Objective: To compare the performance of machine learning (ML) and individualized treatment effect (ITE) models based on deep learning in providing personalized treatment recommendations using real-world clinical datasets, and construct personalized drug treatment recommendation models for pegylated liposomal doxorubicin (PLD) and epirubicin (EPI) in postoperative breast cancer patients, and assist clinical decision-making by evaluating the treatment effects of these drugs. Methods: Clinical data of 904 breast cancer patients admitted at Sir Run Run Shaw Hospital, Zhejiang University School of Medicine was collected retrospectively, including 387 cases treated with PLD and 517 cases treated with EPI. The two groups were compared using propensity score matching to assess the 5-year disease free survival (DFS) outcome. Six ITE models, including CFR_WASS, were used to predict the 5-year DFS probability of patients under two drug treatments. Six machine learning (ML) models, including Random Forest, were used as baselines for performance analysis and comparison. Model's Predictive performance was evaluated based on the AUROC. The effectiveness of treatment recommendations was assessed by calculating the difference of 5-year rates between the group where the actual treatment used was consistent with the treatment recommended by the model and the control group. Results: Among the 153 matched cases, there was no statistically significant difference in 5-year DFS outcomes between the two groups. In 16 pairs of cases, the PLD group showed better clinical outcomes than the EPI group, and in 12 pairs of cases, the EPI group had better clinical outcomes than the PLD group, confirming individual differences in treatment benefit between the two drugs. The CFR_WASS model achieved the optimal predictive performance (AUROC value was 0.736 8), and there was no significant difference in 5-year DFS rates between most ML groups and the control group; The 5-year DFS rate in the ITE group was lower than that in the control group (P < 0.01), showing significant differences. Among them, the 5-year DFS rate in the CFR_WASS group was 2.13% lower than that in the control group. Conclusion: The ITE model is more accurate in estimating the individualized treatment effects of two drugs compared to the ordinary ML model, providing effective individualized treatment recommendations, and has certain clinical application value.

Propensity score weighting methods are crucial tools for adjusting confounding and conducting causal inference in observational studies. This paper introduces various propensity score weighting methods, and details the key features of each to aid researchers in selecting the most suitable method for their analyses. Additionally, the application of these methods in the fields of medicine and public health is demonstrated through an example data analysis. The example is based on data from 996 patients who underwent percutaneous coronary intervention (PCI). Different propensity score weighting methods are used to explore the effect of PCI combined with abciximab on six-month post-operative survival rates. The results indicate that different propensity score weighting methods provide treatment effect estimates tailored to different inference goals, highlighting the importance of selecting the appropriate method based on the specific inference objective. Propensity score weighting methods enhance the balance of covariate distribution between treatment and control groups, making them powerful tools for adjusting confounding factors in observational studies. Their applicability and value in medical and public health research are significant.

Glomerulocystic kidney disease (GCKD), a rare nephrocystic disorder, was the focus of this study. The FAT1 gene, comprising 27 exons, encoded a single-pass transmembrane protein with multiple biological functions. The extracellular portion of this protein consisted of 33 cadherin repeat sequences, 5 epidermal growth factor-like domains, and a G-layer associated domain. This report detailed a case of glomerular cystic disease caused by a mutation in the FAT1 gene, which was treated with the traditional Chinese medicine formula 'Shen Jiu Fang Modified'. This formula included Astragalus, Atractylodes, Poria, Angelica, Clematis, Silkworm, Hedyotis diffusa, Hibiscus leaf, among other ingredients, and aimed to invigorate the spleen and promote diuresis, as well as to nourish the kidneys and promote blood circulation. This case demonstrated the potential of traditional Chinese medicine in addressing rare kidney diseases, offering new insights for the diagnosis and treatment of similar complex cases.

Cantharidin is a terpenoid compound secreted by Mylabrisphalerata Pallas. In a variety of cancer types, cantharidin has shown the effects of inhibiting cancer cell growth, proliferation and migration. The anticancer mechanism of cantharidin involves inducing cell cycle arrest and apoptosis, inhibiting autophagy, enhancing DNA damage, inhibiting DNA repair, and regulating various cell signaling pathways. This article reviews the mechanism and research progress of cantharidin and its derivatives in common cancer types, in order to identify novel targets of cancer treatment using cantharidin, and provide new direction for clinical research to improve anticancer therapy.

Objective: To study and verify the function of de novo interferon regulatory factor (IRF8) frameshift mutation detected in an etiology screening of the cohort of children with recurrent pneumonia at the molecular level. Methods: The recombinant overexpression plasmids with wildtype or mutated IRF8 genes were constructed to transiently transfect HEK293T cells, or packed into lentivirus to infect two kinds of immune cell lines. Q-PCR, Western blot, immunofluorescence and other experimental assays were performed to explore the differences of expression and the regulatory effect on downstream genes associated with inflammation. Results: The recombinant vectors with wildtype or mutated IRF8 genes were constructed successfully, and the efficiency of transfection by plasmids and infection by packed lentivirus was remarkable as well. Compared with wildtype, the molecular weight of IRF8 variant was slightly increased, while the expression level presents in opposition, even if on transcription level. Moreover, the localization of IRF8 variant was detected in abundance in nucleus rather than cytoplasm, and its inhibition effect was enhanced on the downstream ISRE element in comparison with the wildtype IRF8 protein. Conclusion: The de novo frameshift mutation was presumed as gain-of-function (GOF) mutation.

This paper briefly introduced the varying coefficient model and used the varying coefficient Cox model as an example to demonstrate its application in the fields of medicine and public health through real data analysis, thereby provided methodological references for related research. The example was based on chronic disease management data from a Center for Disease Control and Prevention, fitting a varying coefficient Cox model to explore the time-varying association between body mass index (BMI) and mortality risk among a hypertensive population. The results showed that being underweight (BMI < 18.5 kg/m²) was associated with a higher risk of mortality, but this association weakened over time; being overweight (23 kg/m²≤BMI < 25 kg/m²) was associated with a lower risk of mortality, and this association also weakened over time. The varying coefficient model captured how the impact of exposure factors on outcomes changed with other variables, helping to comprehensively understand the complex relationships between variables, and had significant application and promotion value in medical and public health research.

Objective: To achieve automatic segmentation of cell nuclei in gastrointestinal cancer pathological images by using a deep learning algorithm, so as to assist in the quantitative analysis of subsequent pathological images. Methods: A total of 59 patients with gastrointestinal cancer treated in Ruijin Hospital, Shanghai Jiao Tong University School of Medicine from Jan 2022 to Feb 2022, were selected as the research objects. Python and LabelMe were used for data anonymization, image segmentation, and region of interest annotation of patients' pathological images. A total of 944 pathological images were included, and 9 703 nuclei were annotated. Then, a new semantic segmentation model based on deep learning was constructed. The model introduced deformable attention transformer (DAT) to realize automatic, accurate and efficient segmentation of pathological image nuclei. Finally, multiple segmentation evaluation criteria are used to evaluate the model's performance. Results: The mean absolute error of the segmentation results of the model proposed in this paper was 0.112 6, and the dice coefficient (Dice) was 0.721 5. Its effect was significantly better than the U-net baseline model, and it was ahead of models such as ResU-net++, R2Unet and R2AttUnet. Moreover, the segmentation results were relatively stable with good generalization. Conclusion: The segmentation model established in this study can accurately identify and segment the nuclei in the pathological images, with good robustness and generalization, which is helpful to play an auxiliary diagnostic role in practical applications.

