Objective:To retrospectively evaluate the clinical efficacy and safety of total 3D laparo-scopic ileal ureters replacement for bilateral ureters combined with bladder augmentation in patients with post-radiotherapy long-segment bilateral ureteral strictures and contracted bladder.Methods:Clinical data of two patients(aged 72 and 54 years)with radiation-induced long-segment bilateral ureteral stric-tures and reduced bladder capacity,treated at the Sixth Affiliated Hospital of Jinan University from Octo-ber 2023 to June 2024,were analyzed.Both presented with bilateral flank pain,recurrent chills/fever,urinary frequency,and urgency.Preoperative ureteral stricture lengths were measured as follows:left 10.4 cm and right 8.7 cm in the first case;left 10.6 cm and right 11.7 cm in the second case.Bladder capacity assessed by nephrostomy-assisted antegrade urography was 90 mL and 130 mL respectively.Both underwent single-position,one-stage totally 3D laparoscopic bilateral ileal ureteral replacement and blad-der augmentation based on membrane anatomy principles,with regular postoperative follow-up.Results:Procedures were completed by the same experienced urologist.Operative times were 420 min and 355 min,with intraoperative blood loss of 50 mL(no transfusion required).Postoperative bowel function re-sumed at the end of 4.5 and 3 days.No major perioperative complications occurred.Ureteral stents were removed at 2 months postoperatively,with imaging showing improved hydronephrosis,unobstructed ure-teral drainage,symmetrical bladder morphology,and smooth walls.Postoperative bladder capacities were 230 mL and 250 mL.Follow-up durations were 10 and 8 months.Both patients experienced significant relief of flank pain and lower urinary tract symptoms.No complications(enteric fistula,urinary fistula,or metabolic acidosis)were observed.At the final follow-up,one patient had mildly elevated serum cre-atinine,while the other showed reduced levels compared with preoperative values;both remained stable.Conclusion:Membrane anatomy-based dissection facilitates safe mobilization of fibrotic ureters with mini-mal bleeding and collateral damage.Total intracorporeal 3D laparoscopic ileal ureters replacement for bi-lateral ureters combined with bladder augmentation effectively addresses long-segment ureteral obstruction and improves bladder capacity.This approach is technically safe and feasible,though further validation with larger clinical cohorts is warranted.

Objective:To establish a model based on the characteristics of breast cancer ultrasound images through deep learning methods to predict the risk of axillary lymph node metastasis（ALNM）in patients with breast cancer（maximum diameter ≤2.5 cm）before surgery.Methods:A total of 419 patients（3 433 breast tumor ultrasound images）with breast cancer（maximum diameter ≤2.5 cm）who underwent axillary lymph node dissection at Beijing Tiantan Hospital，Capital Medical University from January 2019 to December 2024 were retrospectively included. According to the pathological results of axillary lymph nodes，they were divided into 220 cases in the ALNM occurrence group（positive group）and 199 cases in the non-ALNM occurrence group（negative group）. The breast cancer ultrasound images of the two groups of cases were randomly classified into the training set（2 404 images），the validation set（687 images）and the test set（342 images）according to a ratio of 7∶2∶1. YOLOv8 was used as the basic model of You Only Look Once（YOLO）and optimized. The optimized model was applied to locate and capture the potential ultrasound features of breast cancer cases in the training set. A prediction model was constructed based on the captured ultrasound features. The model was adjusted and optimized through the validation set，and then matched with the case images in the test set. The confusion classification matrix graph and the curve graph for measuring the model performance were used to evaluate the model prediction performance and interpret the model，and the efficacy of this model in identifying breast cancer patients at risk of ALNM was analyzed.Results:There were statistically significant differences between the positive and negative groups in terms of the pathological maximum diameter of breast tumors，pathological T staging，the differentiation degree，the presence of distant metastasis，the maximum diameter measured by ultrasound，the quadrant of breast tumor occurrence，the Breast Imaging - Reporting and Data System（BI-RADS）classification of breast tumors，and the presence of abnormal ultrasound features of lymph node（all P<0.05）. The established deep learning model could automatically perform bounding box localization for the breast cancer of patients.The breast tumors in the positive group had potential ultrasound features that could be captured by the model compared with those in the negative group. The mean average precision（mAP）50 was 0.883，mAP 50-95 was 0.636，PR-AUC was 0.884 5，strict PR-AUC was 0.636 4，the sensitivity was 90.5%，and the specificity was 91.2%，and it had a good predictive efficacy. Conclusions:This prediction model based on the ultrasound characteristics of breast cancer through deep learning can effectively predict breast cancer（maximum diameter ≤ 2.5 cm）with the risk of ALNM，providing an effective basis for the clinical management of axillary lymph nodes in breast cancer patients.

