1.Research on objective evaluation of pilot ground simulation ejection training based on pressure detection
Yanpeng ZHAO ; Weiping BU ; Shu ZHANG ; Fei SHI ; Heqing LIU ; Yan JIANG ; Yining ZHANG ; Xinpo SONG ; Lei WANG
Space Medicine & Medical Engineering 2025;36(3):236-240
Objective In response to the practical issues of pilot ground simulation ejection training relying on subjective evaluation,which requires high professionalism and lacks objectivity and systematicity,an objective evaluation system and method for ejection training based on pressure detection has been designed.Methods According to the rules of ejection training about posture,maneuver and time,and utilizing technologies such as pressure detection,infrared detection,and time-triggered detection,this paper designed separate modules for ejection posture detection,ejection timing detection,and comprehensive analysis and evaluation.These modules are closely integrated with the ejection training process,achieving objective evaluation of the ejection training.Results This system and method can provide a scientific and real-time objective evaluation of the posture,movements,and time in ejection training without affecting the normal organization and process.Conclusion The objective evaluation system can provide a new approach that is highly objective and easy to operate for comprehensively evaluating and enhancing the effectiveness of ejection training,as well as for scientifically conducting ejection training and assessment.
2.Identify drug-drug interactions via deep learning: A real world study.
Jingyang LI ; Yanpeng ZHAO ; Zhenting WANG ; Chunyue LEI ; Lianlian WU ; Yixin ZHANG ; Song HE ; Xiaochen BO ; Jian XIAO
Journal of Pharmaceutical Analysis 2025;15(6):101194-101194
Identifying drug-drug interactions (DDIs) is essential to prevent adverse effects from polypharmacy. Although deep learning has advanced DDI identification, the gap between powerful models and their lack of clinical application and evaluation has hindered clinical benefits. Here, we developed a Multi-Dimensional Feature Fusion model named MDFF, which integrates one-dimensional simplified molecular input line entry system sequence features, two-dimensional molecular graph features, and three-dimensional geometric features to enhance drug representations for predicting DDIs. MDFF was trained and validated on two DDI datasets, evaluated across three distinct scenarios, and compared with advanced DDI prediction models using accuracy, precision, recall, area under the curve, and F1 score metrics. MDFF achieved state-of-the-art performance across all metrics. Ablation experiments showed that integrating multi-dimensional drug features yielded the best results. More importantly, we obtained adverse drug reaction reports uploaded by Xiangya Hospital of Central South University from 2021 to 2023 and used MDFF to identify potential adverse DDIs. Among 12 real-world adverse drug reaction reports, the predictions of 9 reports were supported by relevant evidence. Additionally, MDFF demonstrated the ability to explain adverse DDI mechanisms, providing insights into the mechanisms behind one specific report and highlighting its potential to assist practitioners in improving medical practice.
3.Temporal distribution characteristics of other infectious diarrhea in Shenzhen, 2011-2023
Lixia SONG ; Wenhai LU ; Zhen ZHANG ; Yanpeng CHENG ; Huawei XIONG ; Yan LU ; Qiuying LYU ; Zhigao CHEN
Chinese Journal of Epidemiology 2025;46(9):1610-1616
Objective:To analyze the temporal distribution of other infectious diarrhea (OID) in Shenzhen and provide evidence for the prevention and control of OID.Methods:The incidence data of OID in Shenzhen from 2011 to 2023 were collected. The seasonal and trend decomposition using loess (STL), seasonal index method, concentration degree and circular distribution method were used to analyze the incidence trend and temporal distribution of OID.Results:A total of 477 611 cases of OID were reported in Shenzhen from 2011 to 2023, with an average annual incidence rate of 260.19/100 000 showing a fluctuating upward trend. The seasonal index method indicated that October-January was period with high incidence of OID in Shenzhen and the seasonal intensity began to decrease in 2020. STL revealed an obvious incidence peak in winter. The concentration method showed that OID had a certain seasonality before 2018 except 2016, but the seasonality was not obvious after 2018. The circular distribution results showed that r was 0.05, mean angle ā was 1.92° and angular standard deviation s was 141.93° ( Z=1 033.37, P<0.001), with the peak on January 1 st and the high incidence period from August 11 th to May 25 th. Conclusions:OID had a certain degree of seasonality in Shenzhen, with an obvious incidence peak in winter. Since the seasonal intensity of OID decreased after 2018, the surveillance, early warning and risk assessment of OID should be continued, and prevention and control measures should be adjusted timely according to the change in the characteristics of the epidemic.
