Objective:To develop and evaluate a medical visual question answering (VQA) model for neonatal lung ultrasound (LUS) images to enhance intelligent auxiliary diagnosis of neonatal pulmonary diseases.Methods:Using data from neonates admitted to Beijing Obstetrics and Gynecology Hospital, Capital Medical University (January 2023 to December 2024), an image-question-answer dataset comprising 251 LUS images was constructed [43 pneumonia (17.1%), 42 neonatal respiratory distress syndrome (16.7%), 83 transient tachypnea (33.1%), and 83 normal (33.1%) images] with a four-tier medical question-answer framework. Building upon the Qwen2.5-VL-7B base model and integrating LoRA fine-tuning with chain-of-thought prompting, we developed the NLUS-VQA model to enhance visual-language semantic alignment and enable stepwise clinical reasoning, achieving efficient small-sample adaptation. Model performance was comprehensively assessed through natural language generation metrics (BLEU-4, ROUGE-1/2/L), qualitative evaluation of characteristic recognition, and clinical consistency analysis.Results:(1) Quantitative evaluation demonstrated that NLUS-VQA achieved scores of 22.38 (BLEU-4), 48.26 (ROUGE-1), 22.40 (ROUGE-2), and 37.20 (ROUGE-L), representing significant improvements over baseline models. (2) Qualitatively, the model exhibited strong performance in identifying lung consolidation, coalescent B-lines, and snowflake signs, with its chain-of-thought strategy enhancing clinical interpretability and answer accuracy. (3) Clinically, NLUS-VQA achieved a Cohen's Kappa coefficient of 0.78 and diagnostic accuracy of 80.8% (21/26), indicating substantial agreement with clinical experts.Conclusion:The NLUS-VQA model demonstrates robust interpretability in recognizing key sonographic patterns (e.g. lung consolidation, confluent B-lines, and snowflake signs), providing a scalable framework for small-sample medical image analysis, though diagnostic performance on complex conditions remains limited by dataset scale and minority class representation.

Objective:To illustrate the clinical modeling and commissioning results of Monte Carlo dose calculation algorithm in RayStation treatment planning system (TPS) for a domestically developed synchrotron-based spot scanning proton therapy system (SAPT-PS-01).Methods:The proton pencil beam model in RayStation required integral depth dose curves, spot profiles and absolute dose as the input beam data. It was not necessary to collect beam parameters with range shifter. The integral depth dose curves of a single spot were measured by an 8 cm parallel ion chamber. A 2-dimensional scintillation detector was used to measure the in-air spot profile at 5 different depths including the isocenter plane. The absolute dose was calibrated by a 0.25 cm parallel ion chamber under the single energy layer irradiation with a field size of 10 cm × 10 cm. After modeling, the results of the beam model and the Monte Carlo dose calculation algorithm were validated from the range, spot profile, point-dose in a spread-out Bragg peak, planar dose in a clinical plan, point dose in an end-to-end test.Results:For the 94 energy layers, the maximum deviation between the calculated and measured range was 0.03 cm. The maximum difference between the calculated and measured in-air spot sigma was 0.015 cm, and the deviation of in-water spot sigma was measured within ±15%. Compared with the measured values, the calculated dose deviation of 138 measured points in the spread-out Bragg peak was within 3%. For the planar dose verification of clinical plans, the TPS-calculated dose distribution of 285 planes agreed well with the measurement with a minimum gamma-passing rate of 90%, and the gamma passing rate of almost 95% of planes were greater than 95%. The point dose measurements for 8 beams in the end-to-end tests under 4 clinical scenarios were within 5%.Conclusions:The acceptable beam model validation results and successful end-to-end test confirm that the Monte Carlo dose calculation algorithm modeling for the synchrotron-based spot scanning proton therapy system is accurate, which is applicable for the design of patient treatment plan.

Children with congenital heart disease (CHD) experience intestinal dysbiosis, and an effective assessment of changes in their gut microbiota aids in optimizing screening, diagnosis, treatment, and risk monitoring for CHD.This review focuses on the microecological changes in the gut microbiota of CHD children and their impact on the development of CHD, aiming to provide new directions for the development of personalized therapies and interventions for CHD children.

Children with congenital heart disease (CHD) experience intestinal dysbiosis, and an effective assessment of changes in their gut microbiota aids in optimizing screening, diagnosis, treatment, and risk monitoring for CHD.This review focuses on the microecological changes in the gut microbiota of CHD children and their impact on the development of CHD, aiming to provide new directions for the development of personalized therapies and interventions for CHD children.

