Pulmonary fibrosis（PF） is a progressive and life-threatening disease with limited therapeutic options， highlighting the urgent need for innovative drug discovery strategies. To address this challenge， the authors propose the formula-originated rational intelligent screening&translation（FIRST）， a systematic framework for developing anti-fibrotic monomers derived from classical traditional Chinese medicine（TCM）. The strategy integrates three key dimensions， including tissue-oriented intelligent screening of active compounds， structural optimization based on drug-target spatial interactions and plant biosynthetic pathways， and cross-scale validation of drug. We further highlight its applications in discovering tissue-oriented novel drugs from clinically validated TCM， the development and mechanistic elucidation of anti-fibrotic therapeutics， as well as the clinical translation and secondary development of candidate drugs. This strategy paves the way for first-in-class， formula-derived monomeric drugs with defined structures， clarified mechanisms， and proven safety， offering a transformative avenue to meet the urgent therapeutic needs of PF and setting a new paradigm for TCM-based drug innovation.

Pulmonary fibrosis（PF） is a progressive and life-threatening disease with limited therapeutic options， highlighting the urgent need for innovative drug discovery strategies. To address this challenge， the authors propose the formula-originated rational intelligent screening&translation（FIRST）， a systematic framework for developing anti-fibrotic monomers derived from classical traditional Chinese medicine（TCM）. The strategy integrates three key dimensions， including tissue-oriented intelligent screening of active compounds， structural optimization based on drug-target spatial interactions and plant biosynthetic pathways， and cross-scale validation of drug. We further highlight its applications in discovering tissue-oriented novel drugs from clinically validated TCM， the development and mechanistic elucidation of anti-fibrotic therapeutics， as well as the clinical translation and secondary development of candidate drugs. This strategy paves the way for first-in-class， formula-derived monomeric drugs with defined structures， clarified mechanisms， and proven safety， offering a transformative avenue to meet the urgent therapeutic needs of PF and setting a new paradigm for TCM-based drug innovation.

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.

ObjectiveTo investigate the association of the single nucleotide polymorphism （SNP） of the insulin-like growth factor binding protein-3 （IGFBP3） gene at rs10225396 （A>G） locus with the susceptibility to hepatocellular carcinoma， as well as its potential molecular regulatory mechanisms， and to provide novel genetic biomarkers and a theoretical basis for early screening and precise targeted therapy for hepatocellular carcinoma. MethodsA total of 192 patients with hepatocellular carcinoma who were admitted to The Affiliated Hospital of Yanbian University and Yanbian Cancer Hospital from January 2009 to August 2016 were enrolled as experimental group， and 190 healthy individuals who underwent physical examination in Physical Examination Center of Yanbian Hospital during the same period of time were enrolled as control group. Peripheral blood samples were collected from all subjects， and after DNA was extracted from whole blood， the DNA samples meeting quality standards were sent to Beijing Genomics Institute Research Center Co.， Ltd.， for Mass ARRAY mass spectrometry， while genotyping was completed. The independent-samples t test was used for comparison of continuous data between two groups， and the chi-square test was used for comparison of categorical data between two groups. A binary Logistic regression model was used to analyze the association of the SNP of the IGFBP3 gene at rs10225396 locus with the susceptibility to hepatocellular carcinoma， and odds ratio （OR） and 95% confidence interval （CI） were calculated to assess the risk of developing hepatocellular carcinoma in individuals carrying different genotypes. ResultsThere were two alleles （G and A） at the rs10225396 locus of the IGFBP3 gene， yielding the three genotypes of AA， AG， and GG， and its genotype distribution was consistent with the Hardy-Weinberg equilibrium （χ²=0.072， P=0.789）. The stratified genetic analysis showed that carriers of the IGFBP3 rs10225396 AA genotype had a significant increased risk of hepatocellular carcinoma among individuals of age <63 years， male sex， smoking history， drinking history， and the Chinese Korean population （all P<0.001）. The binary Logistic regression analysis showed that after adjustment for related risk factors in the codominant model， the population with genotype AG and GG had a significant reduction in the risk of hepatocellular carcinoma compared with the population with genotype AA （P<0.001）， and in the dominant model， the population with genotype AG+GG had a significant reduction in the risk of hepatocellular carcinoma compared with the population with genotype AA （P<0.001）. ConclusionThrough a genotyping analysis of hepatocellular carcinoma patients and healthy individuals in Yanbian Korean Autonomous Prefecture of Jilin Province， China， this study shows that genotype AA at rs10225396 locus of the IGFBP3 gene is significantly associated with the susceptibility to hepatocellular carcinoma， and this genotype can significantly increase the risk of developing hepatocellular carcinoma with the presence of specific risk factors， which provides a potential genetic marker for early screening and precise treatment of hepatocellular carcinoma.