Objective:To retrospectively evaluate the clinical efficacy and safety of total 3D laparo-scopic ileal ureters replacement for bilateral ureters combined with bladder augmentation in patients with post-radiotherapy long-segment bilateral ureteral strictures and contracted bladder.Methods:Clinical data of two patients(aged 72 and 54 years)with radiation-induced long-segment bilateral ureteral stric-tures and reduced bladder capacity,treated at the Sixth Affiliated Hospital of Jinan University from Octo-ber 2023 to June 2024,were analyzed.Both presented with bilateral flank pain,recurrent chills/fever,urinary frequency,and urgency.Preoperative ureteral stricture lengths were measured as follows:left 10.4 cm and right 8.7 cm in the first case;left 10.6 cm and right 11.7 cm in the second case.Bladder capacity assessed by nephrostomy-assisted antegrade urography was 90 mL and 130 mL respectively.Both underwent single-position,one-stage totally 3D laparoscopic bilateral ileal ureteral replacement and blad-der augmentation based on membrane anatomy principles,with regular postoperative follow-up.Results:Procedures were completed by the same experienced urologist.Operative times were 420 min and 355 min,with intraoperative blood loss of 50 mL(no transfusion required).Postoperative bowel function re-sumed at the end of 4.5 and 3 days.No major perioperative complications occurred.Ureteral stents were removed at 2 months postoperatively,with imaging showing improved hydronephrosis,unobstructed ure-teral drainage,symmetrical bladder morphology,and smooth walls.Postoperative bladder capacities were 230 mL and 250 mL.Follow-up durations were 10 and 8 months.Both patients experienced significant relief of flank pain and lower urinary tract symptoms.No complications(enteric fistula,urinary fistula,or metabolic acidosis)were observed.At the final follow-up,one patient had mildly elevated serum cre-atinine,while the other showed reduced levels compared with preoperative values;both remained stable.Conclusion:Membrane anatomy-based dissection facilitates safe mobilization of fibrotic ureters with mini-mal bleeding and collateral damage.Total intracorporeal 3D laparoscopic ileal ureters replacement for bi-lateral ureters combined with bladder augmentation effectively addresses long-segment ureteral obstruction and improves bladder capacity.This approach is technically safe and feasible,though further validation with larger clinical cohorts is warranted.

OBJECTIVES: To develop a heterogeneous graph prediction method based on the fusion of multi-layer semantics and topological information for addressing the challenges in drug-target interaction prediction, including insufficient modeling of high-order semantic dependencies, lack of adaptive fusion of semantic paths, and over-smoothing of node features. METHODS: A heterogeneous graph network with multiple types of entities such as drugs, proteins, side effects, and diseases was constructed, and graph embedding techniques were used to obtain low-dimensional feature representations. An adaptive metapath search module was introduced to automatically discover semantic path combinations for guiding the propagation of high-order semantic information. A semantic aggregation mechanism integrating multi-head attention was designed to automatically learn the importance of each semantic path based on contextual information and achieve differentiated aggregation and dynamic fusion among paths. A structure-aware gated graph convolutional module was then incorporated to regulate the feature propagation intensity for suppressing redundant information and redcuing over-smoothing. Finally, the potential interactions between drugs and targets were predicted through an inner product operation. RESULTS: Compared with existing drug-target interaction prediction methods, the proposed method achieved an average improvement of 3.4% and 2.4%, 3.0% and 3.8% in terms of the area under the receiver operating characteristic curve (AUC) and the area under the precision-recall curve (AUPRC) on public datasets, respectively. CONCLUSIONS The drug-target interaction prediction method developed in this study can effectively extract complex high-order semantic and topological information from heterogeneous biological networks, thereby improving the accuracy and stability of drug-target interaction prediction. This method provides technical support and theoretical foundation for precise drug target discovery and targeted treatment of complex diseases.
