ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

Purpose: This study aimed to evaluate the safety and efficacy of gemcitabine and cisplatin (GP) regimen in combination with immune checkpoint inhibitor sintilimab as neoadjuvant therapy for muscle-invasive bladder cancer (MIBC) patients and the feasibility of the following selective bladder sparing surgery. Materials and Methods: Patients with histopathologically confirmed urothelial carcinoma without distant metastases (T2-4a, N ≤ 1, M0, American Joint Committee of Cancer 8th) and with adequate organ function will be enrolled. The therapeutic regimen was sintilimab 200 mg once on day 8, gemcitabine 1,000 mg/m2 and cisplatin 35 mg/m2 once on days 1 and 8, every 21 days for four cycles. The primary endpoint was pathologic complete response (pCR, pT0N0) rate. The secondary end points were ypT < 2 rate, R0 resection rate, event-free survival, and safety. Results: From May 4, 2020, to May 20, 2023, 55 patients were enrolled. Forty-six patients were evaluated for efficacy. Among the 42 patients who underwent surgery, 16 patients (38.0%) achieved pCR. Thirty-three patients (78.6%) achieved pT < 2. With a median follow-up of 15.7 months, the 1-year event-free survival was 91.3%. Notwithstanding the poor pathological baseline characteristic of a high T3-T4a proportion (39.1%), a promising bladder preservation (including 22 patients transurethral resection of bladder tumor, 5 patients partial cystectomy, and 4 surveillances) rate was achieved (67.4%). The most common grade ≥ 3 treatment-related adverse events was neutropenia (n=15, 27.3%), which was related to chemotherapy. There were no grade 3 immune-related adverse events. Conclusion Neoadjuvant GP plus sintilimab is a promising regimen for MIBC patients, with relatively high pT < 2 rate and triggering the emerging roles for the multi-disciplinary team decision-making for bladder sparing surgery.

ObjectiveTo explore the clinical and phenomics characteristics of patients with phlegm-stasis binding syndrome and its accompanied patterns in stable angina pectoris （SAP） of coronary heart disease （CHD）. MethodsA multicenter cross-sectional study design was adopted. A total of 300 patients with SAP of CHD were enrolled and classified into 120 cases of phlegm-stasis binding syndrome， 125 cases of qi deficiency-accompanied syndrome， 38 cases of qi stagnation-accompanied syndrome， and 17 cases of toxin accumulation-accompanied syndrome according to traditional Chinese medicine （TCM） patterns. Data of patients with different TCM patterns were collected， including general condition， TCM symptoms， blood lipids， coagulation function， immune indicators， and serum metabolomics. Metabolic pathway enrichment analysis was used to compare differences in phenomics characteristics among groups. Metabolites with variable importance in projection （VIP） ≥1 and fold change （FC） ≥2 were considered as representative differential metabolites. ResultsPatients in each TCM pattern type were most commonly in the 60-75 years age group， with a relatively high proportion of males. Regarding TCM symptoms， patients with phlegm-stasis binding syndrome most commonly presented with wiry-choppy or wiry-slippery pulse， chest pain， and chest tightness； in patients with qi deficiency-accompanied syndrome， fatigue， chest pain， and weak pulse were most common； in patients with qi stagnation-accompanied sydnrome， wiry-choppy or wiry-slippery pulse， chest pain， and symptoms that increase or decrease with emotional changes， belching， or flatulence were most common； in patients with toxin accumulation-accompanied syndrome， bitter taste in the mouth， chest tightness， irritability， restlessness or manic delirium， and dry and hard stools or foul-smelling diarrhea were most common. Comparisons among the different TCM patterns showed statistically significant differences in coagulation parameters （P＜0.05）， whereas no statistically significant difference was found in blood lipid levels and immune indicators （P＞0.05）.Metabolomics analysis suggested that disorders of glycerophosphate metabolism and valine， leucine， and isoleucine metabolism are characteristic features of phlegm-stasis binding syndrome and its accompanied patterns. The representative differential metabolites between phlegm-stasis binding syndrome and qi deficiency-accompanied syndrome were 7,8-dihydrobiopterin （FC = 3.58） and oxypurinol （FC = 124.50）. Those between phlegm-stasis binding syndrome and qi stagnation-accompanied syndrome were cytidine 5′-diphosphocholine （FC = 2.62） and D-mannosamine （FC = 2.99）. Those between phlegm-stasis binding syndrome and toxin accumulation-accompanied syndrome were anserine （FC = 7.83） and canrenone （FC = 8.94）. ConclusionThere are certain differences in the clinical characteristics and phenomics characteristics among patients with SAP due to CHD exhibiting phlegm-stasis binding syndrome and its accompanied patterns. The phenomics characteristics mainly involve biological alterations such as lipid metabolism disorders， amino acid metabolism abnormalities， and multiple immune indicators activation， which may provide a reference for precise differentiation and treatment of SAP of CHD in TCM clinical practice.

