Objective To evaluate the performance of lightweight Chinese large language models (LLMs) in answering specialized lung cancer questions, and to explore the impact of retrieval-augmented generation (RAG) on model performance. Methods Eleven lightweight Chinese LLMs with parameter sizes ranging from 7B to 32B were included. A lung cancer-specific evaluation dataset consisting of 200 questions [100 A1-type (basic knowledge) and 100 A2-type (clinical case) questions], constructed based on clinical guidelines and thoracic surgery textbooks, was used for assessment. Model performance was evaluated under two conditions (with and without RAG). Accuracy was used to assess model performance, and response latency was recorded to reflect inference efficiency. An accuracy–latency scatter plot was constructed for descriptive analysis of overall model performance. Results All models successfully completed the evaluation. With the introduction of RAG, the overall average accuracy improved from 61.68% to 76.36%. Smaller models demonstrated the most significant improvement (e.g., the accuracy of DeepSeek-7B increased from 32.50% to 60.00%, P<0.001). The average response latency increased from 12.58 s to 13.80 s. The Qwen3 series showed the best overall performance, and Qwen3-32B achieved the highest accuracy under both conditions (76.50% and 84.00%, respectively). After RAG integration, performance differences among model families were markedly reduced. Based on the accuracy-latency trade-off, Qwen3-32B achieved the best balance between accuracy and response latency under the baseline condition, whereas Qwen3-14B demonstrated superior overall performance in terms of accuracy, latency, and computational cost after RAG integration. Conclusion The integration of RAG technology improves the ability of lightweight Chinese LLMs to answer specialized lung cancer questions. Under the dual practical constraints of limited computational resources and medical data security requirements, the "lightweight model+RAG" technical framework may represent a promising deployment solution.

Objective:To explore the effects of combining APR-246 with irradiation for enhancing anti-tumor immune response against 4T1 breast cancer cells, and to develop multiple tumor treatment strategies.Methods:The control group, APR-246 group, irradiation group and irradiation combined APR-246 group were used both in the cell experiment and tumor-bearing mice experiment. The inhibitory effect of APR-246 on the proliferation of 4T1 cells was assessed by using Cell Counting Kit-8. The effect of APR-246 with irradiation on the survival rate of 4T1 cells using clone formation assay was measured. The levels of reactive oxygen species (ROS) and lipid peroxidation (LPO) in tumor cells using a 2’, 7’-dichlorodihydrofluorescein diacetate (DCFH-DA) fluorescent probe and a lipid peroxidation sensor, the tumor inhibition rates of different groups of tumor bearing mice were compared, and the proportions of CD4+ and CD8+ T cells and the ratio of M1/M2 macrophages were determined in the tumor microenvironment by flow cytometry.Results:Compared with irradiation group, 2, 4, 6 Gy irradiation combined APR-246 group significantly reduced the survival rates of 4T1 cells ( t = 2.89, 4.15, 2.62, P < 0.05), the 6 Gy irradiation combined APR-246 group significantly increased the levels of ROS ( t = 16.95, P < 0.05) and LPO ( t = 6.09, P < 0.05) in 4T1 cells, and significantly increased the apoptosis rate of 4T1 cells ( t = 10.99, P < 0.05). Meanwhile, from the 16 th day of tumor inoculation, the 10 Gy irradiation combined APR-246 group showed significantly inhibited tumor growth ( t = 2.38-2.91, P < 0.05) and significantly increased proportions of CD4+ and CD8+ T cells ( t = 9.96, 6.28, P < 0.05) and M1/M2 ratio ( t = 15.30, P < 0.05) in tumor tissues. Conclusions:APR-246 combined with irradiation can effectively increase ROS and LPO levels in 4T1 cells, promote tumor cell apoptosis, and induce anti-tumor immune response, thus potentially inhibiting the growth of 4T1 cells.

