ObjectiveTo investigate the mechanism of Huangqin Tang （HQT） in regulating metabolic reprogramming during the inflammation-cancer transformation in colitis-associated colorectal cancer （CAC）. MethodsCAC mouse model was established using the carcinogen azoxymethane （AOM） combined with the inflammatory agent dextran sulfate sodium （DSS）. HQT treatment was adopted. Serum metabolomics analysis was performed at three stages （inflammation， proliferation， and tumor formation） using liquid chromatography-tandem mass spectrometry （LC-MS/MS） untargeted metabolomics coupled with multivariate statistical analysis to explore the mechanism of HQT intervention in metabolism in CAC. ResultsThe results revealed that HQT significantly reversed the disturbance of key metabolites in CAC mice. A total of 52， 67， and 45 differential metabolites were identified in the model group， compared to the normal group， during inflammation， proliferation， and tumor stages， respectively. Lactate， linoleic acid， oleic acid， elaidic acid， and betaine were characteristic metabolites persistently enriched throughout colitis-cancer transformation. Pathway enrichment analysis of differential metabolites showed that linoleic acid metabolism and arachidonic acid metabolism were the most significantly disturbed in CAC pathogenesis. The proliferation stage featured expanded amino acid metabolic networks， while the tumor stage uniquely exhibited two new pathways of nicotinate and nicotinamide metabolism and phosphoinositide metabolism. HQT exerted stage-specific regulatory effects： targeting arachidonic acid metabolism in the inflammation stage， correcting the dysregulation of choline-carnitine metabolism in the proliferation stage， and rescuing nicotinamide and tryptophan metabolic collapse in the tumor stage. ConclusionHQT exerts regulatory effects on metabolic disorders at various stages of the colitis-cancer transformation process， thereby effectively slowing the progression from colitis to cancer. The study also reveals the dynamic metabolic characteristics of colorectal "inflammation-cancer transformation，"providing new insights for research on the targeted mechanisms of traditional Chinese medicine in anti-tumor therapy based on metabolic reprogramming.

Objective:To develop PixelSIFT-UNet, a novel semantic segmentation model that integrates deep learning with scale-invariant feature transform (SIFT) algorithm to improve the segmentation accuracy of oral mucosal lesions.Methods:This investigation utilized 838 standard clinical white light images of oral mucosal diseases acquired from January 2020 to December 2022 at the Stomatology Hospital Zhejiang University School of Medicine. Randomization was achieved through Python′s random.seed function implementation. The random sample function was subsequently applied for sampling distribution. The dataset was stratified into three subsets with a 6∶2∶2 ratio: training ( n=506), validation ( n=166), and testing ( n=166). Lesion boundaries were annotated using Labelme software, and a PixelSIFT-UNet-based deep learning model was developed with VGG-16 and ResNet-50 backbone networks. Model parameters were optimized using the validation set, and performance metrics [including Dice coefficient, mean intersection over union (mIoU), mean pixel accuracy (mPA), and Precision] were assessed on the test set. The model′s performance was benchmarked against conventional semantic segmentation frameworks (U-Net and PSPNet). Results:The developed PixelSIFT-UNet model could achieve precise segmentation of three common oral mucosal lesions: oral lichen planus, oral leukoplakia, and oral submucous fibrosis. Utilizing VGG-16 as the backbone network, the model achieved Dice coefficient, mIoU, mPA, and Precision values of 0.642, 0.699, 0.836, and 0.792, respectively. Implementation with ResNet-50 backbone network yielded metrics of 0.668, 0.733, 0.872 and 0.817, demonstrating significant improvements across all performance indicators compared to conventional U-Net model (relevant metrics: 0.662, 0.717, 0.861 and 0.809) and PSPNet model (relevant metrics: 0.671, 0.721, 0.858 and 0.813).Conclusions:The proposed PixelSIFT-UNet architecture demonstrates superior performance in oral mucosal lesion segmentation tasks, surpassing conventional semantic segmentation models and providing robust quantitative improvements in segmentation accuracy.