4.Identify drug-drug interactions via deep learning:A real world study
Jingyang LI ; Yanpeng ZHAO ; Zhenting WANG ; Chunyue LEI ; Lianlian WU ; Yixin ZHANG ; Song HE ; Xiaochen BO ; Jian XIAO
Journal of Pharmaceutical Analysis 2025;15(6):1249-1263
Identifying drug-drug interactions(DDIs)is essential to prevent adverse effects from polypharmacy.Although deep learning has advanced DDI identification,the gap between powerful models and their lack of clinical application and evaluation has hindered clinical benefits.Here,we developed a Multi-Dimensional Feature Fusion model named MDFF,which integrates one-dimensional simplified molec-ular input line entry system sequence features,two-dimensional molecular graph features,and three-dimensional geometric features to enhance drug representations for predicting DDIs.MDFF was trained and validated on two DDI datasets,evaluated across three distinct scenarios,and compared with advanced DDI prediction models using accuracy,precision,recall,area under the curve,and F1 score metrics.MDFF achieved state-of-the-art performance across all metrics.Ablation experiments showed that integrating multi-dimensional drug features yielded the best results.More importantly,we obtained adverse drug reaction reports uploaded by Xiangya Hospital of Central South University from 2021 to 2023 and used MDFF to identify potential adverse DDIs.Among 12 real-world adverse drug reaction reports,the predictions of 9 reports were supported by relevant evidence.Additionally,MDFF demon-strated the ability to explain adverse DDI mechanisms,providing insights into the mechanisms behind one specific report and highlighting its potential to assist practitioners in improving medical practice.
5.Temporal distribution characteristics of other infectious diarrhea in Shenzhen, 2011-2023
Lixia SONG ; Wenhai LU ; Zhen ZHANG ; Yanpeng CHENG ; Huawei XIONG ; Yan LU ; Qiuying LYU ; Zhigao CHEN
Chinese Journal of Epidemiology 2025;46(9):1610-1616
Objective:To analyze the temporal distribution of other infectious diarrhea (OID) in Shenzhen and provide evidence for the prevention and control of OID.Methods:The incidence data of OID in Shenzhen from 2011 to 2023 were collected. The seasonal and trend decomposition using loess (STL), seasonal index method, concentration degree and circular distribution method were used to analyze the incidence trend and temporal distribution of OID.Results:A total of 477 611 cases of OID were reported in Shenzhen from 2011 to 2023, with an average annual incidence rate of 260.19/100 000 showing a fluctuating upward trend. The seasonal index method indicated that October-January was period with high incidence of OID in Shenzhen and the seasonal intensity began to decrease in 2020. STL revealed an obvious incidence peak in winter. The concentration method showed that OID had a certain seasonality before 2018 except 2016, but the seasonality was not obvious after 2018. The circular distribution results showed that r was 0.05, mean angle ā was 1.92° and angular standard deviation s was 141.93° ( Z=1 033.37, P<0.001), with the peak on January 1 st and the high incidence period from August 11 th to May 25 th. Conclusions:OID had a certain degree of seasonality in Shenzhen, with an obvious incidence peak in winter. Since the seasonal intensity of OID decreased after 2018, the surveillance, early warning and risk assessment of OID should be continued, and prevention and control measures should be adjusted timely according to the change in the characteristics of the epidemic.