Objective:To develop and evaluate a medical visual question answering (VQA) model for neonatal lung ultrasound (LUS) images to enhance intelligent auxiliary diagnosis of neonatal pulmonary diseases.Methods:Using data from neonates admitted to Beijing Obstetrics and Gynecology Hospital, Capital Medical University (January 2023 to December 2024), an image-question-answer dataset comprising 251 LUS images was constructed [43 pneumonia (17.1%), 42 neonatal respiratory distress syndrome (16.7%), 83 transient tachypnea (33.1%), and 83 normal (33.1%) images] with a four-tier medical question-answer framework. Building upon the Qwen2.5-VL-7B base model and integrating LoRA fine-tuning with chain-of-thought prompting, we developed the NLUS-VQA model to enhance visual-language semantic alignment and enable stepwise clinical reasoning, achieving efficient small-sample adaptation. Model performance was comprehensively assessed through natural language generation metrics (BLEU-4, ROUGE-1/2/L), qualitative evaluation of characteristic recognition, and clinical consistency analysis.Results:(1) Quantitative evaluation demonstrated that NLUS-VQA achieved scores of 22.38 (BLEU-4), 48.26 (ROUGE-1), 22.40 (ROUGE-2), and 37.20 (ROUGE-L), representing significant improvements over baseline models. (2) Qualitatively, the model exhibited strong performance in identifying lung consolidation, coalescent B-lines, and snowflake signs, with its chain-of-thought strategy enhancing clinical interpretability and answer accuracy. (3) Clinically, NLUS-VQA achieved a Cohen's Kappa coefficient of 0.78 and diagnostic accuracy of 80.8% (21/26), indicating substantial agreement with clinical experts.Conclusion:The NLUS-VQA model demonstrates robust interpretability in recognizing key sonographic patterns (e.g. lung consolidation, confluent B-lines, and snowflake signs), providing a scalable framework for small-sample medical image analysis, though diagnostic performance on complex conditions remains limited by dataset scale and minority class representation.

Objective:To illustrate the clinical modeling and commissioning results of Monte Carlo dose calculation algorithm in RayStation treatment planning system (TPS) for a domestically developed synchrotron-based spot scanning proton therapy system (SAPT-PS-01).Methods:The proton pencil beam model in RayStation required integral depth dose curves, spot profiles and absolute dose as the input beam data. It was not necessary to collect beam parameters with range shifter. The integral depth dose curves of a single spot were measured by an 8 cm parallel ion chamber. A 2-dimensional scintillation detector was used to measure the in-air spot profile at 5 different depths including the isocenter plane. The absolute dose was calibrated by a 0.25 cm parallel ion chamber under the single energy layer irradiation with a field size of 10 cm × 10 cm. After modeling, the results of the beam model and the Monte Carlo dose calculation algorithm were validated from the range, spot profile, point-dose in a spread-out Bragg peak, planar dose in a clinical plan, point dose in an end-to-end test.Results:For the 94 energy layers, the maximum deviation between the calculated and measured range was 0.03 cm. The maximum difference between the calculated and measured in-air spot sigma was 0.015 cm, and the deviation of in-water spot sigma was measured within ±15%. Compared with the measured values, the calculated dose deviation of 138 measured points in the spread-out Bragg peak was within 3%. For the planar dose verification of clinical plans, the TPS-calculated dose distribution of 285 planes agreed well with the measurement with a minimum gamma-passing rate of 90%, and the gamma passing rate of almost 95% of planes were greater than 95%. The point dose measurements for 8 beams in the end-to-end tests under 4 clinical scenarios were within 5%.Conclusions:The acceptable beam model validation results and successful end-to-end test confirm that the Monte Carlo dose calculation algorithm modeling for the synchrotron-based spot scanning proton therapy system is accurate, which is applicable for the design of patient treatment plan.

Objective: To obtain chimeric antigen receptor macrophages ( CAR-M) targeting HER2 stably transfected． Methods : CAR lentivirus vector targeting HER2 was constructed and infected with human monocytic leukemia cell line (THP-1) ．CAR THP-1 cells with green fluorescent labeling were selected by sorting flow cytometry and continued to be cultured in vitro．The CAR THP-1 cells targeting HER2 were co-cultured with the endometrial cancer cell line Ishikawa with negative and positive HER2 expression，and their targeted phagocytosis of CAR-M to HER2 positive tumor cells was detected by imaging flow cytometry ，and the targeted phagocytosis efficiency of CAR-M to HER2 positive tumor cells was detected by flow cytometry. Results : CAR lentivirus infection with THP- 1 cells was less efficient ; After co-culture with cancer cells，flow cytometry and imaging flow cytometry showed that CAR THP-1 cells had enhanced phagocytosis of HER2 positive Ishikawa cells compared with the empty body group (P＜0. 01) ． Conclusion In this experiment，CAR THP-1 cell line targeting HER2 was established by constructing CAR lentivirus vector and transfecting THP-1 cells ，and it was proved that CAR THP-1 could phagocytize HER2 positive Ishikawa cells through specific targeting.

