1.Graph Neural Networks and Multimodal DTI Features for Schizophrenia Classification: Insights from Brain Network Analysis and Gene Expression.
Jingjing GAO ; Heping TANG ; Zhengning WANG ; Yanling LI ; Na LUO ; Ming SONG ; Sangma XIE ; Weiyang SHI ; Hao YAN ; Lin LU ; Jun YAN ; Peng LI ; Yuqing SONG ; Jun CHEN ; Yunchun CHEN ; Huaning WANG ; Wenming LIU ; Zhigang LI ; Hua GUO ; Ping WAN ; Luxian LV ; Yongfeng YANG ; Huiling WANG ; Hongxing ZHANG ; Huawang WU ; Yuping NING ; Dai ZHANG ; Tianzi JIANG
Neuroscience Bulletin 2025;41(6):933-950
Schizophrenia (SZ) stands as a severe psychiatric disorder. This study applied diffusion tensor imaging (DTI) data in conjunction with graph neural networks to distinguish SZ patients from normal controls (NCs) and showcases the superior performance of a graph neural network integrating combined fractional anisotropy and fiber number brain network features, achieving an accuracy of 73.79% in distinguishing SZ patients from NCs. Beyond mere discrimination, our study delved deeper into the advantages of utilizing white matter brain network features for identifying SZ patients through interpretable model analysis and gene expression analysis. These analyses uncovered intricate interrelationships between brain imaging markers and genetic biomarkers, providing novel insights into the neuropathological basis of SZ. In summary, our findings underscore the potential of graph neural networks applied to multimodal DTI data for enhancing SZ detection through an integrated analysis of neuroimaging and genetic features.
Humans
;
Schizophrenia/pathology*
;
Diffusion Tensor Imaging/methods*
;
Male
;
Female
;
Adult
;
Brain/metabolism*
;
Young Adult
;
Middle Aged
;
White Matter/pathology*
;
Gene Expression
;
Nerve Net/diagnostic imaging*
;
Graph Neural Networks
2.Risk factor analysis of postoperative gastrointestinal dysfunction after infant intestinal surgery
Zichuan GAO ; Hongxing LI ; Weibing TANG
Journal of Clinical Surgery 2025;33(5):466-469
Objective This study analyzes the risk factors for postoperative gastrointestinal dysfunction(POGD)in infants after intestinal surgery.Methods The perioperative clinical data of 220 infants who underwent intestinal surgery in Children's Hospital of Nanjing Medical University from March 2019 to March 2024 were retrospectively analyzed.Based on the criteria of postoperative"fasting time more than 5 d,vomiting bile fluid or bile fluid drainage from nasogastric tube,abdominal distension",the infants were divided into POGD group(78 cases)and non-POGD group(142 cases),and the clinical data of the two groups were analyzed by one-way analysis,and the independent factors with a P<0.05 were included in the binary Logistic regression analysis to analyze the independent influencing factors affecting the recovery of gastrointestinal function after surgery.Results The time of the first defecation after surgery in the POGD group and the non-POGD group was(1.79±1.78)days and(1.44±0.71)days,respectively,and the time of parenteral nutrition was(9.73±4.64)days and(5.19±2.18)days,respectively,and the hospital stay was(13.31±5.70)days and(8.46±2.90)days,respectively.There was statistically significant difference between the two groups(P<0.05).Multivariate regression analysis showed that a low preoperative weight for age Z(WAZ)score(P<0.05,OR=0.705)and a long operation time(P<0.05,OR=2.642)were risk factors for POGD.Conclusion Preoperative nutritional status and duration of surgery are risk factors for the development of postoperative gastrointestinal dysfunction in infants.
