1. Having aura, older age, male, coronary artery disease, congestive heart failure, diabetes, hypertension, and tobacco use were also associated with a higher risk of adverse vascular outcomes. Our results demonstrated the potential of using AI-ECG AF prediction model output and delta age to identify migraine patients at risk for adverse vascular outcomes.

AbstractCholecystectomy is the standard treatment for acute cholecystitis, but it may not be suitable for all patients. For those who cannot undergo surgery, a percutaneous cholecystostomy tube (PCT) and ERCP-guided transpapillary gallbladder drainage are viable options. We aimed to perform a nationwide study to assess 30-day readmission rates, adverse events (AEs), and mortality rates in these two cohorts.We conducted a nationwide cohort study using data from the Nationwide Readmissions Database (NRD) from 2016 to 2019. We identified patients with acute cholecystitis during the index admission who underwent either PCT or ERCP-guided gallbladder drainage. Propensity score matching along with multivariable regression was used to compare cohorts.During the study period, 3,592 patients (average age 63.0 years) underwent endoscopic drainage, whereas 80,372 patients (average 70.8 years) underwent Interventional Radiology drainage. Utilizing multivariate Cox regression analysis, compared with ERCP, PCT had a higher risk for 30-day readmission (adjusted hazard ratio [aHR] 1.47; 95% confidence interval [CI] 1.27 to 1.71; P < 0.001). The PCT group had a significantly higher rate of readmission for acute cholecystitis compared with the ERCP group (2.72% vs 0.86%; P < 0.005). Cox proportional hazard ratio showed a 3.41-fold increased risk (95% CI 1.99 to 5.84) for readmission in the PCT group. ERCP was consistently associated with lower rates of post-procedural AEs compared with PCT including acute hypoxemic respiratory failure (P < 0.001), acute renal failure (P < 0.001), shock (P < 0.001), and need for blood transfusions (P < 0.001).Our nationwide analysis revealed that ERCP-guided gallbladder drainage should be the preferred approach for managing acute cholecystitis when unfit for surgery.

Abstract Objective The goal is to provide an overview of artificial intelligence (AI) and machine learning (ML) methodology and appraisal tailored to clinicians and researchers in the headache field to facilitate interdisciplinary communications and research. Background The application of AI to the study of headache and other healthcare challenges is growing rapidly. It is critical that these findings be accurately interpreted by headache specialists, but this can be difficult for non‐AI specialists. Methods This paper is a narrative review of the fundamentals required to understand ML/AI headache research. Using guidance from key leaders in the field of headache medicine and AI, important references were reviewed and cited to provide a comprehensive overview of the terminology, methodology, applications, pitfalls, and bias of AI. Results We review how AI models are created, common model types, methods for evaluation, and examples of their application to headache medicine. We also highlight potential pitfalls relevant when consuming AI research, and discuss ethical issues of bias, privacy and abuse generated by AI. Additionally, we highlight recent related research from across headache‐related applications. Conclusion Many promising current and future applications of ML and AI exist in the field of headache medicine. Understanding the fundamentals of AI will allow readers to understand and critically appraise AI‐related research findings in their proper context. This paper will increase the reader's comfort in consuming AI/ML‐based research and will prepare them to think critically about related research developments.

Abstract Struma ovarii (SO) is a rare subtype of ovarian teratoma composed of more than 50% thyroid tissue. Extraovarian spread of SO, called peritoneal strumosis, was previously considered benign given the lack of histological malignant features. However, the 2020 World Health Organization Classification of Female Genital Tumors reclassified peritoneal strumosis as highly differentiated follicular carcinoma of ovarian origin (HDFCO), highlighting its low-grade malignant potential. We present a 38-year-old woman with SO treated initially with right salpingo-oophorectomy, with recurrence 2 years later with multifocal metastatic lesions in the abdomen and pelvis that was successfully treated with surgical resection, total thyroidectomy, and 157 mCi of I-131. Tumor molecular testing revealed a pathogenic DICER1 variant (c.5428G&gt;T, exon 25). A second variant (c.319delinsTGCTGGTTTGACA, exon 4) was of uncertain significance. Germline testing confirmed the second DICER1 variant and also identified an increased risk variant in the APC gene (c.3920T&gt;A). This is a rare case of extensive HFDCO with DICER1 variants, which has been associated with thyroid cancer. Given the germline DICER1 variant, this may also represent the first reported instance of DICER1 syndrome manifesting as HDFCO. Further research into the prognostic factors and optimal treatment of HFDCO is needed.

