Xuejun Zeng

Background: Tophi are crucial for the assessment and treatment monitoring of gout. Musculoskeletal ultrasound serves as an important complementary tool to physical examination of tophi. However, its clinical utility is limited by operator dependency, the need for tedious manual delineation, and the difficulty in identifying tophi with atypical sonographic features. This study aims to explore the feasibility of deep learning in the automatic segmentation of tophi in ultrasound images by evaluating different segmentation algorithms. Methods: A total of 730 original tophi ultrasound images were collected and randomly divided into a training set (558 images), a validation set (89 images) and an internal test set (83 images) at the patient level. An independent external test set containing 30 images was additionally included for generalization assessment. Four deep neural networks—U-Net, Attention U-Net, TransUNet, and Deeplabv3+—were applied for tophus segmentation in this study. The Jaccard Similarity Coefficient (JSC) and Dice Similarity Coefficient (DSC) were adopted as the evaluation metrics on the test set. The number of parameters and per-image inference time of each model were also compared. Segmentation performance of all models was further compared with annotations from junior and senior sonographers, and false positive rates were calculated on non-tophaceous gout ultrasound images to verify model specificity. Results: All four models exhibited good convergence during the training and validation process. In the internal test set, Deeplabv3 + achieved the best performance in terms of average JSC (64.60%) and average DSC (76.77%), followed closely by TransUNet (average JSC 63.59%, average DSC 76.29%); U-Net had the smallest number of parameters (31.00 million), along with the fastest per-image inference time (68.78 ms). Model performance declined slightly on the external test set, with no statistically significant inter-model differences observed. The segmentation accuracy of Deeplabv3 + and TransUNet was comparable to senior physicians and superior to junior physicians. When applied to images without tophi, all models generated only very small segmented regions with overall low false positive rates, and TransUNet showed the lowest false positive rate across most non-tophaceous lesions. Conclusions: This study demonstrates the strong feasibility of deep learning for automated tophus segmentation in ultrasound images, laying a foundation for quantitative assessment of tophus volume and enabling the development of standardized, objective tools to support gout assessment and treatment monitoring.

Gout is a metabolic arthritis caused by the deposition of monosodium urate (MSU) crystals, secondary to elevated serum uric acid (SUA), which triggers joint inflammation [1]. It typically begins as...

Background: VEXAS syndrome is a severe autoinflammatory disease characterized by systemic inflammation, rheumatic manifestations, and hematologic abnormalities. Its clinical heterogeneity and overlap with other conditions complicate diagnosis. No case series have been reported from China. This study aimed to (1) describe the clinical features of Chinese patients with VEXAS syndrome and (2) propose screening recommendations to facilitate earlier identification. Methods: We conducted a retrospective cohort study at Peking Union Medical College Hospital, enrolling 57 male patients with unexplained systemic inflammation and hematologic abnormalities. UBA1 gene sequencing was performed using bone marrow or peripheral blood samples. Clinical and laboratory data were extracted from electronic medical records. A screening model was developed using LASSO regression based on core features of confirmed cases and externally validated in two independent cohorts undergoing UBA1 sequencing. Results: Among 57 suspected cases, 21 carried pathogenic UBA1 variants. Compared with non-VEXAS patients, VEXAS patients more frequently showed older age at onset, skin lesions, chondritis, macrocytic anemia, and characteristic bone marrow vacuolization. Importantly, skin lesions were the initial presenting symptom in over half of VEXAS patients, whereas fever predominated in the non-VEXAS group. Rare manifestations included glomerulonephritis, acute cerebral infarction, and pulmonary arterial involvement. A novel UBA1 variant (p.Thr318Met) was also identified. A screening model incorporating age at onset, skin lesions, chondritis, macrocytic anemia, and bone marrow vacuoles demonstrated good sensitivity and specificity in an external validation cohort. Conclusion: This study is the first to characterize the clinical features of VEXAS syndrome in a Chinese population and to delineate its phenotypic spectrum. The proposed screening tool provides a practical aid for clinicians in identifying suspected cases and may facilitate improved diagnostic pathways and patient outcomes.