3.Risk factor analysis of postoperative gastrointestinal dysfunction after infant intestinal surgery
Zichuan GAO ; Hongxing LI ; Weibing TANG
Journal of Clinical Surgery 2025;33(5):466-469
Objective This study analyzes the risk factors for postoperative gastrointestinal dysfunction(POGD)in infants after intestinal surgery.Methods The perioperative clinical data of 220 infants who underwent intestinal surgery in Children's Hospital of Nanjing Medical University from March 2019 to March 2024 were retrospectively analyzed.Based on the criteria of postoperative"fasting time more than 5 d,vomiting bile fluid or bile fluid drainage from nasogastric tube,abdominal distension",the infants were divided into POGD group(78 cases)and non-POGD group(142 cases),and the clinical data of the two groups were analyzed by one-way analysis,and the independent factors with a P<0.05 were included in the binary Logistic regression analysis to analyze the independent influencing factors affecting the recovery of gastrointestinal function after surgery.Results The time of the first defecation after surgery in the POGD group and the non-POGD group was(1.79±1.78)days and(1.44±0.71)days,respectively,and the time of parenteral nutrition was(9.73±4.64)days and(5.19±2.18)days,respectively,and the hospital stay was(13.31±5.70)days and(8.46±2.90)days,respectively.There was statistically significant difference between the two groups(P<0.05).Multivariate regression analysis showed that a low preoperative weight for age Z(WAZ)score(P<0.05,OR=0.705)and a long operation time(P<0.05,OR=2.642)were risk factors for POGD.Conclusion Preoperative nutritional status and duration of surgery are risk factors for the development of postoperative gastrointestinal dysfunction in infants.
4.Named Entity Recognition of Traditional Chinese Medicine Ancient Records Based on Multi-feature Fusion
Luyao ZHANG ; Jianhua SHU ; Peng WANG ; Hongxing KAN ; Yongxiang XU ; Jie ZHOU ; Shuxuan TANG
Journal of Medical Informatics 2024;45(11):50-58
Purpose/Significance To construct a named entity corpus of traditional Chinese medicine(TCM)ancient records,and to improve the recognition accuracy and applicability of the general domain named entity recognition(NER)model in the field of TCM ancient records.Method/Process Annotation standards for entities in TCM ancient records are formulated,and 2 384 Xin'an medical records are annotated.A RoBERTa-BiLSTM-CRF model is developed,and word vectors with semantic features are generated using the RoBERTa pre-trained language model.The BiLSTM-CRF model is used to learn the global semantic features of sequences and decode and output the optimal label sequence.Dictionary and rule features are incorporated to enhance the model's capability to recognize entity boundaries and categories.Result/Conclusion The model shows a good recognition effect on the named entity corpus of Xin'an medical cases.Integration of domain terminology dictionaries and rule-based features improves the overall Fl score to 72.8%.
5.Research on the Intelligent Assisted Diagnosis and Treatment System of Xin'an Medicine Based on Artificial Intelligence
Shuxuan TANG ; Yongxiang XU ; Jie ZHOU ; Luyao ZHANG ; Peng WANG ; Hongxing KAN ; Fudong NIAN ; Jianhua SHU
Journal of Nanjing University of Traditional Chinese Medicine 2024;40(12):1348-1356
OBJECTIVE To develop an artificial intelligence-based intelligent auxiliary diagnosis and treatment system for Xin'an medicine to address the challenges of integrating ancient Xin'an medical case records into modern clinical applications.METHODS The project involved structuring and standardizing case records from ancient texts of Xin'an medicine to build a compre-hensive Xin'an medicine database.Advanced techniques,such as data annotation,entity relationship extraction,and data mining,were applied to create a Xin'an medicine knowledge base.Furthermore,a knowledge graph of Xin'an medicine was constructed using techniques for knowledge acquisition,integration,storage,and graph-based question-answering,improving the efficiency of knowl-edge organization and retrieval.The LangChain framework was utilized to connect the Xin'an medicine knowledge base to a large lan-guage model,enabling a model-driven local knowledge base question-answering system.RESULTS The study successfully estab-lished a systematic and standardized knowledge base for Xin'an medical case records.The application of knowledge graph technology provided a clear visualization of Xin'an medicine's knowledge structure,and the development of an intelligent question-answering module significantly improved the efficiency of knowledge management and retrieval.The local knowledge base question-answering sys-tem,powered by a large language model and based on Xin'an medicine's theoretical and practical expertise,delivered accurate diag-nostic and treatment support,promoting the heritage and innovation of Xin'an medicine.CONCLUSION This research validates the feasibility of modernizing traditional medical texts and provides an innovative approach to knowledge development and clinical applica-tion in Chinese medicine.The findings have significant academic value and promising clinical implications.