Objective: To investigate the contribution of microglia in the basolateral amygdala (BLA) to pain hypersensitivity and pain-related aversion in knee-joint monoarthritis mice. Methods: A total of 61 mice were used for behavioral tests (14 mice in the control group and 47 mice in the model group), and other 6 mice were used for cell morphology (3 mice in each group). An animal model of knee-joint monoarthritis was established by injection of complete Freund's adjuvant (CFA) into the knee-joint cavity of mice. The von Frey and Hargreaves tests were used to examine mechanical allodynia and thermal hyperalgesia in mice, respectively. The place escape/avoidance paradigm test was used to examine pain-related aversion. Open field test and elevated plus maze test were used to examine anxiety-like behaviors in mice. Morphological changes of microglia in the BLA area after CFA injection were assessed by 3D reconstruction of microglia in the BLA brain region using immunofluorescence staining and Imaris software. Results: Compared with the control group, CFA-arthritic mice produced significant mechanical and thermal hyperalgesia in the ipsilateral hindpaw and maintained for at least 12 and 19 days, respectively. Meanwhile, CFA injection induced pain-related aversion and anxiety-like behaviors in mice, accompanied by significant activation of BLA microglia. Inhibition of BLA microglia activation alleviated CFA-induced hyperalgesia and aversive behaviors but had no significant effects on anxiety-like behaviors. Conclusion: CFA-arthritic mice produce hyperalgesia, pain-related aversion, and anxious behavior, in which hyperalgesia and pain-related aversion may be mediated by the activation of microglia in BLA.

Objective: To explore the experience of migraine in patients undergoing closure of patent foramen ovale, so as to develop targeted interventions for relieving patients' migraine and improving quality of life. Methods: It was a qualitative study. A total of 11 patients who experienced migraine, underwent closure of patent foramen ovale, and admitted to the Department of Cardiology, Zhongshan Hospital, Fudan University from Oct 2022 to Apr 2023 were selected. One-to-one, in-depth semi-structured interviews were conducted. The conventional content analysis was used for data analysis. Results: The migraine experience of patients undergoing closure of foramen ovale could be shown in 5 themes: migraines attacked varied; migraines brought serious physical and mental effects; migraines were dealt with in a variety of ways; patients had expectations for closure; professional support was needed. Conclusion: The experience of migraine in patients undergoing closure of patent foramen ovale varied and was long-lasting. Patients were unable to effectively deal with migraine which had great negative effects. Multidisciplinary pain clinics could be established. Health care professionals should provide patients with professional and personalized pain management, pay attention to the accompanying symptoms and the negative mood, and improve patients' awareness and management level of migraine.

With the development of genetics and advances in genetic testing technology, the demand for cancer genetic counseling has increased dramatically. Advanced practice nurses play a key role in personalized health care delivery. The oncology genetic nurse-led genetic counseling services in foreign countries are becoming more and more mature, but in China, the work of oncology genetic counseling started late, and the combination of genetics/genomics with nursing is still in its infancy. There is still a lack of relevant research on oncology genetic nurses. This article introduced the qualification certification, core competence and clinical practice content of foreign oncology genetic nurses, and summarized the clinical practice effect of oncology genetic nurses and the enlightenment to China's advanced nursing practice, which provided references for the construction of oncology genetic nurses training programs and clinical service models suitable for China's national conditions, so as to meet the needs of the development of advanced nursing practice and the growing demand for precision oncology and high-quality genetic medical care.

Leonurine (SCM-198) was discovered as one of the active constituents of the Herba Leonuri (HL). Now it can be artificially synthesized. Several recent researches has proven that it exhibits anti-inflammatory effect in several systems in animal models and cell culture in vitro. The key mechanism involves downgrading the activity of nuclear transcription factor-κB (NF-κB), thereby inhibiting the phosphorylation of several signal pathways such as PI3K/Akt, MAPK, ERK, and JNK, or upregulating the activity of Nrf2 related pathways, resulting in downregulated expression of inflammatory cytokines such as tumor necrosis factor-α (TNF-α), IL-1β, IL-2, IL-6, IL-8, inducible nitric oxide synthase (iNOS), cyclooxygenase-2 (COX-2), chemokines, adhesion molecules, etc. Owing to the advantages of high safety and efficiency, the ease of administration, as well as its effectiveness in many organs and systems, leonurine has a widely prospect for future research and clinical applications. This article reviews the progress in the fundamental research of leonurine in multiple inflammation-related disease, and it could be expect to offer new possibilities for the treatment of these disease.

Objective: To investigate the role of the TNFRSF12A molecule in the pathogenesis of liver cancer. Methods: Through comprehensive analysis of the Cancer Genome Atlas Program (TCGA) database and single-cell sequencing data, we studied the expression of TNFRSF12A in liver cancer and its correlation with prognosis. HPA database was utilized to analyze the subcellular localization of TNFRSF12A, and GO and KEGG analyses were performed by DAVID. TIME 2.0 was employed to analyze the correlation between TNFRSF12A and immune cell infiltration in liver cancer tissues. Results: TNFRSF12A was found to be highly expressed in liver cancer tissues, significantly correlating with patient survival prognosis (OS: HR=1.61, P=0.007 0; RFS: HR=1.45, P=0.037 0;PFS: HR=1.30, P=0.099 0; DSS: HR=1.67, P=0.027 0), as well as age (P=0.046 7) and BCLC stage (P=0.045 6). TNFRSF12A co-expressed with tumor stem cell markers (CD24, SOX4, ANPEP), indicating a strong link to malignancy. Furthermore, molecular functional analysis unveiled that IL-2R primarily existed in the cell cytoplasm and played a role in processes such as cell apoptosis, invasion, and protein binding. Moreover, TNFRSF12A was associated with Treg cells and immune cell infiltration, further suggesting its role in tumor immune regulation. Conclusion: TNFRSF12A exhibits a significant elevation within liver tumors and shows a notable correlation with patients' prognosis. Tumor cells engage in interactions with cytokines produced by Tregs, thereby reshaping the tumor microenvironment. The potential clinical significance of TNFRSF12A as a prognostic marker for tumors holds promise in offering novel avenues for personalized treatment and prognosis prediction.

