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.

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.

Objective:To compare the the cultivation effects of human embryos in dry and humid incubators.Methods:A total of 479 infertile patients who underwent in vitro fertilization (IVF) treatment at Reproductive Center of Guiyang Maternal and Child Health Care Hospital from October 2020 to April 2022. The study was divided into two stages. The first stage of the study was a self-comparative research with 95 cases from the same period and source. The embryos were divided into dry and humid incubator groups to compare the embryo development indicators. In the second stage of the study, the patients were divided into six groups, including 10 μL humid incubator group ( n=64), 20 μL humid incubator group ( n=64), 30 μL humid incubator group ( n=64), 10 μL dry incubator group ( n=64), 20 μL dry incubator group ( n=64), and 30 μL dry incubator group ( n=64). The general clinical data, embryo development indicators, pregnancy outcomes, and the osmotic pressure and pH values of each group at 24 h, 48 h and 72 h were detected and compared. Results:After cultivation of the same patient's embryos in dry and humid incubator, the total blastocyst formation rate [62.3% (162/260)] and high-quality blastocyst rate [24.6% (64/260)] in dry incubator were lower than those in the humid incubator [71.6% (252/352), P=0.015; 32.1% (113/352), P=0.043]. Compared with the other microdroplet groups, the osmotic pressure of cleavage culture medium in 10 μL group of dry incubator at 48 h and 72 h and blastocyst culture medium were significantly increased, the differences among the groups were significant (cleavage culture medium, all P<0.001; blastocyst culture medium, P=0.006, P=0.008). There was no significant difference in pH value among different microdroplet volume groups at the same period (all P>0.05). There were no significant differences in general data among the different microdroplet groups (all P>0.05). Compared with the other microdroplet groups, 10 μL dry incubator group exhibited significantly lower transferable embryo rate (all P<0.001). When compared with 20 μL and 30 μL groups in both dry and humid incubators, 10 μL dry incubator group showed a lower day 5 blastocyst formation rate, lower total blastocyst formation rate, and lower high-quality blastocyst formation rate, the differences among the groups were significant (all P<0.05). There were no significant differences in the number of transferred embryos, the ratio of cleavage-stage embryos and the ratio of high-quality embryos among different groups (all P>0.05). Compared with the other microdroplet groups, the clinical pregnancy rate, the embryo implantation rate, the live birth rate of fresh transplanted embryos and the cumulative pregnancy rate in 10 μL group in the dry incubator decreased, and the miscarriage rate increased, but all were not significant (all P>0.05). Conclusion:Compared with humid incubators, there are no significant differences in embryo development and pregnancy outcomes for droplet volumes of 20 μL or above in dry incubators. However, the 10 μL microdroplet culture in the dry incubator is not conducive to embryonic development, which may be related to the increased osmotic pressure of the microdroplet.

Descurainiae Semen Lepidii Semen, also known as Tinglizi, originates from Brassicaceae plants Descurainia sophia or Lepidium apetalum. The former is commonly referred to as "Southern Tinglizi(Descurainiae Semen)", while the latter is known as "Northern Tinglizi(Lepidii Semen)". To scientifically and accurately identify the origin of Tinglizi medicinal materials and traditional Chinese medicine products, this study developed a specific DNA mini-barcode based on chloroplast genome sequences. By combining the DNA mini-barcode with DNA metabarcoding technology, a method for the qualitative and quantitative identification of Tinglizi medicinal materials and Chinese patent medicines was established. In this study, chloroplast genomes of Southern Tinglizi and Northern Tinglizi and seven commonly encountered counterfeit products were downloaded from the GenBank database. Suitable polymorphic regions were identified to differentiate these species, enabling the development of the DNA mini-barcode. Using DNA metabarcoding technology, medicinal material mixtures of Southern and Northern Tinglizi, as well as the most common counterfeit product, Capsella bursa-pastoris seeds, were analyzed to validate the qualitative and quantitative capabilities of the mini-barcode and determine its minimum detection limit. Additionally, the mini-barcode was applied to Chinese patent medicines containing Tinglizi to authenticate their botanical origin. The results showed that the developed mini-barcode(psbB) exhibited high accuracy and specificity, effectively distinguishing between the two authentic origins of Tinglizi and commonly encountered counterfeit products. The analysis of mixtures demonstrated that the mini-barcode had excellent qualitative and quantitative capabilities, accurately identifying the composition of Chinese medicinal materials in mixed samples with varying proportions. Furthermore, the analysis of Chinese patent medicines revealed the presence of the adulterant species(Capsella bursa-pastoris) in addition to the authentic species(Southern and Northern Tinglizi), indicating the occurrence of adulteration in commercially available Tinglizi-containing products. This study developed a method for the qualitative and quantitative identification of multi-origin Chinese medicinal materials and related products, providing a model for research on other multi-origin Chinese medicinal materials.
DNA Barcoding, Taxonomic/methods* ; Drugs, Chinese Herbal/chemistry* ; Drug Contamination ; Genome, Chloroplast ; Medicine, Chinese Traditional