6.Research on the Intelligent Assisted Diagnosis and Treatment System of Xin'an Medicine Based on Artificial Intelligence
Shuxuan TANG ; Yongxiang XU ; Jie ZHOU ; Luyao ZHANG ; Peng WANG ; Hongxing KAN ; Fudong NIAN ; Jianhua SHU
Journal of Nanjing University of Traditional Chinese Medicine 2024;40(12):1348-1356
OBJECTIVE To develop an artificial intelligence-based intelligent auxiliary diagnosis and treatment system for Xin'an medicine to address the challenges of integrating ancient Xin'an medical case records into modern clinical applications.METHODS The project involved structuring and standardizing case records from ancient texts of Xin'an medicine to build a compre-hensive Xin'an medicine database.Advanced techniques,such as data annotation,entity relationship extraction,and data mining,were applied to create a Xin'an medicine knowledge base.Furthermore,a knowledge graph of Xin'an medicine was constructed using techniques for knowledge acquisition,integration,storage,and graph-based question-answering,improving the efficiency of knowl-edge organization and retrieval.The LangChain framework was utilized to connect the Xin'an medicine knowledge base to a large lan-guage model,enabling a model-driven local knowledge base question-answering system.RESULTS The study successfully estab-lished a systematic and standardized knowledge base for Xin'an medical case records.The application of knowledge graph technology provided a clear visualization of Xin'an medicine's knowledge structure,and the development of an intelligent question-answering module significantly improved the efficiency of knowledge management and retrieval.The local knowledge base question-answering sys-tem,powered by a large language model and based on Xin'an medicine's theoretical and practical expertise,delivered accurate diag-nostic and treatment support,promoting the heritage and innovation of Xin'an medicine.CONCLUSION This research validates the feasibility of modernizing traditional medical texts and provides an innovative approach to knowledge development and clinical applica-tion in Chinese medicine.The findings have significant academic value and promising clinical implications.
7.Efficacy of microsurgical treatment in 58 patients with primary jugular foramen schwannomas
Guochen YANG ; Xixi LI ; Weijie SU ; Honglin WU ; Hongxing TANG ; Zhong DENG ; Yibin YANG ; Lixuan YANG
Chinese Journal of Neuromedicine 2022;21(11):1138-1142
Objective:To investigate the clinical efficacy of microsurgical resection in primary jugular foramen schwannomas (JFSs).Methods:A retrospective analysis was performed; the clinical data of 58 patients with JFSs treated by microsurgery in Department of Neurosurgery, First Affiliated Hospital of Sun Yat-sen University from May 2012 to June 2021 were collected. Seven patients accepted microsurgery via suboccipital retrosigmoid approach, and 51 patients accepted microsurgery via jugular foramen approach. Fifty-three patients were followed up for 4.5 years (ranged from 0.5-8.5 years); follow-ups included Karnofsky performance status (KPS) scores, postoperative complications, and imaging reexaminations. Results:Fifty patients (86.2%) achieved total tumor resection and 8 (13.8%) subtotal resection. The KPS scores at discharge were 68.6±14.9, which were significantly lower than the preoperative KPS scores (77.6±13.5, t=2.452, P=0.017). During the follow-up, 5 patients(9.4%) had tumor recurrence, and 39 patients (73.6%) had improved symptoms after surgery. One patient (1.9%, modified Samii D type) died of cerebellar hemorrhage and swelling after surgery. The main complications included new/aggravated hoarseness (11/53), cerebrospinal fluid leakage (7/53), new/aggravated dysphagia (5/53), and new facial paralysis (4/53). Conclusion:In microsurgical resection of JFSs, short-term symptoms of the lower cranial nerves may be exacerbated, but long-term results are good.
