1.Integrated Transcriptomic Landscape and Deep Learning Based Survival Prediction in Uterine Sarcomas
Yaolin SONG ; Guangqi LI ; Zhenqi ZHANG ; Yinbo LIU ; Huiqing JIA ; Chao ZHANG ; Jigang WANG ; Yanjiao HU ; Fengyun HAO ; Xianglan LIU ; Yunxia XIE ; Ding MA ; Ganghua LI ; Zaixian TAI ; Xiaoming XING
Cancer Research and Treatment 2025;57(1):250-266
Purpose:
The genomic characteristics of uterine sarcomas have not been fully elucidated. This study aimed to explore the genomic landscape of the uterine sarcomas (USs).
Materials and Methods:
Comprehensive genomic analysis through RNA-sequencing was conducted. Gene fusion, differentially expressed genes (DEGs), signaling pathway enrichment, immune cell infiltration, and prognosis were analyzed. A deep learning model was constructed to predict the survival of US patients.
Results:
A total of 71 US samples were examined, including 47 endometrial stromal sarcomas (ESS), 18 uterine leiomyosarcomas (uLMS), three adenosarcomas, two carcinosarcomas, and one uterine tumor resembling an ovarian sex-cord tumor. ESS (including high-grade ESS [HGESS] and low-grade ESS [LGESS]) and uLMS showed distinct gene fusion signatures; a novel gene fusion site, MRPS18A–PDC-AS1 could be a potential diagnostic marker for the pathology differential diagnosis of uLMS and ESS; 797 and 477 uterine sarcoma DEGs (uDEGs) were identified in the ESS vs. uLMS and HGESS vs. LGESS groups, respectively. The uDEGs were enriched in multiple pathways. Fifteen genes including LAMB4 were confirmed with prognostic value in USs; immune infiltration analysis revealed the prognositic value of myeloid dendritic cells, plasmacytoid dendritic cells, natural killer cells, macrophage M1, monocytes and hematopoietic stem cells in USs; the deep learning model named Max-Mean Non-Local multi-instance learning (MMN-MIL) showed satisfactory performance in predicting the survival of US patients, with the area under the receiver operating curve curve reached 0.909 and accuracy achieved 0.804.
Conclusion
USs harbored distinct gene fusion characteristics and gene expression features between HGESS, LGESS, and uLMS. The MMN-MIL model could effectively predict the survival of US patients.
2.Integrated Transcriptomic Landscape and Deep Learning Based Survival Prediction in Uterine Sarcomas
Yaolin SONG ; Guangqi LI ; Zhenqi ZHANG ; Yinbo LIU ; Huiqing JIA ; Chao ZHANG ; Jigang WANG ; Yanjiao HU ; Fengyun HAO ; Xianglan LIU ; Yunxia XIE ; Ding MA ; Ganghua LI ; Zaixian TAI ; Xiaoming XING
Cancer Research and Treatment 2025;57(1):250-266
Purpose:
The genomic characteristics of uterine sarcomas have not been fully elucidated. This study aimed to explore the genomic landscape of the uterine sarcomas (USs).
Materials and Methods:
Comprehensive genomic analysis through RNA-sequencing was conducted. Gene fusion, differentially expressed genes (DEGs), signaling pathway enrichment, immune cell infiltration, and prognosis were analyzed. A deep learning model was constructed to predict the survival of US patients.
Results:
A total of 71 US samples were examined, including 47 endometrial stromal sarcomas (ESS), 18 uterine leiomyosarcomas (uLMS), three adenosarcomas, two carcinosarcomas, and one uterine tumor resembling an ovarian sex-cord tumor. ESS (including high-grade ESS [HGESS] and low-grade ESS [LGESS]) and uLMS showed distinct gene fusion signatures; a novel gene fusion site, MRPS18A–PDC-AS1 could be a potential diagnostic marker for the pathology differential diagnosis of uLMS and ESS; 797 and 477 uterine sarcoma DEGs (uDEGs) were identified in the ESS vs. uLMS and HGESS vs. LGESS groups, respectively. The uDEGs were enriched in multiple pathways. Fifteen genes including LAMB4 were confirmed with prognostic value in USs; immune infiltration analysis revealed the prognositic value of myeloid dendritic cells, plasmacytoid dendritic cells, natural killer cells, macrophage M1, monocytes and hematopoietic stem cells in USs; the deep learning model named Max-Mean Non-Local multi-instance learning (MMN-MIL) showed satisfactory performance in predicting the survival of US patients, with the area under the receiver operating curve curve reached 0.909 and accuracy achieved 0.804.
Conclusion
USs harbored distinct gene fusion characteristics and gene expression features between HGESS, LGESS, and uLMS. The MMN-MIL model could effectively predict the survival of US patients.
