Citations

e16322 Background: Small intestine neuroendocrine carcinoma (NEC) is a rare, aggressive malignancy comprising less than 0.1% of gastrointestinal cancers. Population-level data remain limited, restricting understanding of long-term survival patterns, demographic disparities, and evolving disease burden. We evaluated survival outcomes, demographic predictors, and incidence trends of small intestine NEC in the United States. Methods: We conducted a retrospective population-based cohort study using SEER Research Data (17 registries, November 2024 submission). Patients diagnosed with malignant small intestine NEC (ICD-O-3 morphology code 9.3.3.1.2) between 2000 and 2022 were identified using SEER*Stat (v9.0.4.2). Variables including age group, sex, race, year of diagnosis, survival time, and vital status were analyzed in R (v4.5.1). Overall survival (OS) was estimated using Kaplan–Meier methods and compared using log-rank testing. Cox proportional hazards models identified independent predictors of OS. Annual case counts were modeled using time-series forecasting to project future incidence. Results: A total of 506 patients were included. Estimated OS was 94.4% (95% CI 92.4–96.4) at 1 year, 88.1% (85.3–91.0) at 3 years, and 82.4% (79.0–85.9) at 5 years, with a subset demonstrating survival beyond 15 years. Increasing age was associated with worse survival (log-rank p = 0.058). Significant racial differences in OS were observed (log-rank p < 0.0001). In multivariable analysis, race remained independently associated with OS. Incidence forecasting projected a modest increase in annual case burden over the next decade. Conclusions: Small intestine NEC demonstrates heterogeneous survival outcomes, significant demographic disparities, and projected growth in disease burden. These findings highlight the need for improved risk stratification, equitable access to care, and integration of molecular and treatment-level data to optimize management of this rare malignancy

Abstract To enhance the prediction of traumatic brain injury (mTBI) outcomes, we propose a deep learning approach that integrates brain computed tomography (CT) scans with corresponding synthetic T1-weighted magnetic resonance imaging (T1-MRI). Our method significantly outperforms the prediction using CT scans alone. TRACK-TBI Pilot dataset, which includes imaging and clinical outcome data from patients with TBI, is studied. The hypothesis is brain CT and T1-MRI complement each other and together will improve TBI prognosis compared to using either CT or T1-MRI alone. Since CT and T1-MRI may not be available for the same individual, we employed a specialized version of a generative adversarial network (GAN), known as fixed-point GAN (FP-GAN). FP-GAN was trained using unpaired CT and T1-MRI scans to generate synthetic T1-MRIs from real CT scans. This process produced pseudo-paired CT-MRI data, which was then used to train a deep learning classifier for outcome prediction. The classifier consists of dual parallel 3D ResNet-18 models, each independently processing T1-MRI and CT scans. We used Glasgow Outcome Scale-Extended (GOSE) scores at 3 months post-TBI as the measure of patient outcomes. To avoid data leakage, the subjects used in FP-GAN and ResNet-18 model have no overlap. We further divided the paired data, allocating 69 samples for 5-fold cross-validation and 17 samples for testing. Prognostic performance was evaluated using the area under the receiver operating characteristic curve (AUC), F1-score (the harmonic mean of precision and recall), sensitivity (true positive rate) and specificity (true negative rate). For binary classification, we defined good recovery as GOSE ≥ 7 (positive) and poor recovery (negative) as 3 ≤ GOSE ≤ 6. Accordingly, our training set consists of 24 subjects with poor recovery and 45 subjects with good recovery, while the testing set includes 5 subjects with poor recovery and 12 subjects with good recovery. A DeLong test on AUC confirms that the improvement from incorporating synthetic T1-MRI (AUC = 0.76 ± 0.10) is statistically significant (P &lt; 0.05) compared to using CT alone (AUC = 0.68 ± 0.13). The significant improvement from using the combination of real CT and synthetic T1-MRI in sensitivity (SEN = 0.95 ± 0.07) and overall performance metrics, such as F1-score (F1 = 0.84 ± 0.03), suggests that the proposed approach provides a robust and effective prognostic approach compared to using CT alone (SEN = 0.83 ± 0.18 and F1 = 0.76 ± 0.07). This pilot research demonstrates the potential of a deep learning-based harmonization model to bridge the gap between CT and T1-MRI in TBI assessment. By integrating synthetic T1-MRI with CT, prediction performance is substantially enhanced.