Semantics ; Humans ; Drug Interactions ; Neural Networks, Computer ; Algorithms

Objective:To establish a model based on the characteristics of breast cancer ultrasound images through deep learning methods to predict the risk of axillary lymph node metastasis（ALNM）in patients with breast cancer（maximum diameter ≤2.5 cm）before surgery.Methods:A total of 419 patients（3 433 breast tumor ultrasound images）with breast cancer（maximum diameter ≤2.5 cm）who underwent axillary lymph node dissection at Beijing Tiantan Hospital，Capital Medical University from January 2019 to December 2024 were retrospectively included. According to the pathological results of axillary lymph nodes，they were divided into 220 cases in the ALNM occurrence group（positive group）and 199 cases in the non-ALNM occurrence group（negative group）. The breast cancer ultrasound images of the two groups of cases were randomly classified into the training set（2 404 images），the validation set（687 images）and the test set（342 images）according to a ratio of 7∶2∶1. YOLOv8 was used as the basic model of You Only Look Once（YOLO）and optimized. The optimized model was applied to locate and capture the potential ultrasound features of breast cancer cases in the training set. A prediction model was constructed based on the captured ultrasound features. The model was adjusted and optimized through the validation set，and then matched with the case images in the test set. The confusion classification matrix graph and the curve graph for measuring the model performance were used to evaluate the model prediction performance and interpret the model，and the efficacy of this model in identifying breast cancer patients at risk of ALNM was analyzed.Results:There were statistically significant differences between the positive and negative groups in terms of the pathological maximum diameter of breast tumors，pathological T staging，the differentiation degree，the presence of distant metastasis，the maximum diameter measured by ultrasound，the quadrant of breast tumor occurrence，the Breast Imaging - Reporting and Data System（BI-RADS）classification of breast tumors，and the presence of abnormal ultrasound features of lymph node（all P<0.05）. The established deep learning model could automatically perform bounding box localization for the breast cancer of patients.The breast tumors in the positive group had potential ultrasound features that could be captured by the model compared with those in the negative group. The mean average precision（mAP）50 was 0.883，mAP 50-95 was 0.636，PR-AUC was 0.884 5，strict PR-AUC was 0.636 4，the sensitivity was 90.5%，and the specificity was 91.2%，and it had a good predictive efficacy. Conclusions:This prediction model based on the ultrasound characteristics of breast cancer through deep learning can effectively predict breast cancer（maximum diameter ≤ 2.5 cm）with the risk of ALNM，providing an effective basis for the clinical management of axillary lymph nodes in breast cancer patients.

Objective:To establish and validate a gallstones classification method based on deep learning.Methods:A total of 618 gallstones samples were collected from East Hospital Affiliated to Tongji University, and 1 023 high-definition cross-sectional gallstones profile images were captured to construct a cross-sectional gallstones profile image dataset. Based on the traditional eight-category gallstones classification method, a lightweight network model, MobileNet V3, was trained using deep learning and transfer learning methods. The classification performance of MobileNet was evaluated using a confusion matrix with metrics such as accuracy rate, precision rate, F1 score, and recall rate. The MobileNet V3 was improved and further validated using accuracy and loss values.Results:The accuracy rate (94.17%), precision rate (94.03%), F1 score (92.96%) and recall rate (92.99%) of the improved MobileNet V3 model were better than other networks. The improved MobileNet V3 model achieved the highest accuracy rate (94.17%) in gallstones profile classification and was validated by the test set. The confusion matrix showed a weighted average of accuracy rate (92.0%), precision rate (92.6%), and F1 score (92.2%) for each category of gallstones.Conclusions:Based on deep learning, a high-accuracy gallstones classification method is proposed, which provides a new idea for the intelligent identification of gallstones.

