Citations

Phonoenterography (PEG) offers a non-invasive and radiation-free technique to assess gastrointestinal activity through acoustic signal analysis. In this feasibility study, 110 high-resolution PEG recordings (44.1 kHz, 16-bit) were acquired from eight healthy individuals, yielding 6314 prominent bowel sound (PBS) segments through automated segmentation. Each event was characterized using a 279-feature acoustic profile comprising Mel-frequency cepstral coefficients (MFCCs), their first-order derivatives (Δ-MFCCs), and six global spectral parameters. After normalization and dimensionality reduction with PCA and UMAP (cosine distance, 35 neighbors, minimum distance = 0.01), five clustering strategies were evaluated. K-Means (k = 5) achieved the most favorable balance between cluster quality (silhouette = 0.60; Calinski–Harabasz = 19,165; Davies–Bouldin = 0.68) and interpretability, consistently identifying five acoustic patterns: single-burst, multiple-burst, harmonic, random-continuous, and multi-modal. Temporal modeling of clustered events further revealed distinct sequential dynamics, with Single-Burst events showing the longest dwell times, random continuous the shortest, and strong diagonal elements in the transition matrix confirming measurable state persistence. Frequent transitions between random continuous and multi-modal states suggested dynamic exchanges between transient and overlapping motility patterns. Together, these findings demonstrate that unsupervised PEG-based analysis can capture both acoustic variability and temporal organization of bowel sounds. This annotation-free approach provides a scalable framework for real-time gastrointestinal monitoring and holds potential for clinical translation in conditions such as postoperative ileus, bowel obstruction, irritable bowel syndrome, and inflammatory bowel disease.

Background: Sepsis survivors experience substantial long-term cardiovascular (CV) morbidity and mortality. The pooled-cohort atherosclerotic cardiovascular disease (ASCVD) score estimates 10-year risk in the general population, but its accuracy soon after sepsis is unknown. Objectives: (1) Describe the absolute five-year incidence of major adverse cardiovascular events (MACE) after sepsis hospitalization. (2) Evaluate the discrimination and calibration of the 10-year ASCVD score for predicting five-year MACE in this setting. Research Questions: The ASCVD score would underpredict five-year MACE in sepsis survivors, reflecting their heightened risk. Methods: We performed a retrospective closed-cohort study using electronic health records from Mayo Clinic hospitals (USA). Adults aged 40–75 years hospitalized for sepsis between 2018 and 2020 were eligible if follow-up was ≥5 years; the age criterion matched the derivation range of the ASCVD score. The primary outcome was five-year MACE—myocardial infarction (MI), ischemic or hemorrhagic stroke, or CV death. Secondly we evaluated the predictive performance of ASCVD score for five-year MACE. Discrimination was assessed with the area under the receiver-operating curve (AUROC); calibration with Calibration-in-the-Large (CITL, ideal = 0) and slope (ideal = 1). Performance metrics were calculated in patients with complete ASCVD variables. Results: A total of 6306 sepsis admissions were screened. 2424 patients were excluded for age outside 40–75 years and 879 for <5 year follow-up, leaving 2718 patients for incidence analyses. Median age was 65 years (IQR 56–70); 55 % were male, 38 % required ICU care, and 91 % were White. Median body-mass index was 28.2 kg/m2 (IQR 23.7-34.7). During follow-up, 125 patients (4.6 %) had MI, 219 (8.1 %) experienced stroke, and 303 (11.1 %) died of CV causes; overall, 1573 patients (57.9 %) met the composite MACE outcome. Complete ASCVD data were available for 409 patients. In this subset, discrimination was moderate (AUROC 0.65; 95 % CI 0.59–0.71). Calibration showed CITL +0.34 (95 % CI 0.09–0.59) and slope 0.50 (95 % CI 0.28–0.72), indicating systematic infra-prediction and over-fitting, respectively. Conclusion: Five-year MACE incidence after sepsis is high. The 10-year ASCVD score moderately discriminates but underestimates five-year CV risk in this population, highlighting the need for sepsis-specific predictive models or recalibration to guide targeted post-sepsis cardiovascular prevention.

Abstract Obesity is a major and growing driver of cardiovascular disease (CVD) in the United States, contributing to hypertension, diabetes, dyslipidemia, and structural cardiac changes that elevate the risk of ischemic heart disease, heart failure, and stroke. After decades of declining CVD mortality, recent national trends show a concerning reversal, coinciding with record-high obesity prevalence. We quantified national trends and disparities in obesity-related CVD mortality from 1999–2020 using de-identified death certificate data from the CDC WONDER database for U.S. residents aged 15–84 years, including deaths with circulatory disease (ICD-10 I00–I99) as the underlying cause and obesity (E66.0–E66.9) as a contributing cause. Age-adjusted mortality rates (AAMR) standardized to the 2000 U.S. population were analyzed using joinpoint regression to estimate annual percent change (APC) and detect inflection points. From 1999–2018, AAMR rose from 2.75 to 6.61 per 100,000 (APC 4.79%; 95% CI 4.60–4.98), followed by a significant post-2018 acceleration (APC 11.05%; 95% CI 6.76–15.51) to 8.36 in 2020. Mortality was highest among adults 65–74 years, males, Black Americans, and nonmetropolitan residents, with a 5.1-fold state variation. Young adults (15–24 years) had the largest relative increase (+210%). The 2020 surge was independent of COVID-19 mortality, indicating gaps in chronic disease management. These results highlight a growing obesity-related CVD mortality burden, with widening disparities by age, sex, race, and geography, emphasizing urgent, equity-focused prevention and targeted interventions.