Objective:To explore the longitudinal effect of learning stress on learning burnout in vocational college students, and the mediating role of academic procrastination.Methods:A total of 1 212 vocational college students were selected, and two follow-up surveys were conducted at 12-week intervals in September (T1) and November (T2) of 2024 using the basic situation questionnaire, the burnout inventory-student survey, the learning pressure questionnaire and the brief academic procrastination scale. SPSS 26.0 software was used to compare the demographic characteristics of students' depersonalization using t test and single factor analysis of variance. Bootstrap was used to analyze the relationship among learning stress, academic procrastination and learning burnout. Results:The scores of learning stress at T1 and T2 for vocational college students were 14.47±3.52 and 14.52 ±3.50, the scores of academic procrastination at T1 and T2 were 27.14±9.07 and 27.21±9.04, and the scores of learning burnout T1 and T2 were 39.38±8.76 and 39.69±8.79.The t-test showed that the score of learning burnout at T1 of students aged 18 and below (36.70±8.72) was lower than students aged 18 above (40.15±8.63, t=-5.81, P<0.01). The score of learning burnout for liberal arts students at T1(40.82±8.54) was higher than that of science students (37.68±8.72, t=6.31, P<0.01). Single factor analysis of variance showed that the score of learning burnout for grade 1 students at T1(35.19±8.45) was lower than that of grade 2 students (41.33±7.98) and grade 3 students (38.92±9.88), and learning burnout score of grade 2 students at T1 was higher than that of grades 3 students ( F=61.59, P<0.01). The score of learning burnout for high-achieving students at T1(36.23±8.34) was lower than that of middle-achieving students (39.82±8.52) and low-achieving students (45.42±9.14), and the score of learning burnout for middle-achieving students at T1 were lower than that of low-achieving students ( F=36.53, P<0.01). Bootstrap test showed that academic procrastination T2 played a partial mediating role in the relationship between learning stress T1 and learning burnout T2 (effect size=0.04, 95% CI=0.03-0.07). Academic procrastination T1 played a partial mediating role in the relationship between learning stress T1 and learning burnout T2 (effect size=0.05, 95% CI=0.04-0.07). Conclusion:Learning stress can directly affect learning burnout in vocational college students, and also can indirectly affect learning burnout through the mediating effect of academic procrastination.

Objective To investigate the effect of monitor unit(MU)optimization on volumetric modulated arc therapy(VMAT)for nasopharyngeal carcinoma(NPC).Methods Totally 21 NPC patients confirmed pathologically at some hospital from October 2022 to March 2023 were selected retrospectively,and for each patient a double-arc VMAT plan named base was designed without using the monitor unit objective(MUO)tool and optimized by the photon optimization algorithm.With the parameter kept constant,Plan base was reoptimized with only the MUO tool by regulating its Maximum MU to 30％of the total MU of Plan base and setting Strength values as 50,80 and 100 respectively to obtain Plan S50,S80 and S100.Plan base was compared with Plan S50,S80 and S100 in terms of dose distributions of the target volumes and organs at risk(OARs),MU,beam-on time and γ pass rate.SPSS 17.0 software was used for statistical analysis.Results In Plan S100 only 4 patients had a decrease of more than 4％in theD98％to the PCTV2 target volume compared with that in Plan base.In Plan S80 only 2 patients had a decrease of between 3％and 4％in the Dmax to PGTV and PGTVnd and D98％to PCTV2 when compared with those in Plan base,and most of the dosimetric parameters to the other target volumes had the changing amplitudes lower than 2％.When compared with Plan base,Plan S50,S80 and S100 all increased theDmaxto the brainstem significantly(all P＜0.05),with the maximum increment not exceeding 1.5％.In Plan S50 the D502％and Dmean to the right and left parotid glands were lowered,and in Plan S80 there were decreases found in the Dmean to the right and left parotid glands andD50％to the left parotid gland,with statistically significant differences(P＜0.001),and in Plan S100 the D50％and Dmean to the right and left parotid glands and the Dmax to the spinal cord increased significantly when compared with those in Plan base(all P＜0.05).In Plan S80 and S100 the V40 of the thyroid relatively rose statistically when compared with that in Plan base(P＜0.05).In Plan S50,S80 and S100 the MU was decreased by 5.1％,21.5％and 30.9％respectively when compared with that in Plan base,and the average beam-on time of all the four plans was kept within 2.5 min.γ pass rates increased sequentially for Plan base,S50,S80 and S100 under 1％/1 mm conditions.Conclusion In NPC VMAT MU optimization effectively decreases the planned MU and improves γ pass rates under 1％/1 mm conditions while ensuring the dose to the target volume and OAR protection.[Chinese Medical Equipment Journal,2025,46(2):56-62]