Objective:To explore the protective effects and mechanisms of an indole-3-acetaldehyde (I3A)-loaded inulin-based hydrogel against radiation-induced intestinal injury.Methods:The gelation properties and injectability of the I3A-loaded inulin-based hydrogel were detected using a rheometer, and its biocompatibility was assessed via a CCK-8 assay. Eighteen C57BL/6 mice (aged: 6-8 weeks) were stratified by body weight and randomly assigned into three groups with 6 mice in each group: blank control, irradiation-only, and irradiation+ hydrogel protection. Abdominal irradiation was administered using 137Cs γ-rays at 17 Gy. The irradiation+ hydrogel protection group received 200 μl/day of I3A-loaded inulin-based hydrogel for two days before and 2-3 days after irradiation. Meanwhile, the irradiation-only group was treated with an equivalent volume of sterile water via gavage. The mice were euthanized four days post-irradiation, and their intestinal tissues were harvested. Hematoxylin-eosin (HE) staining, Ki67 immunohistochemistry, and TUNEL immunofluorescence were performed to assess histopathological damage, epithelial cell proliferation, and apoptosis, respectively. Quantitative real-time PCR (qRT-PCR) was employed to measure mRNA levels of inflammatory and antioxidant factors. Gut microbiota composition was analyzed via 16S rRNA sequencing. Results:The test results of the rheometer confirmed successful hydrogel formation. CCK-8 assays demonstrated excellent biocompatibility. Compared with the irradiation-only group, the irradiation+ hydrogel protection group exhibited preserved intestinal histoarchitecture, a 1.5-fold increase in intestinal cell proliferation ( t = 8.35, P < 0.05), and a 2-fold reduction in radiation-induced apoptosis ( t = 7.94, P < 0.05). Moreover, the hydrogel group showed significantly elevated expression of the anti-inflammatory cytokine IL-10 and antioxidant factors NRF-2 and HO-1 ( t = 3.16, 24.83, 5.92, P < 0.05), alongside reduced levels of pro-inflammatory cytokines IL-1β, IL-6, and TNF-α ( t = 5.15, 3.82, 3.83, P < 0.05). Gut microbiota analysis revealed significant modulation in microbial composition and abundance in the hydrogel group. Conclusions:The I3A-loaded inulin-based hydrogel can significantly promote intestinal cell proliferation, reduce radiation-induced apoptosis, and enhance both anti-inflammatory and antioxidant responses. In addition, it regulates gut microbiota composition and abundance, protecting against radiation-induced intestinal injury.

Objective:To reveal the differences in the transcriptome maps of brain tissues in mice subjected to Flash irradiation and conventional dose rate irradiation with electron beams and to explain the biological effect and mechanisms of Flash irradiation from multiple perspectives.Methods:Following the principle of grouping based on approximate body weights, 36 female C57BL/6J mice were divided into three groups, i. e., the control, conventional dose rate irradiation (CONV), and Flash irradiation (Flash) groups, with 12 mice in each group. Both the CONV and Flash groups received a single 15 Gy whole-brain irradiation with 9 MeV electron beams. At 3 d post-irradiation, the whole-brain tissue specimens were collected for hematoxylin-eosin (HE) staining to observe pathological changes. At 1, 3, and 10 weeks post-irradiation, the motion function, cognitive ability, depression level, and spatial memory capacity of the mice were assessed using ethology. At 1 and 10 weeks after behavioral experiments, brain tissue samples were collected and snap-frozen in liquid nitrogen for reference-based transcriptome sequencing. Accordingly, the differences in the transcriptome maps of radiation-induced brain injury between CONV and Flash groups were analyzed.Results:The HE staining-based pathological result revealed that compared to the CONV group, the Flash group exhibited reduced glial cell hyperplasia and inflammatory cell infiltration in brain tissues. Ethological research result at 1 week post-irradiation showed that the CONV group manifested a significantly decreased total traveled distance compared to the control and Flash groups ( t = 5.51, 2.38, P < 0.05) and a significantly increased immobility time compared to the control group ( t = 3.60, P < 0.05). Ethological research result at 3 weeks post-irradiation indicated that compared to the CONV group, the Flash group displayed significantly alleviated cognitive impairment ( t = 3.35, P < 0.05) and reduced depression levels ( t = 2.39, P < 0.05). Ethological research result at 10 weeks post-irradiation demonstrated that the CONV group showed the worst cognitive performance, significantly differing from the control group ( t = 4.53, P < 0.05). Transcriptome sequencing result revealed that besides immune-related pathways, the Flash group also exhibited multiple upregulated metabolic pathways and fibroblast growth factor (FGF)-related pathways compared to the CONV group. Conclusions:Compared to conventional dose rate irradiation, Flash irradiation can effectively alleviate radiation-induced brain injury in mice. This effect is associated with various metabolic pathways (including amino acid metabolism) and FGF-related pathways besides immune pathways.