6.Prenatal ultrasound diagnosis of fetal congenital pyriform sinus fistula: a case report
Yanpeng SONG ; Ming LI ; Yuhang ZHANG
Chinese Journal of Perinatal Medicine 2024;27(3):238-240
This article reported a case of fetal congenital pyriform sinus fistula (CPSF) diagnosed by prenatal ultrasound. An oval cystic mass in the left side of the fetal neck was detected during routine prenatal ultrasonography at 17 +2 weeks of gestation, which was also found in the systemic ultrasound screening at 23 weeks. Besides, a narrow fistula between the cyst and pharynx was observed in the coronal view of the neck, raising the suspicion of CPSF. On day 21 after birth, the cystic mass bulged out of the neck and the neonate started choking and coughing during feeding alongside wheezing respiration. Ultrasound examination showed a strong gas echo in the cystic mass, which was also confirmed by CT scan on day 25. On day 26, cystectomy, ligation of high fistula, and fistulectomy were performed, and the diagnosis of CPSF was confirmed. The baby recovered after the operation and was healthy during follow-up till three months.
7.Unilateral tubal twin pregnancy: report of two cases and literature review
Mingyuan YUAN ; Yanpeng SONG ; Chengjuan JIN
Chinese Journal of Reproduction and Contraception 2024;44(12):1285-1291
Unilateral tubal twin pregnancy is a relatively rare ectopic pregnancy. This paper reported the diagnosis and treatment of two cases of unilateral tubal twin pregnancy admitted to Jiaozhou Central Hospital of Qingdao in 2015 and 2023, respectively, and the relevant literatures were reviewed. This paper summarized the risk factors, clinical case characteristics, clinical diagnosis and treatment methods of unilateral tubal twin pregnancy, in order to provide useful reference for early recognition, accurate diagnosis and timely treatment of unilateral tubal twin pregnancy, and reduce its incidence rate and mortality.
8.Anatomical study of the ideal lag screw trajectories in the L 5 spondylolysis
Xingguo TAN ; Tao ZHANG ; Xiaohong TIAN ; Mingjia SONG ; Yizhe WANG ; Long CHEN ; Dashuai HUANG ; Yanpeng LU ; Songkai LI
Chinese Journal of Orthopaedics 2024;44(24):1594-1601
Objective:To explore the anatomical parameters of the ideal trajectory for pedicle screw fixation through the lamina in the treatment of L 5 spondylolysis. Methods:CT data from 40 male patients with bilateral L 5 spondylolysis (age, 24.95±4.01 years; range, 20-36 years), treated at the 940th Hospital of PLA Joint Logistics Support Force between January 2021 and June 2024, were analyzed. Three-dimensional vertebral models were reconstructed using this data. Measurements included the lumbosacral angle, the thickness at the midpoint of the superior and inferior lamina edges, mid-lamina thickness, the distance from the lateral edge of the lamina to the spinous process midline, the thickness at the defect of the pars interarticularis, and the vertical diameter of the defect. The screws were inserted from the inferior edge of the lamina, passing through the pars interarticularis defect, and exiting at the superior edge of the pedicle. In the vertical direction of the lamina, the inferior and superior edges of the lamina were divided into three zones, named A, B, C (for the inferior lamina edge) and 1, 2, 3 (for the superior pedicle edge). Seven trajectories (A2, A3, B1, B2, B3, C2, and C3) were designed by combining these zones. Screws with diameters of 5.0, 4.5, 4.0, and 3.5 mm were sequentially inserted along each trajectory. Screw trajectories with an insertion