Objective: To investigate the effect of cordyceps sinensis (CS) on the activation of fibroblasts through IL-6 trans-signaling pathway and its specific mechanism in the treatment of renal fibrosis. Methods: Renal fibrosis mouse model was established by unilateral ischemia/reperfusion (UIR), and the mice were administered intragastrically CS, soluble glycoprotein 130 Fc (sgp130Fc) or Hyper-IL-6. Masson's trichrome staining was utilized to identify tubulointerstitial fibrosis. PAS staining was utilized to assess the extent of renal injury. Western blot was employed to analyze the expression levels of fibrosis markers [alpha-smooth muscle actin (α-SMA), fibronectin (FN)] and proteins associated with IL-6 trans-signaling pathway [phosphorylated signal transducer and activator of transcription 3(p-STAT3), soluble interleukin-6 receptor (sIL-6R)]. The expression and localization of proteins were additionally detected by immunohistochemistry, immunofluorescence and qPCR. The effect of cordyceps sinensis extract cordycepin on IL-6 trans-signaling in fibroblasts was further investigated in vitro. Results: The results from in vivo experiments showed that administration of CS during the chronic phase demonstrated a beneficial protective impact on inflammation and fibrosis in the affected kidney, and serum creatinine levels and collagen deposition were decreased. Western blot analysis revealed a decrease in the expression levels of α-SMA, FN, as well as IL-6 trans-signaling pathway protein p-STAT3, sIL-6R in the treatment group. Additionally, the mRNA expression levels of chemokines monocyte chemoattractant protein-1 (MCP-1) and C-X-C motif chemokine ligand 12 (CXCL12) were also decreased in the CS treatment group. Additionally, Hyper-IL-6 can partially counteract the therapeutic effects of CS. In vitro experiments further demonstrated that cordycepin inhibited the secretion of IL-6 from NRK-52E. Combined treatment of recombinant IL-6 and sIL-6R protein activated NRK-49F, leading to a significant increase in α-SMA, FN, and p-STAT3 expression levels. Cordycepin or sgp130Fc treatment significantly inhibited the proliferation of fibroblasts induced by IL-6 trans-signaling pathway. Conclusion: CS can significantly reduce IL-6 secretion by renal tubular epithelial cells and inhibit the activation of IL-6 trans-signaling pathway in fibroblasts, thereby ameliorating renal interstitial fibrosis.

Objective: To investigate anxious and depressive emotions in patients underwent cervical cancer surgery and to analyze its influencing factors. Methods: A total of 304 patients who underwent primary cervical cancer surgery in Obstetrics and Gynecology Hospital, Fudan University from Oct 2018 to Jun 2021, were recruited to evaluate the clinical effect based on cervical cancer-patient reported outcome 137 scale (CC-PRO137 scale). This study focused on dimensions of depressive and anxious emotions within this scale and explored their influencing factors. Results: The average scores of their depressive and anxious emotions within half a year after surgery were 4.141±0.798 and 4.020±0.616, respectively; and the average scores of depressive and anxious emotions more than one year after surgery were 4.250±0.802 and 4.097±0.613, respectively. By using statistical methods including analysis of variance and t test, it was found that there were statistically significant differences in the scores of depression and anxiety among cervical cancer patients under different postoperative adjuvant treatments and at different postoperative time points (P < 0.05). However, there were no statistically significant differences in the scores of depression and anxiety among patients with different ages, surgical methods, and clinical stages of cervical cancer (P > 0.05). Conclusion: Patients underwent cervical cancer surgery may suffer varying degree of depressive and anxious emotions, and the main influencing factors are different adjuvant treatments and the length of time for postsurgical recovery. Medical practitioners should strengthen comfort and care for patients with cervical cancer, especially those who receive chemotherapy and radiotherapy treatments and are in the primary stage after the surgery. Formulating positive intervention measures can effectively reduce the psychological pain of patients and safeguard their physical and mental health.

Objective: To elucidate the potential mechanisms by which mesothelin (MSLN) contributes to chemotherapy resistance in high-grade serous ovarian cancer (HGSOC). Methods: A Meta-analysis utilizing public ovarian cancer databases was performed to evaluate the correlation between MSLN expression levels and overall survival (OS) in ovarian cancer patients. Pathway enrichment analysis was employed to identify key signaling pathways regulated by MSLN and their roles in chemotherapy resistance. Additionally, the TCGA-HGSOC database was analyzed to examine genomic features associated with MSLN-mediated chemotherapy resistance. To validate the biological function of MSLN in chemotherapy resistance, an intraperitoneal metastasis model was established using MSLN-knockdown ID8 ovarian cancer cells in mice. Results: Elevated MSLN expression was significantly associated with poor patient prognosis (HR: 1.42, 95%CI: 1.16-1.74). Differential gene expression and pathway enrichment analyses revealed that high MSLN expression upregulates resistance-associated genes and pathways involved in drug metabolism and DNA-binding signaling. Genomic association analysis showed a negative correlation between high MSLN expression and chromosomal instability features, specifically CX3, CX11, and CX13 scores. In vivo studies demonstrated that MSLN knockdown enhanced the tumor-suppressive effects of cisplatin. Conclusion: High MSLN expression represents a potential biomarker for poor prognosis and chemotherapy resistance in HGSOC patients, suggesting MSLN as a promising target for therapeutic intervention.

Objective: To investigate the relationship between triglyceride-gluscose (TyG) index in early pregnancy and the delivery of small for gestational age infant (SGA) in patients with gestational diabetes mellitus (GDM). Methods: A total of 1 532 pregnant women from Shanghai Fifth People's Hospital and the Second People's Hospital of Kashgar Region, who were enrolled in the study between Jan 2018 and Jun 2023 and met the inclusion criteria, were categorized into the group with GDM (754 cases) and the group without GDM (778 cases) based on the results of the oral glucose tolerance test (OGTT) conducted at 24-28 weeks of pregnancy. Within the GDM group, patients were further categorized into 3 groups based on neonatal weight: SGA group, large for gestational age infant (LGA) group, and appropriate for gestational age infant (AGA) group. A Logistic regression model was used to analyze the independent influences on SGA delivery in patients with GDM to analyze the correlation between the TyG index and the occurrence of SGA. The predictive value of the TyG index in early pregnancy for SGA delivery in GDM patients was analyzed using ROC curves. Results: The TyG index in the SGA group of GDM patients was significantly lower than that in the LGA, AGA and control without GDM group (P < 0.05). Multifactorial logistic regression analysis revealed that the TyG index was independently correlated with the SGA in GDM patients (P < 0.05). The ROC curve analysis demonstrated a good predictive value of the TyG index in early gestation for SGA delivery in GDM patients (AUC=0.821, 95% CI: 0.763-0.879, P < 0.001). Conclusion: The TyG index in early pregnancy among GDM patients is independently correlated with the occurrence of SGA infants and has a good predictive value for SGA delivery in GDM patients.