ObjectiveTo explore the clinical and phenomics characteristics of patients with phlegm-stasis binding syndrome and its accompanied patterns in stable angina pectoris （SAP） of coronary heart disease （CHD）. MethodsA multicenter cross-sectional study design was adopted. A total of 300 patients with SAP of CHD were enrolled and classified into 120 cases of phlegm-stasis binding syndrome， 125 cases of qi deficiency-accompanied syndrome， 38 cases of qi stagnation-accompanied syndrome， and 17 cases of toxin accumulation-accompanied syndrome according to traditional Chinese medicine （TCM） patterns. Data of patients with different TCM patterns were collected， including general condition， TCM symptoms， blood lipids， coagulation function， immune indicators， and serum metabolomics. Metabolic pathway enrichment analysis was used to compare differences in phenomics characteristics among groups. Metabolites with variable importance in projection （VIP） ≥1 and fold change （FC） ≥2 were considered as representative differential metabolites. ResultsPatients in each TCM pattern type were most commonly in the 60-75 years age group， with a relatively high proportion of males. Regarding TCM symptoms， patients with phlegm-stasis binding syndrome most commonly presented with wiry-choppy or wiry-slippery pulse， chest pain， and chest tightness； in patients with qi deficiency-accompanied syndrome， fatigue， chest pain， and weak pulse were most common； in patients with qi stagnation-accompanied sydnrome， wiry-choppy or wiry-slippery pulse， chest pain， and symptoms that increase or decrease with emotional changes， belching， or flatulence were most common； in patients with toxin accumulation-accompanied syndrome， bitter taste in the mouth， chest tightness， irritability， restlessness or manic delirium， and dry and hard stools or foul-smelling diarrhea were most common. Comparisons among the different TCM patterns showed statistically significant differences in coagulation parameters （P＜0.05）， whereas no statistically significant difference was found in blood lipid levels and immune indicators （P＞0.05）.Metabolomics analysis suggested that disorders of glycerophosphate metabolism and valine， leucine， and isoleucine metabolism are characteristic features of phlegm-stasis binding syndrome and its accompanied patterns. The representative differential metabolites between phlegm-stasis binding syndrome and qi deficiency-accompanied syndrome were 7,8-dihydrobiopterin （FC = 3.58） and oxypurinol （FC = 124.50）. Those between phlegm-stasis binding syndrome and qi stagnation-accompanied syndrome were cytidine 5′-diphosphocholine （FC = 2.62） and D-mannosamine （FC = 2.99）. Those between phlegm-stasis binding syndrome and toxin accumulation-accompanied syndrome were anserine （FC = 7.83） and canrenone （FC = 8.94）. ConclusionThere are certain differences in the clinical characteristics and phenomics characteristics among patients with SAP due to CHD exhibiting phlegm-stasis binding syndrome and its accompanied patterns. The phenomics characteristics mainly involve biological alterations such as lipid metabolism disorders， amino acid metabolism abnormalities， and multiple immune indicators activation， which may provide a reference for precise differentiation and treatment of SAP of CHD in TCM clinical practice.

To address the scarcity of annotated data for pediatric abdominal CT imaging and the insufficient generalization capability of existing models, we constructed a self-supervised pretraining architecture tailored for pediatric CT domain adaptation based on the visual foundation model DINOv3, and validated its performance in the task of pediatric abdominal multi-organ segmentation.

