Objective By applying the long short-term memory(LSTM)neural network model and using lower body landmark coordinates obtained from a markerless motion capture system as inputs,to estimate ground reaction force(GRF)curves during running.Methods The video images and GRF data of 59 amateur runners during running were collected by the markerless motion capture system and three-dimensional(3D)force plates.The LSTM model was established,and the 3D coordinates of 11 lower body landmarks,obtained via the Theia3D markerless system,were used as inputs to estimate the 3D GRF curves during the stance of running.The estimation performance was evaluated using correlation coefficients r,root mean square error(RMSE),and normalized root mean square error(nRMSE)by comparing LSTM model estimation and force plate measurement.Statistical parametric mapping was used to analyze differences in GRF curves estimated by the LSTM model and measured by the force plate,while paired t-tests were used to assess differences in GRF characteristics between model estimation and actual measurement.Results A strong correlation(r＞0.85,P＜0.001)and lower error(RMSE＜0.3 body weight,nRMSE＜15％)was found between the LSTM model estimation and actual measurements.No significant difference was found in GRF curve intervals between LSTM model estimation and actual measurements.There was no significant difference in GRF characteristics between LSTM model estimation and actual measurements(P＞0.05).Conclusions Based on the LSTM model,the 3D GRF curves can be effectively estimated by lower body landmark coordinates obtained from the makerless motion capture system,thereby acquiring the highly accurate GRF characteristics.The LSTM model developed in this study can be used to monitor injury risks during running in outdoor environments.

Objective By applying the long short-term memory(LSTM)neural network model and using lower body landmark coordinates obtained from a markerless motion capture system as inputs,to estimate ground reaction force(GRF)curves during running.Methods The video images and GRF data of 59 amateur runners during running were collected by the markerless motion capture system and three-dimensional(3D)force plates.The LSTM model was established,and the 3D coordinates of 11 lower body landmarks,obtained via the Theia3D markerless system,were used as inputs to estimate the 3D GRF curves during the stance of running.The estimation performance was evaluated using correlation coefficients r,root mean square error(RMSE),and normalized root mean square error(nRMSE)by comparing LSTM model estimation and force plate measurement.Statistical parametric mapping was used to analyze differences in GRF curves estimated by the LSTM model and measured by the force plate,while paired t-tests were used to assess differences in GRF characteristics between model estimation and actual measurement.Results A strong correlation(r＞0.85,P＜0.001)and lower error(RMSE＜0.3 body weight,nRMSE＜15％)was found between the LSTM model estimation and actual measurements.No significant difference was found in GRF curve intervals between LSTM model estimation and actual measurements.There was no significant difference in GRF characteristics between LSTM model estimation and actual measurements(P＞0.05).Conclusions Based on the LSTM model,the 3D GRF curves can be effectively estimated by lower body landmark coordinates obtained from the makerless motion capture system,thereby acquiring the highly accurate GRF characteristics.The LSTM model developed in this study can be used to monitor injury risks during running in outdoor environments.

BACKGROUND:Overactive osteoclasts disrupt bone homeostasis and play a bad role in the pathological mechanisms of related skeletal diseases,such as osteoporosis,fragility fractures,and osteoarthritis.Studies have confirmed that ellagic acid and ellagtannin have the potential to inhibit osteoclast differentiation.As their natural metabolites,urolithin A has antioxidant,anti-inflammatory,anti-proliferative and anti-cancer effects,but its effect on osteoclast differentiation and its underlying molecular mechanisms remain unclear. OBJECTIVE:To explore the effect of urolithin A on osteoclast differentiation induced by receptor activator for nuclear factor-κB ligand and its mechanism. METHODS:Mouse mononuclear macrophage leukemia cells(RAW264.7)that grew stably were cultured in vitro.Toxicity of urolithin A(0,0.1,0.5,1.5,2.5 μmol/L)to RAW264.7 cells were detected by cytotoxic MTS assay to screen out the safe concentration.Different concentrations of urolithin A were used again to intervene with receptor activator for nuclear factor-κB ligand-induced differentiation of RAW264.7 cells in vitro.Then,tartrate-resistant acid phosphatase staining and F-actin ring and nucleus staining were performed to observe its effect on the formation and function of osteoclasts.Finally,the expressions of urolithin A on upstream and downstream genes and proteins in the MAPK signaling pathway were observed by western blot and RT-qPCR assays. RESULTS AND CONCLUSION:Urolithin A inhibited osteoclast differentiation and F-actin ring formation in a concentration-dependent manner and 2.5 μmol/L had the strongest inhibitory effect.Urolithin A inhibited the mRNA expression of Nfatc1,Ctsk,Mmp9 and Atp6v0d2 and the protein synthesis of Nfatc1 and Ctsk,related to osteoclast formation and bone resorption.Urolithin A inhibited the activity of osteoclasts by downregulating the phosphorylation of p38 protein to inhibit the mitogen-activated protein kinase signaling pathway.