Objective: To investigate the role of magnesium ion (Mg²⁺) in cisplatin-induced acute kidney injury (Cis-AKI) in kidney organoids and HK-2 cells, as well as the potential mechanism. Methods: Initially, we utilized human-derived induced pluripotent stem cells (iPSCs) to construct kidney organoids, and then built a Cis-AKI model based on kidney organoids. HE staining was used to observe the structure of kidney organoids, and immunofluorescence staining was used to observe the localization of markers and the expression of cleaved caspase-3. qRT-PCR was conducted to detect mRNA levels of tubular and glomerular markers, as well as inflammatory factors. Subsequently, the kidney organoids were randomly divided into control group, cisplatin group (Cis group), and Mg²⁺ pretreatment group (Cis+Mg²⁺ group). CCK-8 and ATP content assays were employed to evaluate the cell viability of renal tubular epithelial cells. TUNEL staining was performed to detect the apoptosis of renal tubular epithelial cells. Western blot was utilized to detect the expression of apoptosis-associated proteins (Bcl-2, Bax, cleaved caspase-3) and organic cation transporter 2 (OCT2). Immunofluorescence was used to detect the localization and expression of OCT2. Results: On the 10th day, the tubular structure in kidney organoids was visible, with abundant expression of renal markers. Treatment with 10 μmol/L cisplatin resulted in structural damage to kidney organoids, significantly increased expression of cleaved caspase-3 and mRNA levels of inflammatory factors, and significantly decreased ATP content. Compared with the Cis group, the Cis+Mg²⁺ group showed increased ATP content in kidney organoids, reduced number of TUNEL-positive cells, significantly decreased expression of apoptosis-associated proteins, and significantly decreased expression of OCT2. However, there was no significant improvement in HK-2 cell viability, the number of TUNEL-positive cells, or apoptosis-associated proteins in the Cis+Mg²⁺ group, and HK-2 cells did not express OCT2. Conclusion: Kidney organoid is an ideal in vitro model to study the pathogenesis and treatment of Cis-AKI. Mg²⁺ pretreatment can significantly reduce the damage of kidney organoids induced by cisplatin, and the mechanism may be related to the downregulation of OCT2.

Objective: To evaluate the predictors of biochemical recurrence after laparoscopic radical prostatectomy. Methods: The study cohort consisted of 416 consecutive prostatecancer patients who underwent radical prostatectomy in Zhongshan Hospital, Fudan University between Jan 2016 and Jun 2018. No patient received adjuvant therapy until documented biochemical recurrence. Biochemical recurrence-free survival (BCRFS) was estimated using the Kaplan-Meier method. Univariable and multivariable Cox proportional hazards regression models were utilized to determine variables predictive of biochemical recurrence. Results: The median follow-up period was 174 weeks (inter quartile range 133-209 weeks).The overall biochemical recurrence rate was 21.39% (88/416).The time to biochemical recurrence was (51.11±47.71) weeks. Until the end of follow-up, the biochemical recurrence rates were 10%(1/10), 6.12%(3/49), 16.41%(43/262) and 43.16%(41/95) for low-risk, medium-risk, high-risk and locally advanced prostate cancer, respectively.On multivariable analysis, tumor invasion of the seminal vesicles (HR: 2.119, 95%CI: 1.154-3.891, P=0.016);Seminal vesicle invasion (HR: 1.824, 95%CI: 1.034-3.220, P=0.038), Gleason score≥8 (HR: 2.746, 95%CI: 1.268-5.943, P=0.010), prostate-specific antigen (PSA)=10-20 ng/mL(HR: 4.240, 95%CI: 2.550-7.050, P ＜ 0.001);PSA > 20 ng/mL (HR: 6.341, 95%CI: 3.296-12.202, P ＜ 0.001)were the main predictors of biochemical recurrence. Conclusion: Laparoscopic radical prostatectomy provides effective control of biochemical recurrence during the mid-term follow-up period. Tumor invasion of the seminal vesicles, seminal vesicle invasion, GS≥8, PSA=10-20 ng/mL, PSA > 20 ng/mL were important predictors of biochemical recurrence after radical prostatectomy.

Objective: To establish a scoring model for predicting the prognosis and drug sensitivity of colorectal cancer (CRC) based on the expression of disulfidptosis-related genes by bioinformatics analyses combined with the validation with CRC patient-derived organoids (CRC-PDOs). Methods: NMF (non-negative Matrix Factorization) algorithm, Cox and LASSO regression analyses were used to identify disulfidptosis-related genes with predictive value for CRC prognosis, and disulfidptosis-related risk scoring formula was constructed. The differential genes and enrichment pathways among different clusters were analyzed by GO (Gene Ontology) and KEGG (Kyoto Encyclopedia of Genes and Genomes). The sensitivity of the high/low-risk clusters of CRC patients to chemotherapy drugs was predicted using the GDSC database and validated using CRC-PDOs. Results: The results of NMF algorithm showed that CRC patients could be grouped into two clusters based on the disulfidptosis-related genes. COX regression analysis demonstrated that LRPPRC and SLC7A11 were the only two genes with significance to predict the prognosis of CRC patients (P=0.047, 0.033). Low expression of SLC7A11 or high expression of LRPPRC in tumors of CRC patients was significantly correlated with overall survival (OS) (P=0.004, 0.003). Based on LASSO regression analysis, the mortality risk scoring formula for disulfidptosis was as follows: Risk score=LRPPRC×(-0.670 5)+SLC7A11×0.311 2, and the GSE161158 dataset could be re-grouped into high-risk and low-risk clusters accordingly. There were 125 differentially expressed genes (DEGs) between the two clusters. According to the GO and KEGG results, the up-regulated genes in high-risk cluster were mainly enriched in immune regulation, such as leukocyte chemotaxis, granulocyte migration and toll-like receptor binding. Low-risk cluster was characterized by pathways associated with sulfide metabolism, such as sulfur compound transmembrane transporter activity. Based on the GDSC database, the expression level of SLC7A11 and LRPPRC could predict chemotherapy drug sensitivity. As a representative, the efficacy of chemotherapy drug (irinotecan) on inhibiting the growth of CRC-PDOs was shown to be linearly correlated with the relative gene expression levels of SLC7A11 and LRPPRC in CRC tissues of patients (P=0.007, 0.040). Conclusion: According to the results based on the bioinformatics analyses and drug sensitivity testing on CRC-PDOs, disulfidptosis risk score could predict the prognosis and drug sensitivity of CRC patients, with potential clinical application prospect.

Objective: To develop a health belief model (HBM) based adolescent alcohol-related cognition scale to measure adolescent alcohol-related cognition and test its reliability and validity. Methods: The adolescents' alcohol-related cognitive scale was developed based on HBM model. By using purposive sampling, three general high schools in Qingpu District, Shanghai were selected. One-third of the classes from grades 10 and 11 in each school were randomly selected, and the students from these classes were surveyed as the research subjects.Exploratory factor analysis and confirmatory factor analysis were used to analyze its reliability (internal consistency reliability and combination reliability) and validity (structural validity, convergent validity, discriminative validity and criterion validity). Results: A total of 970 questionnaires were collected, of which 948 were valid, with an effective rate of 97.7%. The adolescents' alcohol-related cognitive scale contained 22 items. Five common factors were extracted from exploratory factor analysis, including perceived susceptibility, perceived severity, perceived benefits, perceived obstacles, and self-efficacy. The cumulative variance contribution rate reached 83.89%. The results of confirmatory factor analysis confirmed the overall fit of the model. The average variance extracted value (AVE) of each dimension was greater than 0.5, and the convergent validity of the model was ideal. The AVE square root of each dimension of the scale was greater than its correlation coefficient, indicating good discrimination validity. Cronbach's α coefficient of the total volume table was 0.892, indicating good overall reliability. Conclusion: The adolescents' alcohol-related cognitive scale developed in this study has good reliability and validity, which can be used to measure adolescents' alcohol-related perceptions.