8.Electroacupuncture activates phosphorylated AMPKαto regulate the expression of the mTOR gene in rats with functional dyspepsia
Lei TANG ; Hongxing ZHANG ; Paidi XU ; Zhaoxia KANG
Chinese Journal of Physical Medicine and Rehabilitation 2018;40(11):809-813
Objective To investigate how electroacupuncture activates the AMPKαpathway in the treatment of functional dyspepsia ( FD) . Methods Tail clipping stimulation was combined with an irregular diet to induce FD in 20 Sprague-Dawley rats. They were randomly divided into a model group and an electroacupuncture group, each of 10. Ten others had no FD induced and formed a normal group. The rats in the electroacupuncture group were given 10 days of electroacupuncture treatment, while those in the normal and model groups were restrained and fixed like those in the electroacupuncture group, but not given electroacupuncture. The expression of phosphorylated adenosine mono-phosphate-activated protein kinase alpha (p-AMPKα), phosphorylated tuberin (p-TSC2) and Rheb protein in the stomachs and small intestines of each group were detected using the western blotting. Levels of mTOR mRNA were quantified using the reverse-transcription polymerase chain reaction ( RT-PCR) . Results The western blotting re-sults showed that compared with the normal group, the expression levels of p-AMPKα and p-TSC2 in the stomachs and small intestines of the model group decreased significantly, while that of Rheb protein increased significantly. Compared with the model group, the expression of p-AMPKαand p-TSC2 increased significantly after the electroacu-puncture treatment, while that of Rheb protein decreased significantly. According to the RT-PCR results, the expres-sion of mTOR mRNA in the stomachs and small intestines of the model group was up-regulated significantly compared with the normal group. Compared with the model group, expression of mTOR mRNA in the electroacupuncture group was significantly down-regulated. Conclusion Electroacupuncture can activate the AMPKα pathway, up-regulate the expression of the related protein p-TSC2, and decrease the expression of Rheb protein, thereby down-regulating the transcription of the mTOR gene. That would serve to treat FD.
9.Application of bowel plication combined with early enteral nutrition in the enhanced recovery after surgery for neonates with jejunal atresia.
Huan CHEN ; Qiming GENG ; Changgui LU ; Weiwei JIANG ; Jie ZHANG ; Xiaofeng LYU ; Wei LI ; Hongxing LI ; Weibing TANG
Chinese Journal of Gastrointestinal Surgery 2017;20(5):535-539
OBJECTIVETo evaluate the efficacy of bowel plication combined with early enteral nutrition (EEN) in the enhanced recovery after surgery(ERAS) of jejunal atresia (JA) neonates.
METHODSBetween January 2005 and January 2014, 58 neonates with JA underwent surgical treatment in Children's Hospital of Nanjing Medical University. Their clinical data, including operation procedures, ages, birth weight, concomitant diseases, age at surgery, hospital stay, total parenteral nutrition (TPN), postoperative intestinal function recovery (the time to the first oral feeding and the time to oral feeding volume reaching 150 ml·kg·d), complications and reoperation, were retrospectively analyzed.
RESULTSAccording to the surgical procedures, the 58 neonates were divided into three groups: control group(18 cases, undergoing atretic segments resection and primary anastomosis), bowel plication group(19 cases, undergoing bowel plication after atretic segments resection and primary anastomosis) and bowel plication combined with EEN group (21 cases, undergoing bowel plication combined with EEN). No significant differences of ages, birth weight, age at operation, and concomitant diseases were found among 3 groups (all P>0.05). The time of hospital stay, the time to the first oral feeding, the time to oral feeding volume reaching 150 ml·kg·d, and the time of TPN in bowel plication group were significantly shorter than those of control group [(19.3±4.4) d vs. (22.7±3.1) d, t=2.696, P=0.011; (9.8±3.3) d vs. (12.5±3.0) d, t=2.630, P=0.013; (18.5±4.1) d vs. (21.5±2.5) d, t=2.726, P=0.011; (13.1±2.9) d vs. (15.0±2.3) d, t=2.219, P=0.033]. However, above parameters of bowel plication combined with EEN group were significantly shorter than those of bowel plication group [(15.3±3.5) d vs. (19.3±4.4), t=4.120, P=0.003; (7.7±2.2) d vs. (9.8±3.3) d, t=2.428, P=0.020; (14.8±2.5) d vs. (18.5±4.1) d, t=3.752, P=0.001; (9.5±3.0) vs. (13.1±2.9) d, t=4.370, P=0.000].