Abstract Introduction Hodgkin's lymphoma is known to have favorable survival outcomes; however, limited data exist regarding survival disparities between rural and urban populations. In this study, we aimed to investigate the differences in survival outcomes between the two populations. Methods Data for patients diagnosed with Hodgkin lymphoma (HL) between 2000 and 2015 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Rurality was identified using the Rural-Urban Continuum Codes (RUCC). Patients aged ≥ 20 years with HL as the only primary malignancy were included. Individuals with unknown race or those diagnosed at autopsy were excluded. The Ann Arbor Staging system was utilized for HL staging at diagnosis. HL-specific survival was defined as the time from diagnosis to death attributable to HL and was censored at the time of death from other causes or at last follow-up. Meanwhile, overall survival (OS) was defined as the time from diagnosis to death from any cause and was censored if the patient was still alive. A Cox proportional hazards model was performed to analyze the predictors for 10-year overall and HL-specific mortality. All statistical analyses were conducted using RStudio (version 2024.12.0). Results Between 2000 and 2015, a total of 22,505 patients diagnosed with Hodgkin lymphoma (HL) were included in the analysis. Among them, 13,657 (61%) were aged between 20 and 44 years. Overall, 10,137 patients (47%) were female, 14,885 (66%) were White, 2,672 (12%) were Black, and 3,724 (17%) were Hispanic. The majority of patients resided in urban counties, while only 2,266 (10.1%) lived in rural areas. Among rural patients, 845 (37%) were diagnosed with stage II disease, followed by 475 (21%) with stage III at the time of diagnosis. Notably, both the 10-year overall survival (69.5%) and Hodgkin lymphoma–specific survival (79.7%) were significantly lower among rural patients compared with urban patients (75.3% and 83.5%, respectively; p &lt; 0.001). In the multivariable model for 10-year HL-specific survival, older age (&gt; 65 years) was associated with the highest risk of mortality (HR 10.32; 95% CI 9.42–11.32). Although survival improved over the study period, male patients remained at increased risk of mortality (HR 1.26; 95% CI 1.18–1.35). Urban residence, diagnosis in more recent years, early stage, female sex, White race, and married status were independently associated with improved overall survival (p &lt; 0.01). Conclusion In this population-based study, both the 10-year overall and HL-specific survival were consistently lower among rural vs. urban patients. Male sex, older age, and unmarried status were particularly associated with worse survival outcomes. Further research is warranted to identify possible strategies for improving healthcare access and reducing survival disparities among rural patients.

BACKGROUND: Complex regional pain syndrome (CRPS) is a debilitating and painful condition accompanied by sensory, autonomic, trophic, and/or motor abnormalities. Although CRPS is rare in the general population, the prevalence among individuals at higher risk, particularly posttraumatic and postsurgical patients, remains unknown. This study aims to provide a benchmark that quantifies CRPS prevalence in high-risk groups, and offers insights on potential predictors of developing CRPS. METHODS: We conducted a systematic review and meta-analysis to identify studies reporting prevalence of CRPS after an inciting event (eg, fracture, surgery), specifically 12-month and 24-month prevalence (primary outcomes), as well as 3-month and 6-month prevalence (secondary outcomes). Estimates from individual studies were transformed using double-arcsine transformation, and the resulting estimates with 95% confidence interval (CI) were pooled in a meta-analysis using a random-effects model. RESULTS: We included 214 articles with data from 2491,378 participants worldwide (35 countries), of which 16,873 had CRPS. The pooled 12-month and 24-month global prevalence was 3.04% (95% CI, 2.64–3.48) and 6.46% (95% CI, 5.46–7.53), respectively. Subgroup analysis and meta-regression were performed to understand the impact of population-dependent (mechanism of injury, type of CRPS), contextual-dependent (socioeconomic status), and methodological-dependent (study design, publication year) factors. The 12-month prevalence was higher in countries with a high human development index (HDI) compared to those with a medium or very high HDI, was higher in participants with a traumatic inciting injury only versus those with surgical injury only or traumatic/surgical injury, and was higher in prospective versus retrospective studies. Meta-regression analysis showed that publication year was a significant moderator, with more recent articles reporting lower 12-month prevalence. CONCLUSIONS: This study provides a benchmark of the global prevalence of CRPS, which anesthesiologists and pain specialists can use to prioritize early diagnosis and identify those at the highest risk for CRPS.