Objective To observe the clinical efficacy and safety of Kangyan Jingfang in the treatment of post-operative cancer-related fatigue in Stage ⅡB-Ⅲ colorectal cancer.Methods A total of 66 patients with stage ⅡB-Ⅲ colorectal cancer who were diagnosed as cancer-related fatigue(CRF)based on TCM syndrome differentiation as spleen vacuity and essence depletion were divided into treatment group and control group randomly,with 33 cases in each group.The treatment group was treated with conventional symptomatic treatment combined with Kangyan Jingfang,and the control group was treated with conventional symptomatic treatment,the course of treatment was 60 days.The scores of cancer-related Fatigue(CFS),TCM syndrome accumulation scale,the revised Piper fatigue scale,Karnofsky performance status,hospital anxiety and depression scale(HAD),tumor markers(CEA,CA19-9,CA72-4,CA12-5,CA15-3,CA242,CA50,Septin9 gene methylation,etc.),immune function(CD3+T lymphocyte level,CD4+T lymphocyte level,CD8+T lymphocyte level and CD4+/CD8+ratio)and safety indexes(blood routine,liver and kidney function)were compared between the two groups in order to evaluate the clinical efficacy of Kangyan Jingfang on CRF.Results 60 cases were included in the observation finally,including 30 cases in each group.After treatment,the scores CFS,TCM syndrome accumulation scale,the revised Piper fatigue scale and HAD of two groups were decreased,but the treatment group before treatment showed a significant decrease in all indicators compared to the control group(P<0.05).The KPS score in the treatment group improved after treatment compared to the control group significantly(P<0.05).CEA,CA19-9 and the methylation of Septin9 gene were significantly decreased in the treatment group after treatment(P<0.05).CD3 and CD4 in treatment groups were higher than control group after treatment significantly(P<0.05).There were no significant abnormal changes in safety indexes before and after treatment(P>0.05).Conclusion Kangyan Jingfang can effectively improve the clinical symptoms of cancer-related fatigue in patients with stage ⅡB-Ⅲ colorectal cancer after surgery,reduce the physical and emotional suffering of patients,improve the quality of life,and enhance the immune function of the body.The efficacy is safe and reliable,and it has certain clinical application value.

Objective:To develop PixelSIFT-UNet, a novel semantic segmentation model that integrates deep learning with scale-invariant feature transform (SIFT) algorithm to improve the segmentation accuracy of oral mucosal lesions.Methods:This investigation utilized 838 standard clinical white light images of oral mucosal diseases acquired from January 2020 to December 2022 at the Stomatology Hospital Zhejiang University School of Medicine. Randomization was achieved through Python′s random.seed function implementation. The random sample function was subsequently applied for sampling distribution. The dataset was stratified into three subsets with a 6∶2∶2 ratio: training ( n=506), validation ( n=166), and testing ( n=166). Lesion boundaries were annotated using Labelme software, and a PixelSIFT-UNet-based deep learning model was developed with VGG-16 and ResNet-50 backbone networks. Model parameters were optimized using the validation set, and performance metrics [including Dice coefficient, mean intersection over union (mIoU), mean pixel accuracy (mPA), and Precision] were assessed on the test set. The model′s performance was benchmarked against conventional semantic segmentation frameworks (U-Net and PSPNet). Results:The developed PixelSIFT-UNet model could achieve precise segmentation of three common oral mucosal lesions: oral lichen planus, oral leukoplakia, and oral submucous fibrosis. Utilizing VGG-16 as the backbone network, the model achieved Dice coefficient, mIoU, mPA, and Precision values of 0.642, 0.699, 0.836, and 0.792, respectively. Implementation with ResNet-50 backbone network yielded metrics of 0.668, 0.733, 0.872 and 0.817, demonstrating significant improvements across all performance indicators compared to conventional U-Net model (relevant metrics: 0.662, 0.717, 0.861 and 0.809) and PSPNet model (relevant metrics: 0.671, 0.721, 0.858 and 0.813).Conclusions:The proposed PixelSIFT-UNet architecture demonstrates superior performance in oral mucosal lesion segmentation tasks, surpassing conventional semantic segmentation models and providing robust quantitative improvements in segmentation accuracy.

Objective To observe the clinical efficacy and safety of Kangyan Jingfang in the treatment of post-operative cancer-related fatigue in Stage ⅡB-Ⅲ colorectal cancer.Methods A total of 66 patients with stage ⅡB-Ⅲ colorectal cancer who were diagnosed as cancer-related fatigue(CRF)based on TCM syndrome differentiation as spleen vacuity and essence depletion were divided into treatment group and control group randomly,with 33 cases in each group.The treatment group was treated with conventional symptomatic treatment combined with Kangyan Jingfang,and the control group was treated with conventional symptomatic treatment,the course of treatment was 60 days.The scores of cancer-related Fatigue(CFS),TCM syndrome accumulation scale,the revised Piper fatigue scale,Karnofsky performance status,hospital anxiety and depression scale(HAD),tumor markers(CEA,CA19-9,CA72-4,CA12-5,CA15-3,CA242,CA50,Septin9 gene methylation,etc.),immune function(CD3+T lymphocyte level,CD4+T lymphocyte level,CD8+T lymphocyte level and CD4+/CD8+ratio)and safety indexes(blood routine,liver and kidney function)were compared between the two groups in order to evaluate the clinical efficacy of Kangyan Jingfang on CRF.Results 60 cases were included in the observation finally,including 30 cases in each group.After treatment,the scores CFS,TCM syndrome accumulation scale,the revised Piper fatigue scale and HAD of two groups were decreased,but the treatment group before treatment showed a significant decrease in all indicators compared to the control group(P<0.05).The KPS score in the treatment group improved after treatment compared to the control group significantly(P<0.05).CEA,CA19-9 and the methylation of Septin9 gene were significantly decreased in the treatment group after treatment(P<0.05).CD3 and CD4 in treatment groups were higher than control group after treatment significantly(P<0.05).There were no significant abnormal changes in safety indexes before and after treatment(P>0.05).Conclusion Kangyan Jingfang can effectively improve the clinical symptoms of cancer-related fatigue in patients with stage ⅡB-Ⅲ colorectal cancer after surgery,reduce the physical and emotional suffering of patients,improve the quality of life,and enhance the immune function of the body.The efficacy is safe and reliable,and it has certain clinical application value.