Objective: To explore the safety and management mode of hysteroscopy in three different modes: outpatient, daily and inpatient. Methods: The quality control data of patients who underwent hysteroscopic surgery in Hysterscopy Centre, Obstetrics and Gynecology Hospital, Fudan University from Jan 2019 to Dec 2021 were collected through the electronic information system of the hospital and the monthly quality control report of hysteroscopy center. The amount of surgery, the proportion of grade Ⅳ surgery, the analysis of operation types, the indicator including complications, and unanticipated secondary surgery were retrospectively analyzed. Results: From 2019 to 2021, 5 162 outpatient hysteroscopic patients, 15 331 daily hysteroscopic patients and 5 942 inpatient hysteroscopic patients were admitted in our hospital. The age of inpatient hysteroscopic patients was significantly older than those of outpatient and daily patients (P < 0.001). In the past three years, the proportion of daily hysteroscopy gradually increased, and the proportion of inpatient hysteroscopy gradually decreased (P < 0.001). The total percentage of grade Ⅳ hysteroscopic surgery was 12.9%, in which inpatient was higher than daily, and daily was higher than outpatient (P < 0.001). The incidence of complications and accidents during hysteroscopy was 0.117% (31/26 435), including 17 cases of uterine perforation, 7 cases of hysteroscopy failure, 3 cases of excessive intraoperative bleeding, 2 cases of fluid overload, 1 case of intestinal injury, and 1 case of anesthesia accident. The incidence of hysteroscopy in outpatient, daily and inpatient were 0.020% (1/5 162), 0.137% (21/15 331) and 0.151% (9/5 942) respectively. Conclusion: Hysteroscopy in outpatient, daily and inpatient are all safe and reliable. Outpatient and daily hysteroscopy can improve the efficiency of medical services, which has gradually become a trend.

Objective: To compare the clinical efficacy and pregnancy outcomes of CO₂ laser and loop electrosurgical excision procedure (LEEP) surgery in the treatment of cervical high-grade squamous intraepithelial lesion (HSIL)in women of childbearing age. Methods: Patients with cervical HSIL of appropriate reproductive age (20-34 years old) who underwent satisfied colposcope, subsequently treated with laser or LEEP between Jan 2019 and Dec 2021 in Obstetrics and Gynecology Hospital, Fudan University were collected. All the patients were interpreted as transformation zone (TZ) type 1/2, and colposcopy biopsy diagnosed as cervical high-grade squamous intraepithelial lesion (HSIL) and received laser or LEEP according to the physician's advice and the patient's wishes.The effectiveness of the two surgeries was evaluated, and postoperative cytology, HPV result, and colposcopic pathology were compared. We monitored the surgical pregnancy rate, pregnancy complications, and pregnancy results after the surgery. Results: A total of 566 cases, including 233 laser cases and 333 LEEP cases, were enrolled. There was no significant difference between the two groups in terms of preoperative cytology, HPV distribution, or the length of cervical canal. After 6 months of surgery, 91.5% of patients who underwent laser therapy and 86.9% of those who underwent LEEP had normal cytology. The histology cure rate was 91.5% and the HPV conversion rate was 74.6% in both groups. The results of cytology, HPV, and colposcopic pathology had no significant difference between the two groups. The length of postoperative cervical canal was significantly longer in laser group than that in LEEP group [(29.09±3.15) mm vs. (27.05±3.84) mm, P < 0.001]. There was no statistically significant difference in postoperative pregnancy rate and pregnancy outcome between the two groups. Conclusion: Laser ablation and LEEP produce comparable effects pregnancy outcomes in cervical HSIL patients with TZ types 1 and 2, while LEEP has pathologic feedback, so its misdiagnosis of invasive cancer reduce.

Objective: To describe the epidemiological characteristics of falls among hospitalized elderly patients in Shanghai, to explore the external and self-factors that affect falls in the elderly, and to provide evidence for the elderly to prevent falls. Methods: All the injury cases collected from the hospital information system of a general hospital in Shanghai during 2020 and 2022 based on Haddon Model were taken as the research objects. The epidemiological characteristics were analyzed using constituent ratio, chi-square test and Logistic regression analysis. Results: Falls occured in 11 152 cases (11.63%) within 95 880 hospitalized elderly patients aged 65 years old and over. The ratio of males to females was 0.49∶1. Falls were more likely to occur in females and those who lived with children, and the place was mainly home/residential institutions. The time distribution showed the fewest occurrence in February. The mainly injured part was the trunk and limbs with the nature of fracture. Multiple regression analysis showed that compared to males, 65-75 years old, spring, married status and living with spouses, the risk factors of falls in elderly patients were female, older age, winter, not in marriage and living with children. Conclusion: The elderly have a high risk of falls, which is influenced by a combination of factors such as personal health, the natural environment, and social support. Comprehensive measures should be taken to identify and address the risk of falls in the elderly, including personal health promotion, home environment optimization, harmonious family atmosphere, professional medical support, and social collaboration, in order to achieve effective prevention and management.

Objective: To explore the mediating effect of maternal childbirth self-efficacy between reproductive health literacy during pregnancy and expected mode of delivery, in order to provide a reference for obstetric care providers to carry out scientific and reasonable childbirth decision support interventions. Methods: A questionnaire survey was conducted among pregnant women in late pregnancy seeing doctors in Obstetric Clinic of Obstetrics and Gynecology Hospital, Fudan University from Oct 2020 to Feb 2021 using convenience sampling. A general information questionnaire, the Reproductive Health Literacy Scale, and Childbirth Self-efficacy Scale were used in the survey. Linear regression and Logistic regression analyses were used to assess the associations between reproductive health literacy during pregnancy and childbirth self-efficacy and expected mode of delivery, respectively, and to assess the significance of the mediating effect of childbirth self-efficacy between productive health literacy during pregnancy and expected mode of delivery. The MacKinnon's product-of-coefficients method was used to assess the mediating effect of maternal childbirth self-efficacy. Results: A total of 343 participants were included in the study. Regression analysis results of factors influencing expected mode of delivery showed that reproductive health literacy during pregnancy (c'=1.644, 95%CI: 2.223-12.059) and childbirth self-efficacy (β=2.211, 95%CI: 6.300-13.215) were associated with expected mode of delivery (P < 0.001); childbirth self-efficacy was significantly and positively associated with reproductive health literacy during pregnancy (β=1.280, 95%CI: 0.102-0.165, P < 0.01). Childbirth self-efficacy was a mediating variable for health literacy during pregnancy and expected mode of delivery with an effect size of 24.8%. Conclusion: Childbirth self-efficacy mediated the association between reproductive health literacy during pregnancy and expected mode of delivery. Obstetric care providers should focus on strengthening maternal care and childbirth education, improving pregnant women's childbirth self-efficacy and reproductive health literacy during pregnancy, thereby enabling women to make scientific decisions on expected mode of delivery in order to promote normal childbirth.