Body weight abnormalities, including overweight, obesity, and underweight, have become a dual public health challenge in Chinese adults: overweight and obesity lead to a variety of chronic complications, while underweight increases the risks of malnutrition, sarcopenia, and organ dysfunction. To systematically address these issues, multidisciplinary experts in endocrinology, sports science, nutrition, and psychiatry from various regions have held multiple weight management seminars. Based on the latest epidemiological data and clinical evidence, they expanded the guideline to include assessment and intervention strategies for underweight, in addition to the core content of obesity management. This guideline outlines the etiological mechanisms, evaluation methods, and multidimensional management strategies for overweight and obesity, covering key areas such as diagnosis and assessment, medical nutrition therapy, exercise prescription, pharmacological intervention, and psychological support. It is intended to provide a scientific and standardized approach to weight management across the adult population, aiming to curb the rising prevalence of obesity, mitigate complications associated with abnormal body weight, and improve nutritional status and overall quality of life.

Aim To investigate the effect of chrysoeriol on pulmonary vascular remodeling in pulmonary hyper-tension by animal experiments combined with cell ex-periments,and to explore its potential therapeutic tar-gets by network pharmacology.Methods The target of chrysoeriol was collected in Targetnet,SEA and SwissTargetPrediction database.Pulmonary arterial hy-pertension(PAH)targets were collected in the Dis-GeNET and GeneCards databases,and PPI network map was drawn in the STRING database,and key tar-gets were screened.The GO and KEGG pathway en-richment analysis was carried out through DAVID data-base and Weishengxing platform.AutoDock software was used for molecular docking of key core targets.The PAH model of rats was constructed,and the pulmo-nary hemodynamics and vascular remodeling were de-tected by echocardiography,HE and Masson staining.Primary pulmonary smooth muscle cells were extracted,and the effects of drugs on pathway proteins were de-tected in vitro.Results The results of network phar-macology showed that chrysoeriol exerted therapeutic effects on pulmonary hypertension by affecting key tar-gets such as AKT1,SRC,EGFR,MMP9 and gsk3 β,and signaling pathways such as EGFR and PI3K-AKT.Molecular docking showed that chrysoeriol had good binding ability with 5 key target genes.Animal experi-ments showed that the pulmonary hemodynamic func-tion of PAH rats was significantly improved after ad-ministration of chrysoeriol.The remodeling of small pulmonary arteries was significantly reduced.Cell ex-periments showed that chrysoeriol could inhibit the ex-pression of proliferation,migration and phenotypic transformation genes.Conclusion Chrysoeriol may play a role in the treatment of pulmonary hypertension through multiple targets.