In automatic skin damage analysis,segmentation is a challenging and critical operation due to factors such as the shape and contrast of hair and skin lesions on the skin.Compared with traditional segmentation methods,deep learning seamlessly integrates feature extraction and task-specific decision-making,achieving segmentation tasks more accurately and efficiently,and effectively reducing the burden and cost of skin cancer screening.This article first introduces the background of dermoscopic segmentation and deep learning models,and introduces the application of deep learning in dermoscopic segmentation.Secondly,this article introduces the algorithm models of convolutional neural networks and attention mechanisms,reviews the application of fused attention convolutional neural networks in dermoscopic segmentation since Jan 2022,and summarizes the improvement strategies,the advantages and disadvantages of the model.The model is further analyzed based on commonly used datasets of dermoscopy and evaluation indicators of image segmentation.Finally,the application of fused attention convolutional neural network in dermoscopic segmentation is summarized and prospected.

Objective:To explore the effect of the feedback method education model in patients with recurrent aphthous ulcer (RAU) , so as to provide a new education model for clinical health education.Methods:From May 2019 to May 2020, convenience sampling was used to select 76 RAU patients in the Stomatology Hospital of Southern Medical University. According to the method of random number table, the patients were divided into the experimental group and the control group, with 38 cases in each group. The experimental group implemented feedback method education model, and the control group carried out routine health education. The scores of Morisky Medication Adherence Scale (MMAS-8) , number of ulcers, healing time, pain degree, ulcer diameter and the Oral Health Impact Profile (OHIP-14) scores of patients were compared between the two groups before and after the intervention.Results:The score of MMAS-8 in the experimental group was higher than that in the control group after the intervention, and the difference was statistically significant ( P<0.05) . After the intervention, the number of ulcers in the experimental group was less than that in the control group, and the ulcer diameter was smaller than that in the control group, and the healing time was shorter than that in the control group, and the pain degree was lower than that in the control group, and the differences were statistically significant ( P<0.05) . After the intervention, the total score and dimension scores of OHIP-14 in the experimental group were lower than those in the control group, and the differences were also statistically significant ( P<0.05) . Conclusions:The feedback method education model can improve the adherence of RAU patients, the clinical indicators and the quality of life related to oral health of patients, which is worthy of clinical promotion and application.

OBJECTIVE：To systematically evaluate the effects of CYP1A2 gene polymorphisms on blood concentrations of antipsychotic drugs （haloperidol，clozapine，olanzapine），and to provide evidence-based reference for the clinical drug use . METHODS：Retrieved from Cochrane Library ，PubMed，Embase，CBM，CNKI and Wanfang data ，during the inception to Nov. 2019，cross-sectional study was conducted to investigate the effects of CYP1A2 gene polymorphisms on blood concentrations of antipsychotic drugs （haloperidol，clozapine，olanzapine）were collected. After screening the literature ，extracting the data and quality evaluation with Q-Genie tool ，Meta-analysis was performed by using Rev Man 5.3 software. RESULTS ：A total of 11 cross-sectional studies were included ，with a total of 914 patients. Of these ，haloperidol was used in 2 studies，clozapine was used in 5 studies，and olanzapine was used in 4 studies. Meta-analysis showed that there was no statistically significant difference in the blood concentration of haloperidol between CYP1A2（－2964G＞A）G/G type and G/A+A/A type [SMD ＝－0.22，95％CI（－0.66， 0.23），P＝0.35]. The blood concentration of clozapine in CYP1A2（－163C＞A）A/C type was significantly lower than C/C type [SMD ＝0.31，95％CI（0.01，0.62），P＝0.04]；there was no statistical significance in blood concentration of clozapine between A/A type and C/C type [SMD ＝0.09，95％CI（－0.21，0.40），P＝0.56]，between A/A type and A/C type [SMD ＝－0.22，95％CI（－0.55， 0.10），P＝0.18]，between CYP1A2（－2467delT）delT/delT type and T/T type [SMD ＝－0.11，95％CI（－0.75，0.52），P＝0.72]， between delT/T type and T/T type [SMD ＝0.01，95％CI（－0.33，0.34），P＝0.97]，between delT/delT type and delT/T type [SMD ＝ －0.15，95％CI（－0.80，0.05），P＝0.66]. The blood concentration of olanzapine in CYP1A2（－163C＞A）A/A type was signifi- cantly lower than A/C type [SMD ＝－0.31，95％CI（－0.55， － 0.08），P＝0.009]；there was no statistically significant difference in the blood concentration of olanzapine between A/A liukefeng-num.1@163.com type and C/C type [SMD ＝－0.20，95％CI（－0.61，0.21），P＝ 0.34]，between A/C type and C/C type [SMD ＝0.06，95％CI E-mail：hnmuzj@163.com （－0.35，0.47），P＝0.77]，between CYP1A2（－2467delT）delT/T type and T/T type [SMD ＝0.28，95％CI（－0.15，0.71），P＝0.20]. CONCLUSIONS：CYP1A2（－163C＞A）A/C type is related to the reduction of clozapine blood concentration ，and A/A type is related to the reduction of olanzapine blood concentration. CYP1A2 （－163C＞A）gene polymorphism is significance for guiding individualized medication of schizophrenia patients.