Objective:To design a novel electromagnetic ejection device for endoscopic suturing to achieve continuous deployment of suture nails.Methods:An electromagnetic ejection device and its accompanying suture nail structure were designed and a prototype was fabricated based on electromagnetic ejection principles. A finite element model of the electromagnetic ejection device was constructed to study the effects of armature-coil center distance and different driving voltages on suture nail ejection speed. An experimental platform for testing electromagnetic ejection velocity was constructed, and a high-speed camera was used to detect the ejection velocity. A platform for the suture embedding experiment was built to measure the effects of different voltages on the inserting speed of suture into the gastric wall tissue. A platform for a suture extraction force experiment was built to evaluate the extraction force of sutures embedded in tissues under different driving voltages.Results:A suture nail structure and electromagnetic ejection device were designed, and a prototype was fabricated. The ejection velocity increased and then decreased with the increase of the armature-coil center distance, and the maximum ejection velocity was 15.81 m/s at the center distance of 18 mm. At this distance, the voltage was linearly related to the ejection velocity, and the experimental values of the staple basically coincided with the simulated values. When the driving voltage was in the range of 150 to 180 V, the suture nails could successfully insert in the tissues, and the 180 V voltage group had a greater insertion depth. The extraction force of the suture nails at 120, 150, 180, and 210 V voltages were (0.49 ± 0.19), (1.14 ± 0.19), (1.23 ± 0.15), and (1.85 ± 0.31) N, respectively.Conclusions:A novel electromagnetic ejection device for endoscopic suturing is proposed that is capable of continuous firing of suture nails. This device provides a new long-distance driving method for intelligent, minimally invasive surgical instruments.

Objectives:To investigate the effect of the expression of low-density lipoprotein receptor associated protein (LDLR) on the vascular abnormalities in hepatocellular carcinoma (HCC) and its mechanisms.Methods:Based on the information of Oncomine Cancer GeneChip database, we analyzed the correlation between the expression level of LDLR and the expression level of carcinoembryonic antigen (CEA) and CD31 in hepatocellular carcinoma tissues. Lentiviral transfection of short hairpin RNA target genes was used to construct LDLR-knockdown MHCC-97H and HLE hepatocellular carcinoma cells. The differential genes and their expression level changes in LDLR-knockdown hepatocellular carcinoma cells were detected by transcriptome sequencing, real-time fluorescence quantitative polymerase chain reaction, and protein immunoblotting. The gene-related signaling pathways that involve LDLR were clarified by enrichment analysis. The effect of LDLR on CEA was assessed by the detection of CEA content in conditioned medium of hepatocellular carcinoma cells. Angiogenesis assay was used to detect the effect of LDLR on the angiogenic capacity of human umbilical vein endothelial cells, as well as the role of CEA in the regulation of angiogenesis by LDLR. Immunohistochemical staining was used to detect the expression levels of LDLR in 176 hepatocellular carcinoma tissues, and CEA and CD31 in 146 hepatocellular carcinoma tissues, and analyze the correlations between the expression levels of LDLR, CEA, and CD31 in the tissues, serum CEA, and alanine transaminase (ALT).Results:Oncomine database analysis showed that the expressions of LDLR and CEA in the tissues of hepatocellular carcinoma patients with portal vein metastasis were negatively correlated ( r=-0.64, P=0.001), whereas the expressions of CEA and CD31 in these tissues were positively correlated ( r=0.46, P=0.010). The transcriptome sequencing results showed that there were a total of 1 032 differentially expressed genes in the LDLR-knockdown group and the control group of MHCC-97H cells, of which 517 genes were up-regulated and 515 genes were down-regulated. The transcript expression level of CEACAM5 was significantly up-regulated in the cells of the LDLR-knockdown group. The Gene Ontology (GO) function enrichment analysis showed