CONCLUSIONThe bowel plication combined with EEN contributes to the early use of intestinal function, shorten the time to the first oral feeding, and reduces the use of TPN, which can improve the recovery of jejunal atresia neonates.
Anastomosis, Surgical ; Comparative Effectiveness Research ; Defecation ; Digestive System Surgical Procedures ; methods ; Enteral Nutrition ; methods ; Humans ; Infant, Newborn ; Intestinal Atresia ; rehabilitation ; surgery ; Jejunum ; abnormalities ; surgery ; Length of Stay ; Parenteral Nutrition, Total ; Postoperative Period ; Retrospective Studies
10.Screening and expression analysis of the specific tissue and serum microRNA profile in Hirschsprung disease
Hua XIE ; Hongxing LI ; Qiming GENG ; Xiaoqun XU ; Weiwei JIANG ; Jie ZHANG ; Huan CHEN ; Weibing TANG
Chinese Journal of Applied Clinical Pediatrics 2016;31(5):380-383
Objective To investigate the different expressions of pathological tissue and serum microRNAs (miRNAs)in Hirschsprung disease(HSCR). Methods Pathological colon tissues and serum samples were obtained from 52 confirmed HSCR cases respectively by surgery and pathology and from 52 matched controls,respectively. An initial screening of the tissues and serum microRNA expression were performed through TaqMan Low Density Array. The candidate tissue and serum miRNAs were validated by quantitative real - time - PCR in the 20 paired array samples and extra 32 paired samples after the integration of the screening result. The bioinformatical software online including miR-base,Target Scan,PicTar and MiRanda were used to predict the target mRNA of the consistent microRNAs in the tis-sues and the serum. Results Compared with the controls,47 microRNAs were differently expressed in HSCR tissues, including 17 up - regulated miRNAs and 30 down - regulated miRNAs;32 upregulated miRNAs were also detected to be differently expressed in the HSCR serum. Among these microRNAs,miR - 218 - 1 and miR - 885 - 5p were identi-fied to have a consistent significant different expression in both tissues and the serum,which were validated as high -expressed in microarray samples and expanded 32 paired samples(miR - 218 - 1:tissue array 0. 017 58 ± 0. 002 29 vs 0. 003 37 ± 0. 000 50,P ﹤ 0. 001;tissue expanded expression 0. 013 53 ± 0. 001 74 vs 0. 004 43 ± 0. 000 60,P ﹤0. 001. miR - 885 - 5p:tissue array 0. 000 30 ± 0. 000 11 vs 0. 000 04 ± 0. 0000 08,P = 0. 027 6;tissue expanded ex-pression 0. 004 59 ± 0. 000 16 vs 0. 000 04 ± 0. 000 01,P = 0. 014 5. miR - 218 - 1:serum array 0. 769 60 ± 0. 285 50 vs 0. 045 14 ± 0. 015 07,P = 0. 015 5;serum expanded expression 1. 151 00 ± 0. 430 00 vs 0. 023 07 ± 0. 003 81,P =0. 008 7. miR -885 -5p:serum array 1. 595 00 ±0. 441 70 vs 0. 169 40 ±0. 034 46,P =0. 001 2;serum expanded expres-sion 1. 689 00 ±0. 453 00 vs 0. 146 10 ± 0. 031 24,P = 0. 001 2). Specifically,the target genes of these 2 microRNAs were RET,PLAG1 and NeuroD1,which had been reported to be directly related to HSCR. Conclusions Significantly dif-ferential expressed miRNAs exist in the pathological tissue and the serum of HSCR. MiR - 218 - 1 and miR - 885 - 5p, which showing consistent differential expression,may be involved in the pathogenesis of HSCR.

Result Analysis
Print
Save
E-mail