3.Integrated Transcriptomic Landscape and Deep Learning Based Survival Prediction in Uterine Sarcomas
Yaolin SONG ; Guangqi LI ; Zhenqi ZHANG ; Yinbo LIU ; Huiqing JIA ; Chao ZHANG ; Jigang WANG ; Yanjiao HU ; Fengyun HAO ; Xianglan LIU ; Yunxia XIE ; Ding MA ; Ganghua LI ; Zaixian TAI ; Xiaoming XING
Cancer Research and Treatment 2025;57(1):250-266
Purpose:
The genomic characteristics of uterine sarcomas have not been fully elucidated. This study aimed to explore the genomic landscape of the uterine sarcomas (USs).
Materials and Methods:
Comprehensive genomic analysis through RNA-sequencing was conducted. Gene fusion, differentially expressed genes (DEGs), signaling pathway enrichment, immune cell infiltration, and prognosis were analyzed. A deep learning model was constructed to predict the survival of US patients.
Results:
A total of 71 US samples were examined, including 47 endometrial stromal sarcomas (ESS), 18 uterine leiomyosarcomas (uLMS), three adenosarcomas, two carcinosarcomas, and one uterine tumor resembling an ovarian sex-cord tumor. ESS (including high-grade ESS [HGESS] and low-grade ESS [LGESS]) and uLMS showed distinct gene fusion signatures; a novel gene fusion site, MRPS18A–PDC-AS1 could be a potential diagnostic marker for the pathology differential diagnosis of uLMS and ESS; 797 and 477 uterine sarcoma DEGs (uDEGs) were identified in the ESS vs. uLMS and HGESS vs. LGESS groups, respectively. The uDEGs were enriched in multiple pathways. Fifteen genes including LAMB4 were confirmed with prognostic value in USs; immune infiltration analysis revealed the prognositic value of myeloid dendritic cells, plasmacytoid dendritic cells, natural killer cells, macrophage M1, monocytes and hematopoietic stem cells in USs; the deep learning model named Max-Mean Non-Local multi-instance learning (MMN-MIL) showed satisfactory performance in predicting the survival of US patients, with the area under the receiver operating curve curve reached 0.909 and accuracy achieved 0.804.
Conclusion
USs harbored distinct gene fusion characteristics and gene expression features between HGESS, LGESS, and uLMS. The MMN-MIL model could effectively predict the survival of US patients.
4.Glutamate Receptor Antagonists Attenuate Stereotyped Behaviors via Modulating BDNF Levels in Obsessive-complusive Disorder Model Mice
Weijie WANG ; Yuchong LUO ; Dongmiao HUANG ; Chen YANG ; Jihui YUE ; Xianglan WANG ; Shenglin WEN
Journal of Sun Yat-sen University(Medical Sciences) 2025;46(3):475-485
ObjectiveTo explore whether fluoroethylnormemantine (FENM), an NMDA receptor antagonist, could improve compulsive-like behaviors and to investigate its underlying mechanisms in the RU24969-induced obsessive-compulsive disorder (OCD) mouse model. MethodsThirty-two mice were randomly assigned to four groups: Saline (n=8), RU24969 (n=8), RU+FENM (n=8), and FENM (n=8). Mice received FENM or an equivalent volume of saline for pre-treatment, followed by RU24969 or saline for model induction 30 minutes later. Behavioral tests were performed 1 hour after modeling, and serum samples were collected to measure the level of brain-derived neurotrophic factor (BDNF). Evans Blue dye was intravenously injected to assess dye content in brain tissue, thereby evaluating potential blood-brain barrier damage. ResultsFENM treatment significantly improved repetitive stereotyped circling behavior (F=39.850, P<0.001) and alleviated persistent motor activity (F=50.200, P<0.001) in RU24969 model mice. Additionally, FENM treatment significantly increased serum BDNF level in RU24969-induced OCD mice (F=18.930, P<0.001). ConclusionsFENM , an NMDA receptor antagonist, may alleviate compulsive behaviors in OCD mice by modulating BDNF levels , thereby exerting anti-compulsive effects. Neither the RU24969 model nor FENM treatment significantly affectes blood-brain barrier integrity.