This study aimed to investigate the role of serum uric acid (SUA) in disease progression from baseline to first cardiometabolic disease (FCMD), subsequent cardiometabolic multimorbidity (CMM), and mortality. We conducted a prospective analysis using UK Biobank data (n = 418,765). CMM was defined as the coexistence of at least at least two cardiometabolic conditions out of type 2 diabetes (T2D), ischemic heart disease (IHD), and stroke. Multi-state models were employed to analyze the associations between SUA levels and disease transitions across different stages, including disease-specific analyses for individual cardiometabolic diseases. During follow-up, 54,946 participants developed FCMD, of whom 6,260 progressed to CMM. Each 1 mg/dL increase in SUA was associated with higher risks of FCMD [HR: 1.08 (1.07, 1.09)], transition from FCMD to CMM [HR: 1.07 (1.05, 1.10)], and all-cause mortality from all states [HR 1.04 (1.03, 1.05) from baseline to death; HR 1.03 (1.01, 1.05) from FCMD to death; HR 1.08 (1.03, 1.12) from CMM to death]. When FCMD was further divided into IHD, stroke, and T2D, we found that SUA played different roles in disease-specific transitions even within the same stage, with the strongest associations observed for incident T2D [HR: 1.15 (1.14, 1.17)] and stroke-to-CMM progression [HR: 1.11 (1.04, 1.18)]. These associations remained robust across sensitivity analyses and were stronger in younger participants (< 60 years). Elevated SUA levels serve as a potential marker across the entire trajectory of cardiometabolic disease progression. These findings suggest the importance of SUA monitoring and management in preventing cardiometabolic disease progression.

Arthritis is a global health concern, causing significant social burdens in pain, disability and healthcare costs (Dehlin et al. 2020; Finckh et al. 2022; Giorgino et al. 2023), and a prevalent dis...

The role of vitamin D on the risk and clinical outcomes of sepsis remains incompletely elucidated. This study aims to investigate the association between s

Insulin resistance is a key pathophysiological mechanism potentially linking cardiometabolic diseases, yet the comparative performance of various surrogate

Purpose Pathogenic variants in BTK cause X-linked agammaglobulinemia (XLA), yet some cases remain undiagnosed despite extensive genetic testing. Additionally, autoimmune complications have been increasingly reported, but the underlying etiology is still unclear. This study aimed to use RNA sequencing (RNA-seq) to resolve an undiagnosed case and characterize transcriptomic features associated with disease mechanisms of XLA. Methods We performed genome sequencing and RNA-seq, detected outliers against 89 controls to identify disease-causing variants, and validated functional consequences through RT-PCR. We further conducted gene set enrichment analysis (GSEA) to delineate the transcriptomic features of XLA. Results We recruited an 18-year-old male who experienced recurrent infections since infancy but developed idiopathic portal hypertension recently. He had received four noninformative genetic test results. RNA-seq enabled the identification of significantly decreased BTK expression caused by a de novo hemizygous variant (NM_000061.3:c.-32A > G) located in the 5’ UTR. Outlier analysis revealed reduced expression of genes critical for not only immunoglobulin synthesis, but also T-cell development and inflammation intensification, such as IGHM, TRBV24-1, and PLVAP. GSEA additionally revealed significant enrichment in pathways associated with immune response and inflammatory response, suggesting broader immune dysregulation. These findings align with the growing recognition of autoimmune complications in XLA. Conclusion This study emphasizes the clinical utility of RNA-seq and provides a comprehensive expression profile of XLA. These findings can guide future molecular analyses in undiagnosed cases and enhance understanding of autoimmune complications in XLA, suggesting a Multidisciplinary Team approach in both diagnostic and treatment processes.