success rate ≥95% were selected and evaluated for feasibility. Parameters such as screw length, medial inclination angle, caudal inclination angle, and entry point position were measured. The ideal trajectory and screw dimensions were determined by considering anatomical features, screw characteristics, and insertion safety. Results:The measurement results from the 3D model showed that the lumbosacral angle was 36.22°±5.23°, and the midpoint thickness of the superior lamina edge was 4.14±0.66 mm (left) and 4.18±0.65 mm (right), the mid-lamina thickness was 6.73±0.72 mm (left) and 6.72±0.70 mm (right), the midpoint thickness of the inferior lamina edge was 6.50±0.56 mm (left) and 6.50±0.66 mm (right), the distance from the lateral edge of the lamina to the spinous process midline was 25.95±2.86 mm (left) and 26.39±3.10 mm (right), the thickness at the pars defect was 9.67±0.57 mm (left) and 9.67±0.51 mm (right), and the vertical diameter of the pars defect was 18.76±2.16 mm (left) and 19.26±2.03 mm (right). No statistically significant differences were found between the left and right sides for these parameters ( P>0.05). The trajectories considered feasible and with an insertion success rate ≥95% were B2, B3, C2, and C3. Safe screw diameters were B2 (4.5 mm), B3 (4.0 mm), C2 (4.0 mm), and C3 (3.5 mm). Corresponding screw lengths were B2 (38.28±2.34 mm), B3 (37.03±2.99 mm), C2 (38.37±2.42 mm), and C3 (36.88±2.87 mm). The caudal inclination angles were B2 (52.73°±5.29°), B3 (55.06°± 4.46°), C2 (49.09°±3.92°), and C3 (50.18°±4.36°). The medial inclination angles were B2 (21.21°±3.01°), B3 (5.11°±1.58°), C2 (22.55°±2.46°), and C3 (12.59°±1.80°). The distances from the entry point to the spinous process midline were B2 (13.23±1.68 mm), B3 (13.15±1.46 mm), C2 (11.12±0.64 mm), and C3 (11.09±0.65 mm). The distances from the entry point to the root of the spinous process were B2 (8.23±1.46 mm), B3 (8.21±1.31 mm), C2 (6.65 ±0.76 mm), and C3 (6.67±0.72 mm). Differences in screw length, caudal inclination angle, medial inclination angle, and entry point position across trajectories were statistically significant ( P<0.05). Conclusion:The ideal screw trajectory for L 5 spondylolysis involves insertion through the midpoint of the entry zone, passing through the pars defect, and exiting at the midpoint of the superior edge of the pedicle. The optimal entry point is located on the inferior edge of the lamina, 8.23±1.46 mm from the root of the spinous process and 13.23±1.68 mm from the spinous process midline. The screw should be placed at a caudal inclination angle of 52.73°±5.29° and a medial inclination angle of 21.21°±3.01°. The recommended screw length is 38.28±2.34 mm, with a diameter of 4.5 mm (range, 4.5-5.0 mm).
9.Unilateral tubal twin pregnancy: report of two cases and literature review
Mingyuan YUAN ; Yanpeng SONG ; Chengjuan JIN
Chinese Journal of Reproduction and Contraception 2024;44(12):1285-1291
Unilateral tubal twin pregnancy is a relatively rare ectopic pregnancy. This paper reported the diagnosis and treatment of two cases of unilateral tubal twin pregnancy admitted to Jiaozhou Central Hospital of Qingdao in 2015 and 2023, respectively, and the relevant literatures were reviewed. This paper summarized the risk factors, clinical case characteristics, clinical diagnosis and treatment methods of unilateral tubal twin pregnancy, in order to provide useful reference for early recognition, accurate diagnosis and timely treatment of unilateral tubal twin pregnancy, and reduce its incidence rate and mortality.