NPM1 mutations (NPM1+) are found in about 30% of adult AML. While NPM1+ are generally associated with favorable risk, prior studies have shown that NPM1+ older patients, defined as ≥ 65 years, have significantly worse outcomes. The updated European Leukemia Net 2022 (ELN22) now classifies NPM1+ AML patients with FLT3-ITD (ITD+) and MDS-associated mutations (MDSm+) as adverse-risk, while NPM1+/ITD-/MDSm+ patients remain favorable-risk unless they have an adverse-risk karyotype. A recent report (Chan et al., Blood Adv 2024) found MDSm+ conferred worse OS only in patients ≥ 60 years; however another report (Othman et al., Blood 2024) did not find MDSm+ were significantly associated with adverse outcomes in a larger cohort of NPM1+ patients. To further assess this question, we examined the mutational landscape in a large cohort of pre-treatment, diagnostic specimens from 456 patients enrolled in SWOG clinical trials (SWOG-9031, SWOG-9333, S0106, and S0112) and 126 patients from the UW/FHCC Hematopoietic Repository (Total N=582). Patients were treated with curative intent as per standard practice or clinical trials. Out of the 582 patients, 191 were NPM1+ (33%), of which 148 patients had cytogenetic and long-term outcome data. Kaplan-Meier and cumulative incidence estimates were estimated. C-statistics and AUC for Cox, cause-specific and logistic regression models for overall survival (OS), event-free survival (EFS), time to relapse (TTR, calculated in n=106 who achieved complete remission [CR], death in remission analyzed as a competing risk) and CR were calculated for age, ELN17, and ELN22, as well as ELN17 and ELN22 adjusting for age. ELN17 favorable-risk patients reclassified as adverse-risk (N=5 of 108) demonstrated significantly shorter EFS (HR=2.50, p=0.05) and CR rate (OR=0.20, p=0.024), although with similar OS (HR 1.30, p=0.66) and TTR (HR=1.20, p=0.86). ELN17 favorable-risk patients reclassified as ELN22 intermediate-risk (N=23 of 108) did not show a significant difference in OS (HR 1.47, p=0.20), EFS (HR 1.35, p=0.12), CR rate (OR 0.75, p=0.46), or TTR (HR=1.64, p=0.11) compared to ELN22 favorable-risk. For ELN17 intermediate-risk (N=40), 5 patients were reclassified as ELN22 adverse-risk (12.5%) and had significantly shorter EFS (HR 2.89, p=0.037) compared to ELN22 intermediate risk, although there was not a significant difference in CR rate (OR 0.31, p=0.23) or OS (HR 0.81, p=0.67). We next compared the prognostic significance of age relative to ELN risk classification. Age was associated with higher C-statistics (OS=0.65, TTR=0.57 and EFS=0.60) and AUC (CR=0.64) compared to both ELN17 (OS=0.56, TTR=0.54, EFS=0.54, CR=0.54) or ELN22 (OS=0.57, TTR=0.57, EFS=0.57, CR=0.58). The combination of age and ELN22 resulted in the highest C-statistics for OS, TTR, and EFS (OS=0.68, TTR=0.60, EFS=0.64). To further assess the impact of MDSm on risk-stratification of NPM1+ AML, we performed multivariable analyses adjusting for FLT3-ITD, MDSm, and age in patients without adverse-risk disease. These analyses showed MDSm did not have a significant association with OS when controlling for age or FLT3-ITD. We also compared ELN22 favorable-risk patients with and without MDSm (MDSm+ N=18; MDSm- N=61). Compared with the MDSm-, there was a trend toward an inferior OS (HR 1.85, p=0.054) for MDSm+ patients but not EFS (HR 1.38, p=0.3) or TTR (HR 1.2, p=0.64). When stratified by age (&lt;65, N=53 vs. ≥65, N=26), MDSm+ in younger patients did not show a difference for OS (HR 0.96, p=0.96), EFS (HR 0.63, p=0.53) or TTR (HR 0.4, p=0.37). Older ELN22 favorable-risk patients (≥65) had similarly poor OS (MDSm- HR 3.33 and MDSm+ 3.27, respectively, p=0.001 for both) and EFS (MDSm- HR 3.11 and MDSm+ 2.39, p=0.001 and 0.01) compared to younger MDSm- patients. In our study, the ELN22 guidelines incrementally improve risk stratification for EFS in NPM1+ patients. However, age remains the most prognostic factor, and patients ≥ 65 years have similar adverse clinical outcomes, whether classified as favorable risk by ELN17, ELN22, or incorporating MDSm. Moreover, MDSm do not confer a worse prognosis in NPM1+ patients when accounting for age or FLT3-ITD status. These findings differ from a recent report by Chan et al. and are in keeping with the report by Othman et al. Therefore, further understanding of the underlying biology of age and its effect on NPM1+ AML is needed to improve outcomes for older patients.

INTRODUCTION: This study investigates seasonal variations in eosinophilic esophagitis (EoE)-related emergency department visits among adults. METHODS: We analyzed the National Emergency Department Sample (2016–2021), identifying adult patients with EoE using ICD-10 codes. Generalized additive models assessed seasonal patterns. RESULTS: Among 18,791 EoE-related emergency department visits, a significant seasonal variation was observed, peaking in summer and nadiring in winter. This pattern was consistent across all US regions. CONCLUSION: Seasonal dietary habits and social behaviors likely contribute to EoE exacerbations. Healthcare providers should emphasize management strategies during high-risk periods, particularly summer months and weekends.

Abstract Senescent osteocytes are key contributors to age‐related bone loss and fragility; however, the impact of mechanobiological changes in these cells remains poorly understood. This study provides a novel analysis of these changes in primary osteocytes following irradiation‐induced senescence. By integrating subcellular mechanical measurements with gene expression analyses, we identified significant, time‐dependent alterations in the mechanical properties of senescent bone cells. Increases in classical markers such as SA‐β‐Gal activity and p16 Ink4a expression levels confirmed the senescence status post‐irradiation. Our key findings include a time‐dependent increase in cytoskeletal Young's modulus and altered viscoelastic properties of the plasma membrane, affecting the contractility of primary osteocytes. Additionally, we observed a significant increase in Sclerostin ( Sost ) expression 21 days post‐irradiation. These biophysical changes may impair osteocyte mechanosensation and mechanotransduction, contributing to bone fragility. This is the first study to time‐map senescence‐associated mechanical changes in the osteocyte cytoskeleton. Our findings highlight the potential of biophysical markers as indicators of cellular senescence, providing more specificity than traditional, variable biomolecular markers. We believe these results may support biomechanical stimulation as a potential therapeutic strategy to rejuvenate aging osteocytes and enhance bone health.