A method of high performance liquid chromatography coupled with time-of-flight mass spectrometry was used to detect the endogenous metabolites changes in plasma of normal rats, rats of dampness stagnancy due to spleen deficiency and Astragalus flavone component intervention rats. Metabolism map of rat plasma was obtained and the mechanism of Astragalus flavonoids on dampness stagnancy due to spleen deficiency was studied. Rat model with dampness stagnancy due to spleen deficiency was established by high fat and low protein diet plus load swimming. Liquid chromatography tandem mass spectrometry was used for the analysis of rat plasma sample, and 0.05％ formic acid water with 0.05％ formic acid acetonitrile as the mobile phase was applied in gradient elution with Halo C18 chromatographic column. In this study, partial least squares discriminant analysis and variance analysis were used to screen the potential biomarkers, it was found that the metabolic profile of the Astragalus flavonoids was different from that of the model group, which was close to that of the normal group. A total of 11 potential biomarkers were identified, including glycerol phospholipids, sphingolipids, amino acids, and so on. The metabolic pathways of biomarkers including three tricarboxylic acid cycle, glycerol phospholipid metabolism, fatty acid metabolism, amino acid metabolism and so on, which mainly related to energy metabolism and fat metabolism in the body. Related indexes of rats with syndrome of spleen deficiency of water and dampness were significantly callback after Astragalus flavone intervention, including macro indicators such as body weight, independent activities and micro indicators such as metabolic markers, blood lipids and others. The result showed that Astragalus flavonoids played the role of strengthening the spleen and draining the water mainly through regulating the energy metabolism, fat metabolism and so on.

An App for Android intelligent mobile phone to recite prescriptions in rhyme and learn infused pre-scriptions of traditional Chinese medicine was developed in order to meet the requirement of students majoring in traditional Chinese medicine and lovers of traditional Chinese medicine, which can show the assistant and guide of prescriptions in rhyme by different colors. An interface linked with each corresponding traditional Chinese medicine was added into the App to show its picture, properties, effect and application for the interaction between database server and mobile phone users and the aim to recite prescriptions of traditional Chinese medicine and learn tradi-tional Chinese medicine.

Objective To evaluate the residual risk (i.e.failure risk in detecting aneuploidies abnormalities except for chromosome 13,18,21,X and Y) of cytogenetic abnormalities using interphase fluorescence in situ hybridization (FISH) for the second-trimester amniocytes.Methods The results of interphase FISH and conventional karyotyping of 2 837 consecutive amniotic fluid specimens were analyzed retrospectively.Probes for chromosomes 13,18,21,X and Y were used.The detection rate and residual risk for interphase FISH were calculated for the following three major clinical indications for prenatal diagnosis (advanced maternal age,abnormal maternal serum screening indicating an increased risk for trisomy 18 or trisomy 21,and ultrasound abnormalities).Results Consecutive interphase FISH and karyotyping of second-trimester amniocytes for prenatal diagnosis were performed from January 1,2010 to July 31,2013.Among the 2 837 cases,85 (3.0％) cases with abnormal karyotypes were found,including 73 cases of aneuploidies involving chromosome 13,18,21,X and Y,which were considered detectable by interphase FISH; 12 cases of chromosomal anomalies,other than aneuploidies of chromosome 13,18,21,X and Y,were diagnosed after karyotyping and were not detected by interphase FISH,including six cases of balanced rearrangements,five cases of imbalanced rearrangements,and one case of pseudomosaic of trisomy 20.Of these 12 chromosomal anomalies,three cases of imbalanced rearrangements involving chromosome 21 showed positive FISH results,and the other nine cases showed negative FISH results among which four case of hereditary balanced rearrangemerts and two cases of novel balanced rearrangements.The total detection rate for interphase FISH was 89.4％ (76/85),the misdiagnosis rate of chromosome abnormalities was 14.1％(12/85),and the residual risk was 0.43％ (12/2 761) following interphase FISH of the second-trimester amniocytes.Conclusions Interphase FISH is a useful adjunct to conventional karyotyping,but should not be regarded as a replacement for karyotyping as too many structural chromosomal abnormalities will be missed.Providing patients with a detection rate and residual risk during counselling may help them understand the advantages and limitations of interphase FISH in their prenatal diagnostic evaluation.