ObjectiveTo explore the mechanism by which aerobic exercise improves hippocampal neuroinflammation and learning and memory ability in brain-aged mice via interferon regulatory factor 7 (IRF7). MethodsA total of 60 male specific pathogen-free (SPF) C57BL/6J mice were enrolled in the experiment, including 30 young mice (2 months old) and 30 naturally aging mice (12 months old). Young mice and old mice were respectively randomly assigned to four groups: young control group (group C, n=15), young exercise group (group CE, n=15), and aging control group (group A, n=15), aging exercise group (group AE, n=15). Mice in the exercise groups received an 8-week continuous treadmill aerobic exercise intervention, while control mice were kept under routine feeding without exercise intervention. After the intervention, Barnes maze and Y-maze behavioral tests were performed to comprehensively assess the learning and memory abilities of each group of mice. Transmission electron microscopy was used to observe the ultrastructural morphological changes of hippocampal synapses. Western blot assay was utilized to detect the protein expression and phosphorylation levels of Janus kinase 1 (JAK1), signal transducer and activator of transcription 1 (STAT1), as well as the expression of postsynaptic density protein 95 (PSD95). Immunofluorescence staining was applied to quantify the fluorescence intensity of ionized calcium-binding adapter molecule 1 (IBA1) in hippocampal tissues. The expression levels of inflammatory cytokines including interferon-α (IFN-α), interferon-β (IFN-β), tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β) were measured by enzyme-linked immunosorbent assay (ELISA). Quantitative real-time polymerase chain reaction (qRT-PCR) was performed to detect the mRNA expression level of IRF7, and DNA methylation assay was further conducted to analyze the methylation level of the IRF7 gene in mouse hippocampus. ResultsAt the molecular level, compared with group C, the protein expression levels of JAK1, STAT1, p‑JAK1, and p‑STAT1 in group A were significantly increased (P<0.05), whereas PSD95 expression was decreased (P<0.05). ELISA assays showed that the levels of IFN‑α, IFN‑β, TNF‑α, and IL‑1β were significantly elevated in group A (P<0.001). RT‑PCR analysis revealed that, compared with group C, the mRNA expression of IRF7 in group A was markedly upregulated (P<0.000 1), while the CpG methylation rate of the IRF7 promoter was significantly reduced (P<0.05). In contrast, all the above parameters in the AE group showed improvements relative to those in group A. Immunofluorescence staining indicated positive IBA1 expression in the brain tissues of group A mice. Exercise intervention reversed these abnormal changes. ConclusionIn aged mice, aerobic exercise induces hypermethylation of the IRF7 promoter region, suppresses IRF7 expression and activation of the downstream IFN‑I/JAK1/STAT1 signaling pathway, accompanied by alleviated neuroinflammation, improved learning and memory, and ameliorated brain aging-related phenotypes.