ABSTRACT Transformed follicular lymphoma (tFL) is typically associated with chemotherapy resistance and a poor prognosis. There are limited data regarding outcomes after CD19‐directed chimeric antigen receptor T‐cell (CAR T) therapy in relapsed/refractory (R/R) tFL. A total of 923 adult patients with R/R tFL who received commercial CD19 CAR T therapy between 2017 and 2023 were identified in the Center for International Blood and Marrow Transplant Research registry. Median age was 64 years (range: 30–86) and median prior lines of therapy was 4 (range: 1–18). Most patients (78%) received axicabtagene ciloleucel, with 67% of patients having resistant disease at the time of CAR T infusion. At a median follow‐up of 25 months (range: 1–72) from CAR T infusion, the 2‐year overall survival (OS) was 57% (95% CI: 53–60) and progression‐free survival (PFS) was 43% (95% CI: 40–47). The 2‐year cumulative incidences of relapse or progression (rel/prog) and non‐relapse mortality (NRM) were 47% (95% CI: 44–51) and 9% (95% CI: 7–11), respectively. The overall response rate to CAR T was 76%, with a complete response rate of 63%. Grade ≥ 3 cytokine release syndrome (CRS) was observed in 7.1% and grade ≥ 3 immune effector cell–associated neurologic syndrome (ICANS) in 21.6% of patients. Multivariable analysis suggested that resistant disease status at the time of CAR T, use of bridging therapy, and high comorbidity index ≥ 3 were associated with inferior PFS and OS. Older age ≥ 60 significantly increased the risk of NRM. Our study suggests that CD19 CAR T is effective and safe for tFL.

Accurate diagnosis of gastrointestinal (GI) diseases typically requires invasive procedures or imaging studies that pose the risk of various post-procedural complications or involve radiation exposure. Bowel sounds (BSs), though typically described during a GI-focused physical exam, are highly inaccurate and variable, with low clinical value in diagnosis. Interpretation of the acoustic characteristics of BSs, i.e., using a phonoenterogram (PEG), may aid in diagnosing various GI conditions non-invasively. Use of artificial intelligence (AI) and improvements in computational analysis can enhance the use of PEGs in different GI diseases and lead to a non-invasive, cost-effective diagnostic modality that has not been explored before. The purpose of this work was to develop an automated AI model, You Only Listen Once (YOLO), to detect prominent bowel sounds that can enable real-time analysis for future GI disease detection and diagnosis. A total of 110 2-minute PEGs sampled at 44.1 kHz were recorded using the Eko DUO® stethoscope from eight healthy volunteers at two locations, namely, left upper quadrant (LUQ) and right lower quadrant (RLQ) after IRB approval. The datasets were annotated by trained physicians, categorizing BSs as prominent or obscure using version 1.7 of Label Studio Software®. Each BS recording was split up into 375 ms segments with 200 ms overlap for real-time BS detection. Each segment was binned based on whether it contained a prominent BS, resulting in a dataset of 36,149 non-prominent segments and 6435 prominent segments. Our dataset was divided into training, validation, and test sets (60/20/20% split). A 1D-CNN augmented transformer was trained to classify these segments via the input of Mel-frequency cepstral coefficients. The developed AI model achieved area under the receiver operating curve (ROC) of 0.92, accuracy of 86.6%, precision of 86.85%, and recall of 86.08%. This shows that the 1D-CNN augmented transformer with Mel-frequency cepstral coefficients achieved creditable performance metrics, signifying the YOLO model’s capability to classify prominent bowel sounds that can be further analyzed for various GI diseases. This proof-of-concept study in healthy volunteers demonstrates that automated BS detection can pave the way for developing more intuitive and efficient AI-PEG devices that can be trained and utilized to diagnose various GI conditions. To ensure the robustness and generalizability of these findings, further investigations encompassing a broader cohort, inclusive of both healthy and disease states are needed.