Objective: To analyze the global literature related to functional cure of hepatitis B from 2019 to 2023 by using bibliometric analysis methods, so as to help researchers understand the research hotspots and trends in this field. Methods: The literature related to the topic of functional cure of hepatitis B included in the Science Citation Index Expanded (SCI-Expanded) of the Web of Science Core Collection from 2019 to 2023 was searched. By using VOSviewer and CiteSpace visual analysis tools, analyses were conducted from the perspectives of publication trends, international research cooperation networks, and keyword emergence, and were elaborated with the specific contents of the related literature to elucidate research hotspots and trends. Results: A total of 600 eligible papers in this field were included. Keyword co-occurrence and thematic clustering suggested that the main research directions of functional cure were: serum biomarkers for prediction and monitoring of functional cure, functional cure and immunity, nucleoside analog discontinuation, interferon therapy, and long-term prognosis of functional cure. The research contents of the ESI highly cited original research papers were similar to the clustering of the above, but showed more attention on the novel agents for functional cure. The content of the keyword emergence map showed that hotspots of interest changed from virologic mechanisms and serum markers, to nucleoside analog discontinuation and interferon therapy, and finally to immunologic mechanisms and new drug. Conclusion: The research hotspots and trends of functional cure of hepatitis B were focused on virological mechanism, serum markers, immunological mechanism, nucleoside analog discontinuation, interferon therapy, and long-term prognosis after cure.

The inflammatory effect of dietary is strongly related to the development of cancer, therefore, the diet-related inflammatory index was developed as a methodological tool to investigate the relationship between dietary, inflammation and tumors. In this paper, we summarized the results on diet-related inflammatory indices and common cancers of the digestive system based on relevant cancer epidemiological studies. The available epidemiological evidence suggests that pro-inflammatory diet is associated with an increased risk of gastrointestinal malignancies, with the strongest association with colorectal cancer, followed by esophageal and gastric cancers, and then pancreatic cancer, and the least evidence of studies with liver cancer. Among these studies, the level of evidence for esophageal cancer is lower than colorectal cancer, the study of gastric cancer has gender differences and problems in adjusting for confounders, and the study of pancreatic cancer has heterogeneous results. In view of the current research progress and deficiencies, prospective studies or population-based cohort studies, as well as strengthening nutritional epidemiological studies related to common tumors such as liver cancer could be considered in the future. This review is expecting to provide basic information and scientific basis for strengthening the related healthy eating behavior promotion in the prevention and control of digestive system tumors.

Gallbladder adenoma, a benign tumor of mucosal epithelial origin, is considered to be a pre-cancer to gallbladder cancer (GBC), and its malignant transformation may occur in a relatively short time. The pathological process and molecular mechanism of gallbladder adenoma carcinogenesis are still controversial; ultrasound and CT are widely used to examine gallbladder diseases, and the application of contrast-enhanced ultrasound (CEUS) and high-frame-rate contrast-enhanced ultrasound (H-CEUS) has improved the diagnostic accuracy. In this paper, we summarize the literature on the clinicopathological features, imaging manifestations, gene expression, treatment, and clinical prognosis of gallbladder adenoma carcinogenesis.

Objective: To investigate the role of the GPR120 gene in the progression of sepsis, explore the molecular mechanisms through which GPR120 gene regulates NOD-, LRR- and pyrin domain-containing protein 3 (NLRP3) inflammasome activation and macrophage polarization. Methods: The blood and pleural fluid samples were collected from the sepsis patients and the control group. The expression of inflammatory factors and the associated proteins were detected by flow cytometry and ELISA. C57BL/6 mice and monocyte-macrophage cell line (Raw264.7) were treated with lipopolysaccharide (LPS) to construct the sepsis models. After the intervention of GPR120 agonist TUG891, the expression of GPR120 gene, NLRP3 inflammasome protein and macrophage polarization protein were detected between the control group and the sepsis group. Results: The expression of inflammatory factors, such as IL-1β in the serum of septic patients, significantly increased compared with the control (P < 0.001). And the expression of inflammasome proteins such as NLRP3, Caspase-1 and IL-1β in the pleural fluid also increased (all P < 0.05). In vivo, LPS could induce severe inflammation in lung tissue, the GPR120 gene expression decreased in lung tissue, and inflammatory factors were up-regulated in mouse serum (P < 0.01). The inflammasome-associated protein and M1 type polarization of macrophages were enhanced, the TUG891 could reduce the inflammatory response, inhibit the NLRP3 inflammasome activating, and promote the M2 polarization of macrophages (P < 0.01). In vitro, LPS could inhibit the intracellular GPR120 expression. The inflammatory factors secreted more in LPS-induced sepsis cells. TUG891 could promote the up-regulation of GPR120 protein and alleviate the secretion of inflammatory factors (P < 0.05). Conclusion: In sepsis, GPR120 gene activation could inhibit the NLRP3 inflammasome activation, promote macrophage polarization, and reduce the inflammatory damage, thereby delay the rapid progression of sepsis.

Objective: To explore the cut-off value of three dimensional (3D)vena contracta area (VCA)in diagnosing severe tricuspid regrugitation (TR) under different etiologies and its accuracy and practicality in clinical application. Methods: From Mar 2019 to May 2021, ninety-two patients with confirmed TR underwent two dimensional (2D)and 3D transthoracic echocardiography. The correlation and consistency between 3D VCA 3D calculated based on the proximal isokinetic surface area (PISA) effective regurgitant orifice area (EROA) was calculated. Comprehensive 2D multi-parameter method was used as a reference method to calculate the cut-off value of the diagnosis of severe TR. Results: A total of 85 patients were ultimately included. 3D VCA and 3D PISA EROA had similar and acceptable correlations in both primary TR and secondary TR (primary TR: r=0.831, P < 0.01; secondary TR: r=0.806, P < 0.01). Bland-Altman analysis showed that 3D VCA overestimated TR compared with 3D PISA EROA (62% overestimated in the total patient population, 51% overestimated in primary TR, and 74% overestimated in secondary TR). In secondary TR, the cut-off value of 3D VCA for diagnosing severe TR was 0.45 cm² (sensitivity 89%, specificity 82%); combining clinical symptoms, positive 2D PISA EROA results and 3D VCA results for severe TR, the chi-square value was higher than those only included clinical symptoms or incorporated clinical symptoms and positive 2D PISA EROA results (42.168 vs. 26.059 and 16.759, P < 0.01). Conclusion: 3D VCA would overestimate TR, and had high and incremental diagnostic value for evaluating severe TR in secondary TR.