5.Characteristics and influencing factors of hearing loss among noise-exposed workers in a large machinery maintenance enterprise
Fang JI ; Jian ZHANG ; Xiaowen DING ; Xianglan CUI ; Li RONG ; Xuefeng WANG ; Jue LI
Journal of Environmental and Occupational Medicine 2024;41(1):77-82
Background The current increasing trend of new cases of occupational noise-induced deafness indicates that the hearing loss of occupational population has not been effectively controlled in China. It is of great significance to study the characteristics of hearing loss among noise-exposed workers and its related factors. Objective To investigate characteristics and influencing factors of hearing loss among occupational noise-exposed workers in a large machinery maintenance enterprise, and to provide a scientific basis to prevent and control noise-induced hearing loss. Methods A cross-sectional survey was conducted to investigate male Han occupational noise-exposed workers in a large mechanical maintenance enterprise. We acquired demographic characteristics, occupational exposure history, and individual life behavior characteristics of the workers through questionnaires, collected occupational exposure level data from annual occupational disease hazard factor surveillance reports, obtained pure tone hearing threshold test data through occupational health examinations, and estimated individual noise exposure levels using cumulative noise exposure (CNE). According to the results of pure tone air conduction hearing threshold test, the workers were divided into a hearing loss group and a normal hearing group. The chi-square test was employed to compare the occupational exposure characteristics and individual life behavior characteristics between the two groups. Additionally, the trend chi-square test was utilized to analyze the changing trends of age, length of service, CNE, and hearing loss rate within the two groups. The relationship between high-frequency hearing loss in both ears and its related influencing factors was assessed by a multiple logistic regression model. Results The M (P25, P75) of CNE for the 2531 occupational noise-exposed workers was 97.51 (95.39, 99.96) dB(A)·year. The incidence of hearing anomaly, binaural high-frequency hearing anomaly, random ear high-frequency hearing anomaly, binaural low-frequency hearing anomaly, and random ear low-frequency hearing anomaly were 22.48%, 16.59%, 22.13%, 2.77%, and 3.52%, respectively. High-frequency hearing threshold increase was the main reason for hearing anomaly (98.42%). In comparison to the CNE ≤ 97 dB(A)·year group, the 97 dB(A)·year
6.Influence of automated flexible endoscope channel brushing system on endoscopic cleaning quality
Xianglan WANG ; Renduo SHANG ; Jun LIU ; Xingmin HUANG ; Zi LUO ; Xuan CAI ; Honggang YU
Chinese Journal of Digestive Endoscopy 2024;41(2):142-146
Objective:To evaluate the effect of automated flexible endoscope channel brushing system (AFECBS) on endoscope reprocessing.Methods:A prospective randomized controlled study was conducted. The used endoscopes were divided into automatic group and manual group by random number table method, 200 in each group. In the automatic group, the AFECBS was used to scrub each tube 3 times during endoscope cleaning; and in the manual group, scrubbing and disinfection personnel routinely brushed each pipeline for 3 times. The primary end point was the qualified rate of endoscopic cleaning quality in the two groups, and the secondary end point was the time spent by the scrubbing and disinfection personnel on the two groups.Results:The qualified rate of overall cleaning in the automatic group was 90.0% (180/200), and in the manual group was 81.0% (162/200). The qualified rate of the automatic group was higher than that of the manual group ( χ2=6.534, P=0.011). The qualified rate of gastroscope cleaning in the automatic group was higher than that in the manual group [92.0% (127/138) VS 81.6% (120/147), χ2=6.658, P=0.010]. There was no significant difference in the qualified rate of colonoscope cleaning between the automatic group and the manual group [85.5% (53/62) VS 79.2% (42/53), χ2=0.774, P=0.379]. When the cleaning personnel scoured 5 endoscopes in each of the two groups, the time of the automatic group (5.17±0.42 min) was shorter than that of the manual group (9.60±0.53 min) ( t=92.644, P<0.001). Conclusion:Compared with manual scrubbing, AFECBS can improve the qualified rate of endoscope cleaning and the work efficiency of scrubbing and disinfection personnel, which is worthy of clinical application.
7.Investigation of dementia awareness among stroke patients in China:A cross-sectional study
Zhou JING ; Wang YING ; Zhao JIAYI ; Chen BAOXIN ; Liu XUEMEI ; Fu CHEN ; Hou XIAOBING ; Niu HUANMIN ; Zhang YUNLING ; Jin XIANGLAN
Journal of Traditional Chinese Medical Sciences 2021;8(2):155-160
Objective: To provide guidelines for the prevention of post-stroke dementia, we investigated whether stroke patients were aware of post-stroke dementia and their level of dementia-related knowledge. Methods: Five dementia-related questions were designed. A field survey was conducted in a question-and-answer method using the designed questions. The surveys assessed 3000 stroke patients (2 weeks-6 months after stroke attacks) from 14 hospitals/clinical centers in 7 provinces and cities across China. Results: Among 3000 stroke patients, 80.5%had heard of dementia, 39.1%knew that stroke can lead to dementia, 55.7% believed dementia can be prevented, 50.8% thought dementia can be cured, and only 8.8%had ever seen a doctor because of memory deterioration. Then, patients were classified into three groups, including a no cognitive impairment (NCI) group, a mild cognitive impairment (MCI) group, and a mild dementia (MD) group. Among the MCI and MD groups, only 8.7% (75/861) and 9.9% (64/649) of patients, respectively, had ever seen a doctor because of memory deterioration. According to our results, patients with a higher level of cognitive impairment had a lower awareness of dementia (P<.001). Conclusion: The awareness of dementia in stroke patients in China is low, and the consultation rate is even lower. Moreover, patients with a higher level of cognitive impairment have a lower awareness of dementia. To improve public awareness and improve prevention, more emphasis should be put on ed-ucation regarding post-stroke dementia. Routine cognitive function screening should be conducted on stroke patients as an effective way to assess dementia.