Objective To explore the practical effects of multiple intelligence theory applied in Obstetrics and Gynecology clinical internship teaching for the medical students.Methods The Eight-year program medical students in the classes of 2009 and 2010 were collected,among which 45 students in theexperimental group were taught by multiple intelligence theory,and 40 students in the control group were taught by traditional teaching methods.The results determined by the questionnaire survey,written test scores,operation test scores,medical documents writing results,usual performance and teachers' job evaluation.The data were analyzed by SPSS 17.0 statistical software,chi-square test and t test.Results Compared with control group,the written test scores,operation test scores,medical documents writing results,usual performance,the student questionnaire and teachers' self work evaluation of the experimental group were significantly higher (89.720 ± 8.147) vs.(81.140 ± 9.142),(88.230 ± 4.328) vs.(80.360 ± 5.621),(86.530 ± 2.641) vs.(81.580 ± 2.036),(87.650 ± 3.957) vs.(82.360 ± 6.392),which showed statistically significant difference (P＜0.05).The experimental group students think the teaching methods with multiple intelligence theory are novel,producing following positive effects in fully inspiring learning initiative,fostering the ability of independent study,enlightening clinical thinking,strengthening collective cooperation and improving the ability of doctor-patient communication and problem-solving etc.Conclusions It shows that the application of multiple intelligences theory in Obstetrics and Gynecology clinical internship for the medical students could motivate the students' self-confidence and enthusiasm,and contribute to train high quality medical talent,Meanwhile,the higher requirements are needed for teachers.It is worth promoting the application.

Objective: To study the segment of liver according to the large amount of three-dimensional(3D) reconstructive images of normal human livers and the vascular system, and to recognize the basic functional liver unit based on the anatomic features of the intrahepatic portal veins. Methods: The enhanced CT primitive DICOM files of 1 260 normal human livers from different age groups who treated from October 2013 to February 2017 provided by 16 hospitals were analyzed using the computer-aided surgery system.The 3D liver and liver vascular system were reconstructed, and the digital liver 3D model was established.The vascular morphology, anatomical features, and anatomical distributions of intrahepatic portal veins were statistically analyzed. Results: The digital liver model obtained from the 3D reconstruction of CAS displayed clear intrahepatic portal vein vessels of level four.Perform a digital liver segments study based on the analysis of level four vascular distribution areas.As the less anatomical variation of left hepatic portal vein, the liver was classified into four types of liver segmentation mainly based on right hepatic portal vein.Type A was similar to Couinaud or Cho′s segmentation, containing 8 segments(537 cases, 42.62%). Type B contained 9 segments as there are three ramifications of right-anterior portal vein(464 cases, 36.82%). The main difference for Type C was the variation of right-posterior portal vein which was sector shape(102 cases, 8.10%). Type D contained the cases with special portal vein variations, which needs three-dimensional simulation to design individualized liver resection plan(157 cases, 12.46%). These results showed that there was no significant difference in liver segmental typing between genders(χ²=2.179, P=0.536) and did not reveal any significant difference in liver segmental typing among the different age groups(χ²=0.357, P=0.949). Conclusions The 3D digital liver model can demonstrate the true 3D anatomical structures, and its spatial vascular variations.The observation of anatomic features, distribution areas of intrahepatic portal veins and individualized liver segmentation achieved via digital medical 3D visualization technology is of great value for understand the complexity of liver anatomy and to guide the precise hepatectomy.