Objective: To investigate the brain regions that play a key role in the impact of childhood adversity on cognitive function, based on the mediation effect analysis of debiased machine learning framework. Methods: The dataset, consisting of 460 volunteers, was randomly partitioned into a training set (n=368) and a validation set (n=92). To enhance the stability of the results, this study utilized a debiased machine learning mediation analysis method to develop models and draw conclusions from the training set, which were then validated with the validation set. Finally, the Bonferroni method was employed to correct for significance in the identified brain regions. Results: The experimental results for the training set and validation set were similar. The efficacy of the left superior parietal lobule was significant (validation set P ＜ 0.001), exhibiting a positive effect (validation set Effect Size=0.78). The right frontopolar cortex, anterior cingulate cortex, and left inferior temporal gyrus were also significant (P ＜ 0.001 for both training set and validation set) and were identified as having a negative effect in this process (validation set Effect Size =-0.037 and -0.034, respectively). The associations between the brain regions with significant effects and cognitive function are supported by existing literature. Conclusion: This study employed mediation analysis methods to investigate the mechanisms underlying the impact of childhood adversity on cognitive function. The results indicate that the left superior parietal lobule, right rostral anterior cingulate cortex, and left inferior temporal gyrus play crucial roles in the process of childhood adversity affecting cognitive function. Specifically, the left superior parietal lobule exhibits a positive effect, while the right rostral anterior cingulate cortex and left inferior temporal gyrus exhibit negative effects.

Small cell neuroendocrine cervical carcinoma (SCNECC) is a rare gynecological malignancy characterized by early invasion and metastasis, resulting in a poorer prognosis compared to cervical squamous cell carcinoma and adenocarcinoma. The clinical management of SCNECC remains in the exploratory phase. Recently, as this uncommon tumor has garnered increasing attention both domestically and internationally, some progress has been made in improving its prognosis. This article summarizes the advancements in combined modality therapy for SCNECC, discussing and providing insights into key issues related to current treatment strategies of surgery, radiotherapy and chemotherapy, as well as targeted and immunotherapies.

Objective: To investigate the trends of major thyroid function indices in Chinese adolescent females at different ages and the differences from adult reference intervals. Methods: A total of 791 female students from 4 junior high schools were enrolled in the study by selecting one junior high school in each of the following locations: Minhang District of Shanghai, Haimen City of Jiangsu Province, Yuhuan City of Zhejiang Province, and Deqing County of Zhejiang Province from Oct to Nov 2017 and from Jan to Mar 2019. The subjects were subjected to physical examination as well as thyroid hormone levels; and the Pubertal Developmental Events Self-Assessment Scale (PDS) was used to evaluate the staging of pubertal development. Follow-ups were conducted after 2 years, with the same survey content. Thyroid function levels were assessed in 5 age groups between 11 to 15 years old, 95% CI were calculated, and mixed linear models were used to analyze the effects of age and pubertal developmental stage on hormone levels. Results: The reference intervals for thyroid stimulating hormone (TSH), free triiodothyronine (FT3)and free thyroxine (FT4) in adolescent females differed significantly from those of adults, with misclassification rates ranging from 2.98% to 5.17%. Statistically significant differences were found for age, pubertal development staging, and the interaction of age and pubertal staging after correcting for BMI, waist circumference (all P < 0.05). TSH levels were more affected by age before the completion of pubertal development, the level of the 12-year-old group was higher than that of the 11-year-old group (P=0.001 2) and the 13-year-old group (P < 0.000 1); FT3 levels showed greater variability with age during late pubertal stage, with levels significantly higher at 13 years of age than at 11 and 12 years of age (P < 0.0001), and gradually decreasing after 13 years of age (P < 0.000 1). In contrast, FT4 levels were generally less affected by age between 11 and 15 years of age, with levels slightly higher at 13-15 years of age than at 11-12 years of age (P < 0.000 1). Conclusion: The levels of TSH, FT3, and FT4, as indicators of thyroid function in adolescent females, differ significantly from those of adults, and are differently affected by age and the stages of pubertal development; further refinement of the reference intervals for age- and pubertal-development-specific thyroid indicators is necessary.

Objective: To investigate the impact of radiotherapy simulant Zeocin induced double-stranded RNA (dsRNA)on double-stranded RNA-dependent protein kinase (PKR) activation and explore its molecular mechanism in inhibiting the growth and migration of malignant gliomas. Methods: We employed scratch assays, colony formation assays, and CCK8 assays to assess the impact of Zeocin on glioma growth and migration inhibition. Western blot and RT-qPCR were employed to measure the expression levels of methyl-modifying enzymes in glioma cells. The J2 antibody was used to detect endogenous dsRNA in malignant glioma cells treated with Zeocin. Western blot was used to assess the phosphorylation levels of PKR and eukaryotic initiating factor 2α (eIF2α). Results: Zeocin significantly inhibited growth and migration of glioma cells; Zeocin treatment induced the production of endogenous dsRNA in malignant glioma cells; as the concentration of Zeocin increased, phosphorylated PKR and eIF2α protein level also significantly increased; Zeocin downregulated the total m6A level of glioma cells in a dose-dependent manner; Zeocin selectively downregulated the protein level of methylase METTL14;Zeocin had no effect on mRNA levels of m6A modifying enzymes. Conclusion: Zeocin probably triggered more dsRNA by downregulating the m6A level of RNA in malignant glioma cells, which in turn activated the PKR/eIF2α pathway, ultimately leading to tumor growth inhibition.

The incidence of premature ovarian insufficiency (POI) is increasing and showing a trend of affecting younger women, which not only affects women's fertility, but also has the risk of low estrogen in the short and long term. The etiology of POI is complex and highly heterogeneous, including genetic factors, iatrogenic factors, autoimmune dysfunction, infection and environmental factors, but idiopathic POI still accounts for the majority. A lot of researches have focused on the correlation between autoimmune diseases and POI, among which autoimmune thyroid disease (AITD) plays an important role. Therefore, this article will review the current status of POI, the possible pathogenesis of AITD and related treatment progress.

The incidence of decreased ovarian reserve (DOR) is gradually increasing, with fluctuations in hormone levels causing a series of clinical syndromes.Among these, sleep disorders stand out as prominent symptoms, clinically manifested by difficulties in falling asleep, shortened sleep duration, and poor sleep quality, severely impacting the physical and mental health as well as the quality of life of women. Clinical observations have found a close relationship between decreased ovarian reserve and sleep disorders, where the two interact causally and reciprocally. Western medicine primarily focuses on improving behavioral cognition, administering sedatives and hypnotics, and hormone replacement therapy, albeit with certain limitations in efficacy. Traditional Chinese medicine emphasizes holistic views and syndrome differentiation in treatment, offering low side effects and unique advantages. This review summarizes recent literature on the research progress of sleep disorders and DOR from both Western and traditional Chinese medicine perspectives, providing reference for clinical treatment.