Background The aim of this study was to examine the potential added value of including neuropsychiatric symptoms (NPS) in machine learning (ML) models, along with demographic features and Alzheimer's disease (AD) biomarkers, to predict decline or non-decline in global and domain-specific cognitive scores among community-dwelling older adults. Objective To evaluate the impact of adding NPS to AD biomarkers on ML model accuracy in predicting cognitive decline among older adults. Methods The study was conducted in the setting of the Mayo Clinic Study of Aging, including participants aged ≥ 50 years with information on demographics (i.e., age, sex, education), NPS (i.e., Neuropsychiatric Inventory Questionnaire; Beck Depression and Anxiety Inventories), at least one AD biomarker (i.e., plasma-, neuroimaging- and/or cerebrospinal fluid [CSF]-derived), and at least 2 repeated neuropsychological assessments. We trained and tested ML models using a stepwise feature addition approach to predict decline versus non-decline in global and domain-specific (i.e., memory, language, visuospatial, and attention/executive function) cognitive scores. Results ML models had better performance when NPS were included along with a) neuroimaging biomarkers for predicting decline in global cognition, as well as language and visuospatial skills; b) plasma-derived biomarkers for predicting decline in visuospatial skills; and c) CSF-derived biomarkers for predicting decline in attention/executive function, language, and memory. Conclusions NPS, added to ML models including demographic and AD biomarker data, improves prediction of downward trajectories in global and domain-specific cognitive scores among community-dwelling older adults, albeit effect sizes are small. These preliminary findings need to be confirmed by future cohort studies.

AbstractObjectiveThis study investigated the comparative performance of ear, nose, and throat (ENT) physicians in correctly detecting ear abnormalities when reviewing digital otoscopy imaging using 3 different visualization methods, including computer‐assisted composite images called “SelectStitch,” single video frame “Still” images, and video clips. The study also explored clinicians' diagnostic confidence levels and the time to make a diagnosis.Study DesignClinician diagnostic reader study.SettingOnline diagnostic survey of ENT physicians.MethodsNine ENT physicians reviewed digital otoscopy examinations from 86 ears with various diagnoses (normal, perforation, retraction, middle ear effusion, tympanosclerosis). Otoscopy examinations used artificial‐intelligence (AI)‐based computer‐aided composite image generation from a video clip (SelectStitch), manually selected best still frame from a video clip (Still), or the entire video clip. Statistical analyses included comparisons of ability to detect correct diagnosis, confidence levels, and diagnosis times.ResultsThe ENT physicians' ability to detect ear abnormalities (33.2%‐68.7%) varied depending on the pathologies. SelectStitch and Still images were not statistically different in detecting abnormalities (P > .50), but both were different from Video (P < .01). However, the performance improvement observed with Videos came at the cost of significantly longer time to determining the diagnosis. The level of confidence in the diagnosis was positively associated with correct diagnoses, but varied by particular pathology.ConclusionThis study explores the potential of computer‐assisted techniques like SelectStitch in enhancing otoscopic diagnoses and time‐saving, which could benefit telemedicine settings. Comparable performance between computer‐generated and manually selected images suggests the potential of AI algorithms for otoscopy applications.