Ruijin Hospital

Although many previous studies have highlighted the advances in prediction models, instruments for pathological and histological diagnosis and treatment, as well as individualized treatment modalities in mental disorders, these previous syntheses usually study the research outcomes separately and ignore the holistic integration of research regarding artificial intelligence technological approaches, data sets used and applications in mental health research. We used the BioBERT pretrained language model to systematically extract relevant information and develop an extensive knowledge graph that includes 3158 entities connected with 3248 different relationships. Our knowledge graph delineates essential artificial intelligence technological frameworks and explicitly maps out the relationships linking artificial intelligence methods, mental disorders, and diverse data modalities. The synthesis, centered on the analytical axis of "method-disease-data," highlights key research areas where artificial intelligence and neuropsychiatry meet. Specifically, it focuses on key applications in early detection, improved accuracy of diagnosis, and individualized therapeutic interventions. In addition, we summarized the applications derived from basic research findings that extend to ongoing clinical trials, revealing the path toward future clinical application in psychiatric work. Importantly, the research paid special attention to the application of artificial intelligence in identifying key brain regions and neural circuits, providing important clues for elucidating the neural mechanisms of mental disorders and developing targeted interventions. Although artificial intelligence presents great opportunities, there are also significant challenges, including imbalanced data sets, ethical issues, and clinical concerns about trustworthiness and transparency. Strategies to address these challenges are proposed, and a perspective on emerging methods enabled by artificial intelligence is provided, which are expected to greatly change the management and treatment of the future.

Skeletal aging is characterized by increased fragility, reduced bone mass, and deterioration of bone microstructure. Although aging-related immune dysfunction of macrophages, namely immunosenescence, is known to contribute to this process, the underlying mechanism remains poorly understood. Here, we find that the senescence of macrophages leads to a decrease in the expression of Sirtuin3 (Sirt3), which in turn leads to increased basal and lipopolysaccharides (LPS)-induced protein expression of NLRP3 and facilitates the assembly of NLRP3 inflammasome in macrophages that mediates aging-related osteoporosis. Given the phagocytic property of macrophages, we develop a genetically engineered apoptotic body-based platform for targeted delivery of Sirt3 to macrophages and verify that Sirt3-enriched apoptotic bodies (ABs-Sirt3) delay skeletal aging by promoting ubiquitination and degradation of NLRP3. Our work reveals that Sirt3 plays a key role in regulating aggravated inflammatory responses that accelerate skeletal aging during macrophage senescence and illustrates a novel nanotechnology-based therapeutic approach targeting immune senescence-induced acceleration of skeletal aging, which may provide potential therapeutic value for human patients with age-related osteoporosis.

An editorial on “Human cytomegalovirus reactivation in cirrhosis patients with acute decompensation” by Hong et al. Title: Human cytomegalovirus reactivation in decompensated cirrhosis: marker of immunosuppression or contributor to severity?

High-definition transcranial alternating current stimulation (HD-tACS) is an emerging neuromodulatory intervention for major depressive disorder (MDD). Here, we evaluate the efficacy and safety of a portable, low-intensity (1.6 mA), 10 Hz HD-tACS protocol targeting the bilateral dorsolateral prefrontal cortex (DLPFC) in adults with MDD. This randomized clinical trial was conducted including 120 adult MDD participants. Participants underwent 20 sessions (5 sessions/week for 4 weeks) of either active 10 Hz HD-tACS (1.6 mA) or sham stimulation a ratio of 1:1 and were followed for an additional 4 weeks. The primary outcome was the percentage reduction in 17-item Hamilton Depression Rating Scale scores at week 4. The active group showed significantly greater improvement in depression than the sham group (48% ± 14% vs 24% ± 13%, p < 0.001), sustained through the 4-week follow-up period. Additionally, significant improvements in the active group were also found in anxiety, sleep efficiency, quality of life components of physical, vitality, social functioning, and mental health. Adverse events including two cases of new-onset hypomania occurred, but no serious adverse events were observed. Thus, this trial supported the efficacy and good tolerability of 10 Hz HD-tACS for patients with MDD. Further research requires more focus on the underlying neural mechanisms and the optimized stimulation protocols. Trial Registration: China Clinical Trial Registry Identifier: ChiCTR2500103570. ClinicalTrials.gov Identifier: NCT06891326.

The association between nonalcoholic fatty liver disease (NAFLD) and atrial fibrillation (AF) remains controversial. This study aimed to investigate the potential relationship between NAFLD and AF risk, utilizing data from the UK Biobank. In this large-scale population-based cohort study, participants without a history of AF were captured from the UK Biobank from 2006 to 2010. NAFLD was identified using a fatty liver index threshold of ≥60. Cox regression models with progressive covariate adjustment were employed. Propensity score matching and competing risk models were utilized in sensitivity analysis. Subgroup analyses were performed to explore potential heterogeneity in the association. Overall, 92 581 participants (47.6% male, median [interquartile range] age, 53.00 [47.00-58.00] years, 30 972 [33.5%] NAFLD, 61 609 [65.5%] non-NAFLD) were included at baseline. During a median follow-up of 12.88 years, 1633 new-onset AF were diagnosed in NAFLD (5.27%), while 1583 new-onset AF were diagnosed in non-NAFLD (2.57%). Incidence curves showed that the incidence of new-onset AF was consistently higher in NAFLD participants throughout the observation period. However, in the fully-adjusted multivariable Cox regression model, this association was markedly attenuated and no longer significant (hazard ratio, 0.93; 95% confidence interval, 0.84-1.02; P = 0.12). Sensitivity and subgroup analyses yielded results largely consistent with the primary findings. In this large cohort study from the UK Biobank, NAFLD was not an independent risk factor for AF. This implies that the increased incidence of AF observed in participants with NAFLD may be more closely related to cardiometabolic risk factors contributing to the development of NAFLD rather than the condition itself.