Objective:To prepare heparin (Hep)-modified gelatin methacryloyl (GelMA) microspheres and to investigate their application in liver-targeted delivery of adipose-derived mesenchymal stem cells (ADSCs).Methods:GelMA microspheres were modified with Hep to obtain GelMA-Hep microspheres. The surface morphology of the GelMA-Hep microspheres was observed by scanning electron microscopy. The changes of carbon atoms, nitrogen atoms and sulfur atoms on the surface of the GelMA-Hep microspheres were detected by X-ray photoelectron spectroscopy. The surface chemical group composition of the GelMA-Hep microspheres was analyzed by Fourier transform infrared spectrometer. The swelling properties of the GelMA-Hep microspheres were detected by water absorption swelling experiment. Human liver HL-7702 cells transfected with lentivirus were co-cultured with GelMA, GelMA-dopamine (GelMA-dop) and GelMA-Hep microspheres. The effects of microspheres on cell proliferation activity were evaluated by cell counting kit-8 method and live/dead cell staining experiment. The adhesion of microspheres to cells was observed by confocal microscopy. The GelMA-Hep microspheres loaded with ADSCs were injected into C57BL/6 mice through the tail vein, and its efficiency of liver-targeted delivery of ADSCs was observed by a small animal in vivo imaging system. The data were compared by independent sample t test or one-way analysis of variance. Results:The GelMA-Hep microspheres were prepared by modifying the GelMA microspheres with Hep. Compared with the GelMA microspheres, the size of the GelMA-Hep microspheres did not change significantly, and the surface did not collapse and showed some crystalline particles. The binding energy of sulfur atoms on the surface of the GelMA-Hep microspheres increased from 166 eV to 168 eV. On the surface of the GelMA-Hep microspheres, the characteristic peaks of sulfonic acid and sulfate groups of Hep were detected at 1 490 cm ?1 and from 1 135 cm ?1 to 1 050 cm ?1, respectively. The swelling rate of the GelMA-dop microspheres was uniform, while the swelling rate of the GelMA microspheres and the GelMA-Hep microsphere was quite different, but the final swelling mass of the three microspheres tended to be consistent at 5 min. After 12, 24, 36 and 48 h of culture, the relative proliferation of cells in the GelMA-Hep group (1.61±0.29, 1.78±0.05, 2.27±0.08, 2.26±0.33) were higher than those in the negative control group (1.00±0.00, 1.28±0.06, 1.39±0.02, 1.41±0.04) (all P<0.05). After 36 h of culture, the relative proliferation of cells in the GelMA-Hep group was higher than that in the GelMA-dop group (1.63±0.21), with significant difference ( P<0.05). Live/dead cell staining experiment showed that after 12 h of cell culture in the GelMA-Hep group, only a few microspheres had cell adhesion; at 24 h, the cells were densely distributed on the surface of the microspheres. After 36 h, the number of cells increased further. At 48 h, live cells were distributed throughout the microspheres. Confocal microscopy showed that after 24 h of culture, cells adhered to the surface of the microspheres in the GelMA-Hep group and showed a stretched morphology. The liver of the GelMA-Hep+ADSCs group showed strong fluorescence at 0.5 h, and the fluorescence brightness continued to 48.0 h. The number of ADSCs reaching the liver was more than that of ADSCs group and GelMA+ADSCs group. Conclusions:GelMA-Hep microspheres were successfully prepared, which can improve the efficiency of liver-targeted delivery of ADSCs.

Non-functioning pituitary adenomas(NFPAs)are relatively common.Apart from hyperprolactinemia caused by pituitary compression,they typically lack overt hormonal hypersecretion and usually present with clinical symptoms due to mass effects.Previously considered a uniform entity,NFPAs are actually a highly heterogeneous group of tumors,including aggressive subtypes like silent corticotroph adenomas(SCA)and null cell adenomas.The 2022 WHO new classification of pituitary tumors employs transcription factors[e.g.,pituitary-specific transcription factor 1(PIT-1),T-box transcription factor 19(TBX19,also known as TPIT),steroidogenic factor 1(SF-1)]for detailed categorization,allowing precise subclassification of NFPAs into multiple subtypes derived from distinct cell lineages,including silent gonadotroph adenomas,SCA,and plurihormonal PIT-1-positive adenomas.This helps identify highly invasive subtypes with high recurrence risk,guiding clinical diagnosis and treatment,prognostic assessment,and individualized management.The new classification also provides a theoretical basis for targeted therapies of NFPAs(e.g.,somatostatin analogs and temozolomide).This review comprehensively discusses the latest pathological classification of NFPAs and its clinical implications,aiming to enhance understanding of this disease and offer valuable insights for precise diagnosis,treatment,and prognostic assessment.

Research on immune checkpoints such as programmed death protein-1(PD-1)and cytotoxic T lymphocyte antigen-4(CTLA-4)has provided new directions for tumor treatment.V-domain immunoglobulin suppressor of T-cell activation(VISTA)is an emerging immune checkpoint within the B7 family.Functioning as both a ligand and a receptor,VISTA is an inhibitory immune checkpoint molecule expressed in tumor cells,myeloid cells and T lymphocytes.It plays a crucial role in regulating autoimmunity,inflammatory response and tumor immunity.The non-redundant interactions between VISTA and other immune checkpoints,such as PD-1,may offer new therapeutic strategies and serve as a new target for overcoming immunotherapy resistance.This review summarizes the recent research progress on VISTA in hematological tumors,aiming to provide new insights into its application in the treatment of these malignancies.