Based on the understanding of "kidney storing essence and marrow nourishing bone"， and in combination with the modern pathological characteristics of ankylosing spondylitis （AS）， this paper systematically elucidates the dynamic pathogenesis evolution of "heat bi （inflammatory activity）-marrow deficiency （bone marrow edema）-bone damage （bone structural damage）" in AS. It is proposed that in the heat bi stage， there is kidney deficiency with pathogenic invasion and heat congesting the meridians. In the marrow deficiency stage， heat consumes kidney essence， leading to marrow injury and collateral obstruction. In the bone damage stage， kidney exhaustion and marrow emptiness occur， with phlegm-stasis congealing on bone. Accordingly， a stage-specific treatment strategy for AS is established. In the heat bi stage， treatment focuses on supplementing the kidney and replenishing essence， clearing heat and unblocking collaterals. In the marrow deficiency stage， the main approach is to supplement the kidney and benefit marrow， unblock collaterals and reduce edema. In the bone damage stage， it is suggested to supplement the kidney and replenish essence， benefit marrow and strengthen bone， resolve phlegm and expel stasis.

Based on the understanding of "kidney storing essence and marrow nourishing bone"， and in combination with the modern pathological characteristics of ankylosing spondylitis （AS）， this paper systematically elucidates the dynamic pathogenesis evolution of "heat bi （inflammatory activity）-marrow deficiency （bone marrow edema）-bone damage （bone structural damage）" in AS. It is proposed that in the heat bi stage， there is kidney deficiency with pathogenic invasion and heat congesting the meridians. In the marrow deficiency stage， heat consumes kidney essence， leading to marrow injury and collateral obstruction. In the bone damage stage， kidney exhaustion and marrow emptiness occur， with phlegm-stasis congealing on bone. Accordingly， a stage-specific treatment strategy for AS is established. In the heat bi stage， treatment focuses on supplementing the kidney and replenishing essence， clearing heat and unblocking collaterals. In the marrow deficiency stage， the main approach is to supplement the kidney and benefit marrow， unblock collaterals and reduce edema. In the bone damage stage， it is suggested to supplement the kidney and replenish essence， benefit marrow and strengthen bone， resolve phlegm and expel stasis.

The Huangdi Neijing proposes the " using bitter herbs to nourish or purge" theory to guide clinical prescription and formulation of herbal remedies based on the physiological characteristics and functions of the five zang viscera, along with the properties and flavors of medicinal herbs. This study explored diabetic kidney disease pathogenesis and treatment based on the " using bitter herbs to nourish or purge" theory. Kidney dryness is a key pathological factor in diabetic kidney disease, and the disharmony of kidney dryness is an essential aspect of its pathogenesis. Strengthening is the primary therapeutic principle, and kidney dryness is a persistent factor throughout the occurrence and progression of diabetic kidney disease. In the early stage, the pathogenesis involves heat-consuming qi and injuring yin, leading to kidney dryness. In the middle stage, the pathogenesis manifests as qi deficiency and blood stasis in the collaterals, resulting in turbidity owing to kidney dryness. In the late stage, the pathogenesis involves yin and yang deficiency, with kidney dryness and disharmony. This study proposes the staging-based treatment based on the " need for firmness" characteristic of the kidney. The aim is to provide new insights for clinical diagnosis and treatment in traditional Chinese medicine by rationally using pungent, bitter, and salty medicinal herbs to nourish and moisturize the kidney. This approach seeks to promote precise syndrome differentiation and personalized treatment for different stages of diabetic kidney disease, thereby enhancing clinical efficacy.