Abstract Brexucabtagene autoleucel (brexu-cel) is a chimeric antigen receptor T-cell therapy approved for relapsed/refractory mantle cell lymphoma (R/R MCL). Here, we report real-world effectiveness and safety outcomes of brexu-cel in a prospective study of patients with R/R MCL, including subgroups based on prior treatment with Bruton's tyrosine kinase inhibitor, bendamustine, or autologous hematopoietic cell transplant (auto-HCT) and number of prior therapy lines, using Center for International Blood and Marrow Transplant Research registry data. A total of 476 patients with R/R MCL who received brexu-cel between July 2020 and December 2022 were included in the analysis. With a median follow-up of 13.5 months, the overall response rate was 91% and complete response rate was 82%. One-year overall survival and progression-free survival rates were 76% and 63%, respectively. One-year cumulative incidence of nonrelapse mortality was 8%. Prior auto-HCT was associated with better duration of response within 6 months after infusion (hazard ratio [HR], 0.49; 95% confidence interval [CI], 0.28-0.85) but greater risk of immune effector cell–associated neurotoxicity syndrome (odds ratio [OR], 1.66; 95% CI, 1.06-2.60). Prior bendamustine was associated with increased risk of prolonged thrombocytopenia (OR, 1.90; 95% CI, 1.13-3.21). In patients with 1 to 2 prior therapy lines, relapse or progression was less frequent compared with those with ≥3 prior lines (HR, 0.64; 95% CI, 0.42-1.00). Collectively, our results suggest that real-world outcomes with brexu-cel were consistent with those of the ZUMA-2 trial, regardless of prior therapy type or number of prior therapy lines.

Accurate analysis of medical videos remains a major challenge in deep learning (DL) due to the need for effective spatiotemporal feature mapping that captures both spatial detail and temporal dynamics. Despite advances in DL, most existing models in medical AI focus on static images, overlooking critical temporal cues present in video data. To bridge this gap, a novel DL-based framework is proposed for spatiotemporal feature extraction from medical video sequences. As a feasibility use case, this study focuses on gastrointestinal (GI) endoscopic video classification. A 3D convolutional neural network (CNN) is developed to classify upper and lower GI endoscopic videos using the hyperKvasir dataset, which contains 314 lower and 60 upper GI videos. To address data imbalance, 60 matched pairs of videos are randomly selected across 20 experimental runs. Videos are resized to 224 × 224, and the 3D CNN captures spatiotemporal information. A 3D version of the parallel spatial and channel squeeze-and-excitation (P-scSE) is implemented, and a new block called the residual with parallel attention (RPA) block is proposed by combining P-scSE3D with a residual block. To reduce computational complexity, a (2 + 1)D convolution is used in place of full 3D convolution. The model achieves an average accuracy of 0.933, precision of 0.932, recall of 0.944, F1-score of 0.935, and AUC of 0.933. It is also observed that the integration of P-scSE3D increased the F1-score by 7%. This preliminary work opens avenues for exploring various GI endoscopic video-based prospective studies.

e19033 Background: CLL tends to affect older individuals, and older patients often endure greater health disparities and treatment burdens that may impact their access to healthcare resources, adherence to treatment regimens and treatment outcomes (UN, Health Inequalities in Old Age). This study aimed to assess treatment utilization patterns, discontinuation, and healthcare resource utilization (HCRU) of 1L use of BTKi in CLL, focusing on older patients ≥65 years (yrs). Methods: A retrospective observational study using the Symphony Integrated Dataverse was conducted to identify adult CLL patients initiating a 1L BTKi (zanubrutinib [ZANU], acalabrutinib [ACA] or ibrutinib [IBR]) treatment regimen from 02/2022 and 09/2024 (index period). Patients were required to have ≥30 days of continuous enrollment in the pre- and post-index periods. Patients initiating a BTKi were further categorized by age at index date. Patient sociodemographic, clinical characteristics, and outcomes were reported in overall and older (≥65 yrs) CLL patients by each BTKi group. Treatment discontinuation rate was examined at 60, 90, 120, 150, 180, and 360 days, and by median time to discontinuation (TTD), estimated using the Kaplan-Meier method, censoring patients at the end of follow-up. HCRU (outpatient, inpatient, and other services) was evaluated during time on treatment and over the follow-up period and reported as per-patient-per-year (PPPY). Results: A total of 6,846 patients initiated a 1L BTKi (ZANU: n=1,853; ACA: 3,519; IBR: 1,474) for CLL during the index period. Patients receiving ZANU were older (mean age: 70.9 yrs; 78% ≥65 yrs) than those receiving ACA (69.4 yrs; 71% ≥65 yrs) and IBR (69.6 yrs; 73% ≥65 yrs). Payer type and geography were significantly different across each BTKi group ( p <.001). There were no significant differences across the BTKis regarding baseline comorbidities. ZANU patients had the longest estimated median TTD overall (ZANU: 18.9 months [m], ACA: 17.8 m, IBR: 14.5 m; p <.001) and in patients ≥65 yrs (ZANU: 17.2 m, ACA: 15.2 m, IBR: 12.6 m; p <.001). ZANU patients also had the lowest estimated discontinuation rates at 60- , 90-, 120-, 150-, 180-, and 360-days post BTKi initiation in the overall and ≥65 yrs groups. There were significant differences across the BTKi group for inpatient visits PPPY in all patients and ≥65 yrs with ZANU having the lowest visits of the BTKis (ZANU: 2.6, ACA: 3.6, IBR: 4.6; p =.029 and ZANU: 2.5, ACA: 4.3, IBR: 5.3; p =.004, respectively). Conclusions: This real-world study demonstrated that CLL patients treated with ZANU had longer TTD, lower discontinuation rates, and less HCRU than ACA and IBR across all patients and specifically in older patients ≥65 yrs. Further studies on long-term and clinical outcomes are warranted.