Objective To observe the value of 2D SECara-Net and 3D U2-Net models constructed based on 2D maximal intensity projection(MIP)and 3D time-of-flight MR angiography(3D TOF-MRA)images,respectively,also of their combination for MRA detecting unruptured saccular intracranial aneurysms(USIA).Methods Totally 973 patients with single USIA and 300 subjects who underwent healthy physical examination were retrospectively collected and divided into training set(n=923,containing 723 cases of USIA and 200 healthy subjects)and test set(n=350,containing 250 cases of USIA and 100 healthy subjects)at the ratio of 7:3.Pre-processed 3D TOF-MRA and the obtained 2D-MIP images in training set were imported into 3D U2-Net and 2D SECara-Net models for training and adjusting parameters,respectively.The efficiency of 2 models and their combination for detecting USIA were evaluated.Results The sensitivity,specificity and accuracy of 2D SECara-Net model for detecting USIA in test set was 78.80％(197/250),95.00％(95/100)and 83.43％(292/350),of 3D U2-Net model was 82.80％(207/250),86.00％(86/100)and 83.71％(293/350),respectively.The specificity of 2D SECara-Net model was higher than that of 3D U2-Net model(P=0.030),while no significant difference of sensitivity nor accuracy was found between 2 models(both P＞0.05).The specificity of the combination of the 2 models was 99.00％(99/100),higher than that of 3D U2-Net model(P＜0.05),and the sensitivity and accuracy of the combination was 91.20％(228/250)and 93.43％(327/350),respectivelty,both higher than those of 2 single models(all P＜0.05).Conclusion 2D SECara-Net and 3D U2-Net models had similar,sensitivity and accuracy for MRA detecting USIA.Combination of them could improve the detecting efficacy.

Aim To investigate the effect of chrysoeriol on pulmonary vascular remodeling in pulmonary hyper-tension by animal experiments combined with cell ex-periments,and to explore its potential therapeutic tar-gets by network pharmacology.Methods The target of chrysoeriol was collected in Targetnet,SEA and SwissTargetPrediction database.Pulmonary arterial hy-pertension(PAH)targets were collected in the Dis-GeNET and GeneCards databases,and PPI network map was drawn in the STRING database,and key tar-gets were screened.The GO and KEGG pathway en-richment analysis was carried out through DAVID data-base and Weishengxing platform.AutoDock software was used for molecular docking of key core targets.The PAH model of rats was constructed,and the pulmo-nary hemodynamics and vascular remodeling were de-tected by echocardiography,HE and Masson staining.Primary pulmonary smooth muscle cells were extracted,and the effects of drugs on pathway proteins were de-tected in vitro.Results The results of network phar-macology showed that chrysoeriol exerted therapeutic effects on pulmonary hypertension by affecting key tar-gets such as AKT1,SRC,EGFR,MMP9 and gsk3 β,and signaling pathways such as EGFR and PI3K-AKT.Molecular docking showed that chrysoeriol had good binding ability with 5 key target genes.Animal experi-ments showed that the pulmonary hemodynamic func-tion of PAH rats was significantly improved after ad-ministration of chrysoeriol.The remodeling of small pulmonary arteries was significantly reduced.Cell ex-periments showed that chrysoeriol could inhibit the ex-pression of proliferation,migration and phenotypic transformation genes.Conclusion Chrysoeriol may play a role in the treatment of pulmonary hypertension through multiple targets.