10.Anatomical study of the ideal lag screw trajectories in the L 5 spondylolysis
Xingguo TAN ; Tao ZHANG ; Xiaohong TIAN ; Mingjia SONG ; Yizhe WANG ; Long CHEN ; Dashuai HUANG ; Yanpeng LU ; Songkai LI
Chinese Journal of Orthopaedics 2024;44(24):1594-1601
Objective:To explore the anatomical parameters of the ideal trajectory for pedicle screw fixation through the lamina in the treatment of L 5 spondylolysis. Methods:CT data from 40 male patients with bilateral L 5 spondylolysis (age, 24.95±4.01 years; range, 20-36 years), treated at the 940th Hospital of PLA Joint Logistics Support Force between January 2021 and June 2024, were analyzed. Three-dimensional vertebral models were reconstructed using this data. Measurements included the lumbosacral angle, the thickness at the midpoint of the superior and inferior lamina edges, mid-lamina thickness, the distance from the lateral edge of the lamina to the spinous process midline, the thickness at the defect of the pars interarticularis, and the vertical diameter of the defect. The screws were inserted from the inferior edge of the lamina, passing through the pars interarticularis defect, and exiting at the superior edge of the pedicle. In the vertical direction of the lamina, the inferior and superior edges of the lamina were divided into three zones, named A, B, C (for the inferior lamina edge) and 1, 2, 3 (for the superior pedicle edge). Seven trajectories (A2, A3, B1, B2, B3, C2, and C3) were designed by combining these zones. Screws with diameters of 5.0, 4.5, 4.0, and 3.5 mm were sequentially inserted along each trajectory. Screw trajectories with an insertion success rate ≥95% were selected and evaluated for feasibility. Parameters such as screw length, medial inclination angle, caudal inclination angle, and entry point position were measured. The ideal trajectory and screw dimensions were determined by considering anatomical features, screw characteristics, and insertion safety. Results:The measurement results from the 3D model showed that the lumbosacral angle was 36.22°±5.23°, and the midpoint thickness of the superior lamina edge was 4.14±0.66 mm (left) and 4.18±0.65 mm (right), the mid-lamina thickness was 6.73±0.72 mm (left) and 6.72±0.70 mm (right), the midpoint thickness of the inferior lamina edge was 6.50±0.56 mm (left) and 6.50±0.66 mm (right), the distance from the lateral edge of the lamina to the spinous process midline was 25.95±2.86 mm (left) and 26.39±3.10 mm (right), the thickness at the pars defect was 9.67±0.57 mm (left) and 9.67±0.51 mm (right), and the vertical diameter of the pars defect was 18.76±2.16 mm (left) and 19.26±2.03 mm (right). No statistically significant differences were found between the left and right sides for these parameters ( P>0.05). The trajectories considered feasible and with an insertion success rate ≥95% were B2, B3, C2, and C3. Safe screw diameters were B2 (4.5 mm), B3 (4.0 mm), C2 (4.0 mm), and C3 (3.5 mm). Corresponding screw lengths were B2 (38.28±2.34 mm), B3 (37.03±2.99 mm), C2 (38.37±2.42 mm), and C3 (36.88±2.87 mm). The caudal inclination angles were B2 (52.73°±5.29°), B3 (55.06°± 4.46°), C2 (49.09°±3.92°), and C3 (50.18°±4.36°). The medial inclination angles were B2 (21.21°±3.01°), B3 (5.11°±1.58°), C2 (22.55°±2.46°), and C3 (12.59°±1.80°). The distances from the entry point to the spinous process midline were B2 (13.23±1.68 mm), B3 (13.15±1.46 mm), C2 (11.12±0.64 mm), and C3 (11.09±0.65 mm). The distances from the entry point to the root of the spinous process were B2 (8.23±1.46 mm), B3 (8.21±1.31 mm), C2 (6.65 ±0.76 mm), and C3 (6.67±0.72 mm). Differences in screw length, caudal inclination angle, medial inclination angle, and entry point position across trajectories were statistically significant ( P<0.05). Conclusion:The ideal screw trajectory for L 5 spondylolysis involves insertion through the midpoint of the entry zone, passing through the pars defect, and exiting at the midpoint of the superior edge of the pedicle. The optimal entry point is located on the inferior edge of the lamina, 8.23±1.46 mm from the root of the spinous process and 13.23±1.68 mm from the spinous process midline. The screw should be placed at a caudal inclination angle of 52.73°±5.29° and a medial inclination angle of 21.21°±3.01°. The recommended screw length is 38.28±2.34 mm, with a diameter of 4.5 mm (range, 4.5-5.0 mm).

Result Analysis
Print
Save
E-mail