Nonunion of osteoporotic vertebral fractures (OVF), predominantly affecting the elderly, can lead to intractable pain, vertebral collapse, progressive kyphotic deformity, and neurological impairment, significantly compromising patients′ quality of life. There exists considerable debate on diagnosis and management of OVF, encompassing key issues such as clinical diagnosis and staging criteria for nonunion, surgical indications and procedure selection, and postoperative rehabilitation planning. Currently, there lacks standardized clinical guideline and expert consensus on the diagnosis and management of OVF nonunion in China. To address this gap, Minimally Invasive Surgery Group of Chinese Orthopedic Association, Osteoporosis Committee of Chinese Association of Orthopedic Surgeons, Prevention and Rehabilitation Committee for Osteoporosis of Chinese Association of Rehabilitation Medicine and Minimally Invasive Orthopedic Surgery Branch of China Association for Geriatric Care jointly organized domestic experts in spinal surgery, endocrinology, and rehabilitation to formulate the Clinical guideline for the diagnosis and treatment for nonunion of osteoporotic vertebral fractures ( version 2025), based on existing literature and clinical experience and adhering to principles of scientific rigor and practicality. The guideline provided 13 evidence-based recommendations encompassing diagnosis and treatment of OVF nonunion, aiming to standardize its clinical management.

Acute symptomatic osteoporotic thoracolumbar compression fracture (ASOTLF) can lead to chronic low back pain, kyphosis deformity, pulmonary dysfunction, loss of mobility, and even life-threatening complications. Vertebral augmentation is currently the mainstream treatment method for this condition. In 2019, the Editorial Board of Chinese Journal of Trauma and the Spinal Trauma Group of Orthopedic Surgeons Branch of Chinese Medical Doctor Association collaboratively led the development of Clinical guideline for vertebral augmentation for acute symptomatic osteoporotic thoracolumbar compression fractures. Six years later, with advances in clinical diagnosis and treatment techniques as well as accumulating evidence in related fields, the 2019 guideline requires updating. To this end, the Spinal Trauma Group of Orthopedic Surgeons Branch of Chinese Medical Doctor Association, the Spinal Health Professional Committee of China Human Health Science and Technology Promotion Association, and the Minimally Invasive Orthopedics Professional Committee of Shaanxi Medical Doctor Association have organized experts in the field to develop the Clinical guideline for vertebral augmentation of acute symptomatic osteoporotic thoracolumbar compression fractures ( version 2025) , based on the latest evidence-based medical researches. This guideline incorporates 3 recommendations retained from the 2019 version with updated strength of evidence, along with 12 new recommendations. It provides recommendations from six aspects of diagnosis, pain management, treatment option selection, prevention of postoperative complications, anti-osteoporosis therapy, and postoperative rehabilitation, aiming to provide a reference for standard treatment of vertebral augmentation for ASOTLF in hospitals at all levels.

Nonunion of osteoporotic vertebral fractures (OVF), predominantly affecting the elderly, can lead to intractable pain, vertebral collapse, progressive kyphotic deformity, and neurological impairment, significantly compromising patients′ quality of life. There exists considerable debate on diagnosis and management of OVF, encompassing key issues such as clinical diagnosis and staging criteria for nonunion, surgical indications and procedure selection, and postoperative rehabilitation planning. Currently, there lacks standardized clinical guideline and expert consensus on the diagnosis and management of OVF nonunion in China. To address this gap, Minimally Invasive Surgery Group of Chinese Orthopedic Association, Osteoporosis Committee of Chinese Association of Orthopedic Surgeons, Prevention and Rehabilitation Committee for Osteoporosis of Chinese Association of Rehabilitation Medicine and Minimally Invasive Orthopedic Surgery Branch of China Association for Geriatric Care jointly organized domestic experts in spinal surgery, endocrinology, and rehabilitation to formulate the Clinical guideline for the diagnosis and treatment for nonunion of osteoporotic vertebral fractures ( version 2025), based on existing literature and clinical experience and adhering to principles of scientific rigor and practicality. The guideline provided 13 evidence-based recommendations encompassing diagnosis and treatment of OVF nonunion, aiming to standardize its clinical management.