Background: Normalized creatinine-to-cystatin C ratio (NCCR) was reported to approximate relative skeletal muscle mass and diabetes risk. However, the association between NCCR and cardiometabolic multimorbidity (CMM) remains elusive. This study aimed to explore their relationship in a large-scale prospective cohort. Methods: This study included 5,849 middle-age and older participants from the China Health and Retirement Longitudinal Study (CHARLS) enrolled between 2011 and 2012. The baseline NCCR was determined as creatinine (mg/dL)/cystatin C (mg/L)×10/body mass (kg). CMM was defined as the simultaneous occurrence of two or more of the following conditions: heart disease, stroke, and type 2 diabetes mellitus. Logistic regression analysis and Cox regression analysis were employed to estimate the relationship between NCCR and CMM. The joint effect of body mass index and NCCR on the risk of CMM were further analyzed. Results: During a median 4-year follow-up, 227 (3.9%) participants developed CMM. The risk of CMM was significantly decreased with per standard deviation increase of NCCR (odds ratio, 0.72; 95% confidence interval, 0.62 to 0.85) after adjustment for confounders (P<0.001). Further sex-specific analysis found significant negative associations between NCCR and CMM in female either without or with one CMM component at baseline, which was attenuated in males but remained statistically significant among those with one basal CMM component. Notably, non-obese individuals with high NCCR levels had the lowest CMM risk compared to obese counterparts with low NCCR levels in both genders. Conclusion High NCCR was independently associated with reduced risk of CMM in middle-aged and older adults in China, particularly females.

ObjectivePrimary liver cancer, predominantly hepatocellular carcinoma (HCC), is a significant global health issue, ranking as the sixth most diagnosed cancer and the third leading cause of cancer-related mortality. Accurate and early diagnosis of HCC is crucial for effective treatment, as HCC and non-HCC malignancies like intrahepatic cholangiocarcinoma (ICC) exhibit different prognoses and treatment responses. Traditional diagnostic methods, including liver biopsy and contrast-enhanced ultrasound (CEUS), face limitations in applicability and objectivity. The primary objective of this study was to develop an advanced, light-weighted classification network capable of distinguishing HCC from other non-HCC malignancies by leveraging the automatic analysis of brightness changes in CEUS images. The ultimate goal was to create a user-friendly and cost-efficient computer-aided diagnostic tool that could assist radiologists in making more accurate and efficient clinical decisions. MethodsThis retrospective study encompassed a total of 161 patients, comprising 131 diagnosed with HCC and 30 with non-HCC malignancies. To achieve accurate tumor detection, the YOLOX network was employed to identify the region of interest (ROI) on both B-mode ultrasound and CEUS images. A custom-developed algorithm was then utilized to extract brightness change curves from the tumor and adjacent liver parenchyma regions within the CEUS images. These curves provided critical data for the subsequent analysis and classification process. To analyze the extracted brightness change curves and classify the malignancies, we developed and compared several models. These included one-dimensional convolutional neural networks (1D-ResNet, 1D-ConvNeXt, and 1D-CNN), as well as traditional machine-learning methods such as support vector machine (SVM), ensemble learning (EL), k-nearest neighbor (KNN), and decision tree (DT). The diagnostic performance of each method in distinguishing HCC from non-HCC malignancies was rigorously evaluated using four key metrics: area under the receiver operating characteristic (AUC), accuracy (ACC), sensitivity (SE), and specificity (SP). ResultsThe evaluation of the machine-learning methods revealed AUC values of 0.70 for SVM, 0.56 for ensemble learning, 0.63 for KNN, and 0.72 for the decision tree. These results indicated moderate to fair performance in classifying the malignancies based on the brightness change curves. In contrast, the deep learning models demonstrated significantly higher AUCs, with 1D-ResNet achieving an AUC of 0.72, 1D-ConvNeXt reaching 0.82, and 1D-CNN obtaining the highest AUC of 0.84. Moreover, under the five-fold cross-validation scheme, the 1D-CNN model outperformed other models in both accuracy and specificity. Specifically, it achieved accuracy improvements of 3.8% to 10.0% and specificity enhancements of 6.6% to 43.3% over competing approaches. The superior performance of the 1D-CNN model highlighted its potential as a powerful tool for accurate classification. ConclusionThe 1D-CNN model proved to be the most effective in differentiating HCC from non-HCC malignancies, surpassing both traditional machine-learning methods and other deep learning models. This study successfully developed a user-friendly and cost-efficient computer-aided diagnostic solution that would significantly enhances radiologists’ diagnostic capabilities. By improving the accuracy and efficiency of clinical decision-making, this tool has the potential to positively impact patient care and outcomes. Future work may focus on further refining the model and exploring its integration with multimodal ultrasound data to maximize its accuracy and applicability.