Research and development on inexpensive large-diameter polymer fibers with high mechanical and hydrophobic properties has become a hot topic in functional composites. Herein, boron nitride nanosheets (BNNS) were modified with hyperbranched polylysine (HBPL) via noncovalent interactions to obtain BNNS-HBPL, which were then incorporated into a poly(vinyl alcohol) (PVA) matrix to fabricate large-diameter PVA/BNNS-HBPL composite fibers by wet–dry spinning. The results demonstrated that the composite fibers exhibited optimal comprehensive performance at a BNNS-HBPL loading of 0.7 wt %, the crystallinity and orientation degree reached 53.9% and 96.8%, respectively, and the diameter was 130.5 dtex (113.1 μm). The tensile strength and initial modulus increased to 10.0 and 166.8 cN/dtex, representing improvements of 47.1% and 69.9% over pure PVA fibers, and the water contact angle increased to 101.8°. Finite element analysis (FEA) simulations revealed that the PVA/BNNS-HBPL composite fibers could effectively reduce the strain of the concrete through bridging and force-transferring effects in the concrete matrix. Accordingly, this work provides alternative insights into obtaining PVA/BNNS-HBPL composite fibers with strong mechanical and hydrophobic properties, indicating broad applications in concrete.

BackgroundArtificial intelligence (AI) has shown promise for automating spinal alignment assessment in adolescent idiopathic scoliosis (AIS). However, AI models typically exhibit reduced accuracy and robustness when deployed across multiple medical centers due to variability in imaging protocols and data characteristics, potentially compromising clinical diagnosis and treatment decisions.ObjectiveThis study aimed to develop a real-time, plug-and-play data transformation method to enhance the robustness of deep learning models against data heterogeneity in radiographs, thereby improving their performance in assessing AIS across multiple medical centers.MethodsIn this retrospective multicenter study, 3899 full-spine radiographs from 7 hospitals (2 from Hong Kong and 5 from Mainland China), collected between January 2012 and August 2024, were included. Data from 2 hospitals in Hong Kong (n=3034) were used for model training and internal validation, while radiographs from the 5 mainland hospitals (n=865) formed 5 independent external validation datasets. A novel pixel intensity–based data transformation method was developed to standardize image contrast and brightness across datasets and integrated into the model training process to enhance our previously developed AI model, SpineHRNet+. The enhanced model's accuracy and robustness for cobb angle (CA) prediction and severity classification were evaluated using both internal and external datasets. Data heterogeneity across centers was quantified by brightness and contrast differences. CA prediction accuracy was evaluated using residual analysis, linear regression (coefficient of determination [R²]), and Bland-Altman analyses. Model performance for disease severity classification was assessed using sensitivity, specificity, precision, negative predictive value, accuracy, and confusion matrix analysis. The transformation method aligns pixel intensity distributions across datasets using statistical profiling and optimization, ensuring consistent image characteristics while preserving anatomical integrity.ResultsThe developed data transformation method significantly reduced contrast variability between datasets, improving consistency in image characteristics and enabling more reliable AI analysis. The enhanced SpineHRNet+ achieved consistent and accurate CA predictions across external validation datasets, with mean prediction errors within 4° (SD 3.12°), and maintained an R² greater than 0.90 for all centers. The sensitivity and negative predictive value for disease severity grading improved to 90.18% and 93.16%, respectively. Bland-Altman analyses demonstrated robust agreement, with 95% limits of agreement within 7.51° across all datasets.ConclusionsThe proposed data transformation approach effectively addressed data heterogeneity, significantly improving the accuracy and robustness of SpineHRNet+ in multicenter AIS assessments. The real-time processing capability and preservation of anatomical integrity underscore the method’s clinical practicality, enabling scalable and reliable AI applications in diverse health care environments.

Raloxifene is a selective estrogen receptor modulator; which plays beneficial roles in the treatment of cardiovascular diseases in women. However, its underlying mechanism remains unclear. In the present study, we investigated whether mitochondria-derived reactive oxygen species (ROS) are involved in raloxifene-induced vasodilation. Results showed that endothelium removal attenuated the vasodilatory effect of raloxifene in rat aortic rings, whereas treatment with the cyclooxygenase inhibitor indomethacin or the estrogen receptor antagonist ICI-182,780 had no significant effect, indicating that neither COX2 nor estrogen receptors are involved. The NADPH oxidase inhibitor apocynin also did not affect raloxifene-induced vasodilation. In contrast, the mitochondrial oxidase inhibitor rotenone and the mitochondrial antioxidant mito-tempo significantly attenuated raloxifene-induced vascular relaxation. In cultured vascular smooth muscle cells (VSMCs), phenylephrine (PE) markedly increased both total intracellular and mitochondrial ROS production, which was significantly inhibited by raloxifene. Accordingly, raloxifene effectively prevented the PE-induced increase in mitochondrial membrane potential, a key component of the mitochondrial electrochemical gradient. Finally, PE induced an acute decrease in uncoupling protein-2 (UCP2) expression, which was significantly prevented by raloxifene. In conclusion, this study demonstrates that raloxifene relaxes the rat aorta, at least in part, by inhibiting mitochondria-derived ROS.