that the differential genes were most obviously enriched in the angiogenesis function. The Kyoto Encyclopedia of Genes and Genomes (KEGG) signaling pathway enrichment analysis showed that the relevant pathways involved mainly included the cellular adhesion patch, the extracellular matrix receptor interactions, and the interactions with the extracellular matrix receptors. The CEA content in the conditioned medium of the LDLR-knockdown group was 43.75±8.43, which was higher than that of the control group (1.15±0.14, P＜0.001). The results of angiogenesis experiments showed that at 5 h, the number of main junctions, the number of main segments, and the total area of the lattice formed by HUVEC cells cultured with the conditioned medium of MHCC-97H cells in the LDLR-knockdown group were 295.3±26.4, 552.5±63.8, and 2 239 781.0±13 8211.9 square pixels, which were higher than those of the control group (113.3±23.5, 194.8±36.5, and 660 621.0±280 328.3 square pixels, respectively, all P＜0.01).The number of vascular major junctions, the number of major segments, and the total area of the lattice formed by HUVEC cells cultured in conditioned medium with HLE cells in the LDLR-knockdown group were 245.3±42.4, 257.5±20.4, and 2 535 754.5±249 094.2 square pixels, respectively, which were all higher than those of the control group (113.3±23.5, 114.3±12.2, and 1 565 456.5±219 259.7 square pixels, respectively, all P＜0.01). In the conditioned medium for the control group of MHCC-97H cells,the number of main junctions, the number of main segments, and the total area of the lattice formed by the addition of CEA to cultured HUVEC cells were 178.9±12.0, 286.9±12.3, and 1 966 990.0±126 249.5 spixels, which were higher than those in the control group (119.7±22.1, 202.7±33.7, and 1 421 191.0±189 837.8 square pixels, respectively). The expression of LDLR in hepatocellular carcinoma tissues was not correlated with the expression of CEA, but was negatively correlated with the expression of CD31 ( r=-0.167, P=0.044), the level of serum CEA ( r=-0.061, P=0.032), and the level of serum ALT (r=-0.147, P=0.05). The expression of CEA in hepatocellular carcinoma tissues was positively correlated with the expression of CD31 ( r=0.192, P=0.020). The level of serum CEA was positively correlated with the level of serum ALT ( r=0.164, P=0.029). Conclusion:Knocking down LDLR can promote vascular abnormalities in HCC by releasing CEA.

Objectives:To investigate the effect of the expression of low-density lipoprotein receptor associated protein (LDLR) on the vascular abnormalities in hepatocellular carcinoma (HCC) and its mechanisms.Methods:Based on the information of Oncomine Cancer GeneChip database, we analyzed the correlation between the expression level of LDLR and the expression level of carcinoembryonic antigen (CEA) and CD31 in hepatocellular carcinoma tissues. Lentiviral transfection of short hairpin RNA target genes was used to construct LDLR-knockdown MHCC-97H and HLE hepatocellular carcinoma cells. The differential genes and their expression level changes in LDLR-knockdown hepatocellular carcinoma cells were detected by transcriptome sequencing, real-time fluorescence quantitative polymerase chain reaction, and protein immunoblotting. The gene-related signaling pathways that involve LDLR were clarified by enrichment analysis. The effect of LDLR on CEA was assessed by the detection of CEA content in conditioned medium of hepatocellular carcinoma cells. Angiogenesis assay was used to detect the effect of LDLR on the angiogenic capacity of human umbilical vein endothelial cells, as well as the role of CEA in the regulation of angiogenesis by LDLR. Immunohistochemical staining was used to detect the expression levels of LDLR in 176 hepatocellular carcinoma tissues, and CEA and CD31 in 146 hepatocellular carcinoma tissues, and analyze the correlations between the expression levels of LDLR, CEA, and CD31 in the tissues, serum CEA, and alanine transaminase (ALT).Results:Oncomine database analysis showed that the expressions of LDLR and CEA in the tissues of hepatocellular carcinoma patients with portal vein metastasis were negatively correlated ( r=-0.64, P=0.001), whereas the expressions of CEA and CD31 in these tissues were positively correlated ( r=0.46, P=0.010). The transcriptome sequencing results