Objective To investigate the therapeutic effects of upper limb rehabilitation robot training combined with intermittent theta burst stimulation(iTBS)on upper limb motor and neurological function in stroke patients with hemiplegia.Methods This study retrospectively consecutive enrolled 46 stroke hemiparetic patients from the Department of Rehabilitation Medicine,Nanjing Pukou People's Hospital.The patients were randomly assigned to a control group and an experimental group(23patients in each)using a random number table.Baseline data,including sex,age,disease duration,side of hemiplegia,and stroke type,were collected from patients enrolled.All patients received conventional treatment.The control group received upper limb rehabilitation robot training combined with iTBS sham stimulation(coil placed perpendicular to the skull),while the experimental group received upper limb rehabilitation robot training combined with iTBS real stimulation(coil placed parallel to the skull).Both groups underwent treatment for 3 weeks.Upper limb motor function was assessed using the Fugl-Meyer upper extremity(FMA-UE)scale and Wolf motor function test(WMFT);while neurological function was evaluated using the motor-evoked potentials(MEP)latency,amplitude,and central motor conduction time(CMCT)of the affected thumb abductor muscle.Activities of daily living were assessed using the modified Barthel index(MBI).Results(1)No significant differences in baseline data were found between the two groups(all P＞0.05).(2)Before treatment,the FMA-UE and WMFT scores in the experimental group were 27.48±7.87 and 28.22±3.87,respectively;and in the control group were 26.35±4.78 and 28.35±3.33,respectively;there were no significant differences in both FMA-UE and WMFT scores between the two groups(all P＞0.05).After 3weeks of treatment,the FMA-UE and WMFT scores in the experimental group were 40.35±8.96 and 37.74±4.11,respectively;and in the control group were 32.78±4.50 and 32.57±4.11,respectively;there were significant interaction effects of time and group(Ftime×group values of 19.613 and 31.522,both P＜0.01),main effects of group(Fgroup values of 5.401 and 5.897,both P＜0.05),and main effects of time(Ftime values of 176.516 and 211.478,both P＜0.01).(3)Before treatment,the MEP latency,amplitude,and CMCT in the experimental group were(24.39±3.56)ms,(137.77±42.67)μV,and(10.62±2.76)ms,respectively;and in the control group were(24.64±2.77)ms,(136.74±48.77)μV,and(10.73±1.84)ms,respectively,there were no significant differences between the two groups(all P＞0.05).After 3weeks of treatment,the MEP latency,amplitude,and CMCT in the experimental group were(20.39±1.83)ms,(239.91±43.70)μV,and(6.58±1.23)ms,respectively,and in the control group were(22.53±3.53)ms,(198.54±50.37)μV,and(9.19±1.60)ms,respectively,there were significant interaction effects of time and group(Ftime×group values of 7.270,15.554,and 20.110,all P＜0.05)and main effects of time(Ftime values of 76.540,256.706,and 100.629,all P＜0.01),the main effect of group for CMCT was significant(Fgroup=7.406,P＜0.01),but there were no significant difference in the main effect of group on MEP latency,amplitude between two groups(Fgroup values of 2.145,2.778,both P＞0.05).(4)Before treatment,the MBI score in the experimental group was 42.83±7.36,and in the control group was 43.91±6.56,with no significant difference between two groups(P＞0.05).After 3 weeks of treatments,the MBI score in the experimental group was 67.83±12.69,and in the control group was 54.13±5.57,there were significant interaction effects of time and group(Ftime×group=39.862,P＜0.01),main effects of group(Fgroup=8.083,P=0.007),and main effects of time(Ftime=226.241,P＜0.01).Conclusions Upper limb rehabilitation robot training combined with iTBS can improve upper limb motor function and neurological function and enhance the daily living activity ability of stroke patients.Real iTBS combined with robot training has a more significant effect than sham iTBS.

Objective To observe the value of 2D SECara-Net and 3D U2-Net models constructed based on 2D maximal intensity projection(MIP)and 3D time-of-flight MR angiography(3D TOF-MRA)images,respectively,also of their combination for MRA detecting unruptured saccular intracranial aneurysms(USIA).Methods Totally 973 patients with single USIA and 300 subjects who underwent healthy physical examination were retrospectively collected and divided into training set(n=923,containing 723 cases of USIA and 200 healthy subjects)and test set(n=350,containing 250 cases of USIA and 100 healthy subjects)at the ratio of 7:3.Pre-processed 3D TOF-MRA and the obtained 2D-MIP images in training set were imported into 3D U2-Net and 2D SECara-Net models for training and adjusting parameters,respectively.The efficiency of 2 models and their combination for detecting USIA were evaluated.Results The sensitivity,specificity and accuracy of 2D SECara-Net model for detecting USIA in test set was 78.80％(197/250),95.00％(95/100)and 83.43％(292/350),of 3D U2-Net model was 82.80％(207/250),86.00％(86/100)and 83.71％(293/350),respectively.The specificity of 2D SECara-Net model was higher than that of 3D U2-Net model(P=0.030),while no significant difference of sensitivity nor accuracy was found between 2 models(both P＞0.05).The specificity of the combination of the 2 models was 99.00％(99/100),higher than that of 3D U2-Net model(P＜0.05),and the sensitivity and accuracy of the combination was 91.20％(228/250)and 93.43％(327/350),respectivelty,both higher than those of 2 single models(all P＜0.05).Conclusion 2D SECara-Net and 3D U2-Net models had similar,sensitivity and accuracy for MRA detecting USIA.Combination of them could improve the detecting efficacy.