8.Prevalence of eating behavior problems among preschool children in rural Hunan and its determinants
LI Huixia, WANG Hua, HUANG Guangwen, ZHANG Xianglan, HUANG Qun, XIAO Juan, FENG Na
Chinese Journal of School Health 2020;41(5):692-696
Objective:
To understand the prevalence of eating behavior problems among preschool children in rural Hunan, and to explore its determinants, so as to provide a reference for solving their eating behavior problems.
Methods:
A multistage stratified cluster sampling method was used to randomly select 3 288 preschool children aged 3-6 years and their caregivers from 24 kindergartens between August and November 2019. Eating behavior problems questionnaire survey and peripheral blood hemoglobin detection were conducted among the preschool children, and unconditional Logistic regression model was used to analyze the determinants.
Results:
The total prevalence of eating behavior problems among preschool children in rural Hunan was 66.8%(2 195/3 288), and the three most common eating behavior problems were poor appetite(40.1%, 1 317/3 288), picky eating(43.6%, 1 434/3 288), poor eating habits(44.4%, 1 460/3 288), respectively. The risk factors for general eating behavior problems included left-behind children, caregiver occupation as professional and technical staffs, average daily static activity time≥3 hours, anemia, and second-hand smoke exposure, and older age and daily moderate/high intensity activities were protective factors. In the analysis of the three most common eating behavior problems: The risk factors for poor appetite included left-behind children, anemia, and second-hand smoke exposure, and daily moderate/high intensity activities was protective factors. The risk factors for picky eating included left-behind children, caregiver occupation as professional and technical staffs, high family income, average daily static activity time≥3 hours, and daily moderate/high intensity activities was protective factors. The risk factors for poor eating habits included left-behind children, caregiver occupation as professional and technical staffs, average daily static activity time≥3 hours, and older age and daily moderate/high intensity activities was protective factors.
Conclusion
The prevalence of eating behavior problems among preschool children in rural Hunan was high, which is influenced by age, left-behind, caregivers occupation, family income, activities, anemia, and second-hand smoke exposure.
9.The Gene Polymorphism of VMAT2 Is Associated with Risk of Schizophrenia in Male Han Chinese
Hongying HAN ; Xiaowei XIA ; Huirong ZHENG ; Chongbang ZHAO ; Yanming XU ; Jiong TAO ; Xianglan WANG
Psychiatry Investigation 2020;17(11):1073-1078
Objective:
To investigate the association between gene polymorphism of vesicular monoamine transporter type 2(VMAT2) and schizophrenia in Han Chinese population.
Methods:
430 patients with schizophrenia and 470 age-sex matched controls were recruited from four mental health centers. All patients were diagnosed by two psychiatrists based on the Structured Clinical Interview for DSM Disorders (SCID). The ligase detection reactions (LDR) method was used to assess the polymorphism of the two SNPs (rs363371 and rs363324) of VMAT2.
Results:
No associations of two SNPs with schizophrenia was found. When we stratified males and females for the analysis, we found that that in the recessive model of rs363371, there was an obvious significant association between rs363371 and schizophrenia in males (OR=0.564, 95% CI=0.357–0.892, p=0.014) but not females. For the association between rs363324 and schizophrenia, no association was found in either males or females. No association was found when stratifying early-onset schizophrenia and late-onset schizophrenia.
Conclusion
Our findings indicate that both rs363371 and rs363324 were not associated with schizophrenia, while it seemed that the AA genotype of rs363371 plays a protective effect in male Chinese in developing schizophrenia.
10. Association between behavioral problems and parenting stress in preschool children
Hengheng ZHONG ; Xianglan WANG
Chinese Journal of Practical Nursing 2018;34(21):1631-1634
Objective:
To analyze the relationship between parenting stress in parents and behavioral problems in preschool children, to provide scientific basis for prevention and intervention of preschool children′s behavioral problems.
Methods:
A sample of 208 preschool children were selected from Changsha. Parents were invited to complete the Conners Child Behavior Checklist (Parent Question) and Parenting Stress Index (PSI).
Results:
The proportion of preschool children with behavioral problems in this surveyed sample was 19.71% (41/208). The rate of higher level parenting stress was 19.71% (41/208). Total parenting stress scores in preschool children with behavioral problems were 83.46±16.76, which were higher than the scores 77.07±15.02 in those without behavioral problems (


Result Analysis
Print
Save
E-mail