showed that there were a total of 1 032 differentially expressed genes in the LDLR-knockdown group and the control group of MHCC-97H cells, of which 517 genes were up-regulated and 515 genes were down-regulated. The transcript expression level of CEACAM5 was significantly up-regulated in the cells of the LDLR-knockdown group. The Gene Ontology (GO) function enrichment analysis showed that the differential genes were most obviously enriched in the angiogenesis function. The Kyoto Encyclopedia of Genes and Genomes (KEGG) signaling pathway enrichment analysis showed that the relevant pathways involved mainly included the cellular adhesion patch, the extracellular matrix receptor interactions, and the interactions with the extracellular matrix receptors. The CEA content in the conditioned medium of the LDLR-knockdown group was 43.75±8.43, which was higher than that of the control group (1.15±0.14, P＜0.001). The results of angiogenesis experiments showed that at 5 h, the number of main junctions, the number of main segments, and the total area of the lattice formed by HUVEC cells cultured with the conditioned medium of MHCC-97H cells in the LDLR-knockdown group were 295.3±26.4, 552.5±63.8, and 2 239 781.0±13 8211.9 square pixels, which were higher than those of the control group (113.3±23.5, 194.8±36.5, and 660 621.0±280 328.3 square pixels, respectively, all P＜0.01).The number of vascular major junctions, the number of major segments, and the total area of the lattice formed by HUVEC cells cultured in conditioned medium with HLE cells in the LDLR-knockdown group were 245.3±42.4, 257.5±20.4, and 2 535 754.5±249 094.2 square pixels, respectively, which were all higher than those of the control group (113.3±23.5, 114.3±12.2, and 1 565 456.5±219 259.7 square pixels, respectively, all P＜0.01). In the conditioned medium for the control group of MHCC-97H cells,the number of main junctions, the number of main segments, and the total area of the lattice formed by the addition of CEA to cultured HUVEC cells were 178.9±12.0, 286.9±12.3, and 1 966 990.0±126 249.5 spixels, which were higher than those in the control group (119.7±22.1, 202.7±33.7, and 1 421 191.0±189 837.8 square pixels, respectively). The expression of LDLR in hepatocellular carcinoma tissues was not correlated with the expression of CEA, but was negatively correlated with the expression of CD31 ( r=-0.167, P=0.044), the level of serum CEA ( r=-0.061, P=0.032), and the level of serum ALT (r=-0.147, P=0.05). The expression of CEA in hepatocellular carcinoma tissues was positively correlated with the expression of CD31 ( r=0.192, P=0.020). The level of serum CEA was positively correlated with the level of serum ALT ( r=0.164, P=0.029). Conclusion:Knocking down LDLR can promote vascular abnormalities in HCC by releasing CEA.

Objective:The microstructure, tensile strength, and bursting strength of different brands of hernia meshes were compared and analyzed through experiments to evaluate the performance of different meshes.Methods:The balance and microscope were used to test the weight and microstructure of 15 common meshes and the tensile testing machine and burst testing machine were used to test the tensile and bursting properties of the mesh, and the mechanical properties of the mesh were analyzed.Results:The woven structures of the mesh are diamond, polygon and circle. The average weight of inguinal meshes is 0.08 mg/mm 2, and the average weight of abdominal wall hernia meshes is 0.18 mg/mm 2. The wire diameters of G3 - G6 meshes are larger, while the mesh opening ratio of G12 is lower. In the tensile performance test, it is known that G15 has the highest tensile strength, G12 and G14 have lower tensile strengths in lightweight meshes, and G1, G2, and G7 have lower tensile strengths in lightweight meshes. In the burst performance test, it is known that G3, G9, and G15 have the highest burst strength, while G12, G13, and G14 have the lowest burst strength in lightweight meshes. G1, G2, and G4 have the lowest burst strength in lightweight meshes. Conclusions:The mesh with a polygonal mesh and a large mesh opening ratio has better mechanical properties. The results of this study provide experimental evidence for optimizing hernia meshes, which is expected to provide better support for related research and applications.