Objective: To investigate the clinical characteristics of female migraine patients with patent foramen ovale (PFO) and design a risk prediction model for PFO in female migraine patients (migraineur patients PFO risk prediction model, MPRPM). Methods: Female migraine patients who visited Zhongshan Hospital, Fudan University from Jun 1, 2019 to Dec 31, 2022 were included. Preoperative information and follow-up results after discontinuation of medication were collected. Patients were divided into PFO-positive and PFO-negative groups based on transesophageal echocardiography results. A multivariate Logistic regression model and a random forest model were constructed, and the random forest model was validated multidimensionally. Key features were selected based on the mean decrease accuracy (MDA) to construct MPRPM. Results: A total of 305 female patients were included in the study, with 204 patients in the PFO-positive group and 101 patients in the PFO-negative group. Multivariate Logistic regression analysis showed that age at migraine onset, attack frequency, severe impact on life during attacks, exercise-related headaches, menstruation-induced headaches, aura migraines, and a history of cryptogenic stroke were predictive factors for PFO positivity. The random forest model effectively predicted the incidence of PFO in female migraine patients, with an AUC of 0.895 (95%CI: 0.847-0.943). MPRPM demonstrated a sensitivity of 71.6% and specificity of 91.1% (AUC: 0.862, 95%CI: 0.818-0.906, P ＜ 0.001). The optimal cut-off value was 2.5 points. Patients correctly classified by the model showed a higher rate of symptom improvement compared to incorrectly classified patients (94.3%vs. 82.0%, P=0.023). Conclusion: We identified predictive factors for PFO in migraine patients. MPRPM can provide guidance in the diagnostic process and therapeutic decision-making for female migraine patients, assist in patient triage, and reduce the healthcare burden.

Objective: To assess the nutritional status and dietary quality of medical students and its influencing factors. Methods: Students from Shanghai Medicial College of Fudan University were recruited to complete the 7-day and 24-hour dietary records, with food consumption measured by weight. A comprehensive evaluation of the students' dietary quality was carried out by comparing their actual nutrient intake against recommended levels and calculating the component score of the Chinese Healthy Eating Index (CHEI). Factor analysis was used to extract dietary patterns. All subjects were further divided into high score group and low score group based on total CHEI. Binary Logistic regression was used to determine the factors influencing dietary quality. Results: The study participants had an average daily energy intake of (2 057.02±501.87) kcal/d, 80% from on-campus canteen meals. Carbohydrates, proteins and fats contributed to 48.90%, 16.55% and 36.07% of the total energy intake, respectively. The CHEI median score was 67.55. Component scores for tubers, vegetables, fruits, dairy, fish and seafood, and nuts were below 60%, indicating an unbalanced diet. Being in the graduate stage (OR=0.53, 95%CI: 0.28-0.98) and having unreasonable body weight expectations (OR=0.37, 95%CI: 0.17-0.81) were associated with lower CHEI scores. Conversely, higher CHEI scores were associated with the fruit-dairy dietary pattern (OR=8.20, 95%CI: 3.39-19.84), the tuber-vegetable dietary pattern (OR=3.41, 95%CI: 1.58-7.32), and lower rates of on-campus dining (OR=1.92, 95%CI: 1.02-3.59). Conclusion: The energy intake of students at Shanghai Medical College of Fudan University appears to be adequate. However, a relatively high proportion of energy supply from fat intake indicates a need for improvement in their dietary structure. The dietary quality of students is mainly influenced by their educational level, weight expectations and dietary patterns.

Objective: To develop an MRI-based habitat radiomics model for the preoperative prediction of endometrial cancer (EC) molecular subtypes. Methods: Patients with pathologically proven EC from two hospitals were included in the training (n=270) and testing (n=70) cohorts. All patients had preoperative MRI and histological and molecular diagnoses. First, the tumor was divided into habitat subregions based on diffusion-weighted imaging (DWI) and contrast-enhanced (CE) images. Subsequently, habitat radiomic features were extracted from different subregions of T1-weighted imaging (T1WI), T2-weighted imaging (T2WI), DWI, and CE images. Three machine learning classifiers, including logistic regression, support vector machines, and random forests, were applied to develop predictive models for p53-abnormal endometrial cancer, with model performance validated. The model demonstrating the best overall predictive performance was selected as the habitat radiomics model. Using the same procedure, a whole-region radiomics model based on T1WI, T2WI, DWI, and CE sequences and a clinical model were constructed. The performance of the models was evaluated using receiver operating characteristic curves, and DeLong's test was employed to compare differences between the models. Decision curve analysis was used to assess the clinical benefits of the models' application. Results: After feature selection, eight habitat radiomic features were retained to construct the habitat radiomics model, ten features for the whole-region radiomics model, and three clinical features for the clinical model. The habitat radiomics model achieved the highest area under the curve (AUC), with 0.855 (0.788-0.922) in the training cohort and 0.769 (0.631-0.907) in the testing cohort. DeLong's test showed that the habitat radiomics model outperformed the whole-region radiomics model in the training cohort (P=0.001), but there was no significant difference in the testing cohort (P=0.543). In both cohorts, the habitat radiomics model outperformed the clinical model (P=0.007, training cohort; P=0.038, testing cohort). Decision curve analysis (DCA) demonstrated that this model provided clinical benefit for diagnosis within a threshold probability range of approximately 0.2-0.8. Conclusion: The MRI-based habitat radiomics model can accurately predict p53-abnormal EC, outperforming both the whole-region radiomics model and the clinical model, and is useful for the non-invasive molecular subtyping of endometrial cancer before surgery.

This paper introduces a transfer learning approach based on regression models and demonstrates its application in the medical field through an example. Using data from the 2013-2014 U.S. National Health and Nutrition Examination Survey, the study investigates the association of sleep duration with depression levels and depressive disorder. It employs demographic characteristics and lifestyle factors as predictor variables to predict depression levels and depressive disorder across different racial groups. Compared to models built solely on target racial groups, transfer learning enhances the accuracy of estimating the effect of sleep duration in the target group and improves the prediction accuracy for depression levels and depressive disorder. The results illustrate that transfer learning effectively integrates source data to significantly improve estimation and prediction capabilities of target models, especially in situations with limited target data and heterogeneous data sources.

Objective: To explore the construction and verification of the classification model for the five properties of traditional Chinese medicine: warm, cool, cold, hot, and neutral. Methods: Urine samples of mice after taking Chinese medicine of different properties were selected as research objects, and Raman spectroscopy-related technology was used for detection. The obtained data set was classified into training set and test set, and the classification model was constructed using four machine learning methods: random forest, extreme gradient boosting, support vector machine, and logistic regression. The model performance was evaluated using precision, recall, F1 score, and accuracy. Results: A total of 4 888 sets of spectra were collected in this study, of which 80%, totaling 3 910 sets of spectral data, were used to build the model, and the remaining 20%, totaling 978 sets of spectral data, were used to test model performance. The accuracy of the random forest model was 92%, the extreme gradient boosting model was 87%, the support vector machine model was 83%, and the logistic regression model was 75%. The Raman shifts with the highest classification weights were 872, 1 012, 1 108, 1 190 and 1 668 cm^-1. Conclusion: Raman spectroscopy combined with machine learning algorithms can be used to classify the five medicinal properties of traditional Chinese medicine, among which the random forest model has the best effect.