BACKGROUND:Studies have shown that imbalance of bone metabolism during glucocorticoid-induced osteonecrosis of the femoral head necrosis is closely related to oxidative stress. OBJECTIVE:To investigate the pathological mechanism by which oxidative stress-induced ferroptosis promote apoptosis in osteoblasts involved in steroid-induced osteonecrosis of the femoral head. METHODS:General data and serum specimens were collected from 47 patients with steroid-induced osteonecrosis of the femoral head.In addition,six femoral head specimens were collected from these patients.According to the Association Research Circulation Osseous(ARCO)staging system,serum specimens were grouped into ARCO Ⅱ,Ⅲ,and IV,while femoral head specimens were classified into ARCO Ⅲ and IV.Serum levels of malondialdehyde and superoxide dismutase 1 were measured.The protein expression of superoxide dismutase 1,glutathione peroxidase 4,Bcl-2 in the femoral head was detected and verified by Data independent acquisition(DIA)for quantitative sequencing,western blot and alkaline phosphate detection. RESULTS AND CONCLUSION:The ARCO stage of patients with steroid-induced osteonecrosis of the femoral head was independent of age,sex and necrotic side.The serum levels of malondialdehyde and superoxide dismutase 1 were higher in patients with ARCO stage Ⅲ compared with those with ARCO stage Ⅱ and IV.The results of DIA protein quantification showed that the function of differential proteins was mainly related to redox.The levels of superoxide dismutase 1,glutathione peroxidase 4,and Bcl-2 in the necrotic region were lower than in the normal region,as well as lower in ARCO stage IV than in ARCO stage Ⅲ.Western blot verified the results of DIA protein quantification.The alkaline phosphatase activity was lower in the necrotic region than in the normal region,as well as lower in ARCO stage IV than in ARCO stage Ⅲ.In the necrotic and sclerotic regions,the function of differential proteins was also related to redox,and superoxide dismutase 1,glutathione peroxidase 4,Bcl-2 protein expression and alkaline phosphatase activity were lower in the necrotic area than in the sclerotic region,as well as lower in ARCO stage IV than in ARCO stage Ⅲ.To conclude,glucocorticoids can influence the progression of steroid-induced osteonecrosis of the femoral head by upregulating oxidative stress levels,inducing osteoblast ferroptosis,and inhibiting osteogenic function.

Objective:To develop a deep transfer learning method for the differential diagnosis of osteonecrosis of the femoral head (ONFH) with other common hip diseases using anteroposterior hip radiographs.Methods:Patients suffering from ONFH, DDH, and other hip diseases including primary hip osteoarthritis, non-infectious inflammatory hip disease, and femoral neck fracture treated in the First Affiliated Hospital of Guangzhou University of Chinese Medicine from January 2018 to December 2020 were enrolled in the study. A clinical data set containing anteroposterior hip radiographs of the eligible patients was created. Data augmentation by rotating and flipping images was performed to enlarge the data set, then the data set was divided equally into a training data set and a testing data set. The ResNet-152, a deep neural network model, was used in the study, but the original Batch Normalization was replaced with Transferable Normalization to construct a novel deep transfer learning model. The model was trained to distinguish ONFH and DDH from other common hip diseases using anteroposterior hip radiographs on the training data set and its classification performance was evaluated on the testing data set.Results:The clinical data set was comprised of anteroposterior hip radiographs of 1024 hips, including 542 with ONFH, 296 with DDH, and 186 with other common hip diseases (56 hips with primary osteoarthritis, 85 hips with non-infectious inflammatory osteoarthritis, 45 hips with femoral neck fracture). After data augmentation, the size of the data set multiplied to 6144. The model was trained 100 050 times in each task. Accuracy was used as the representative parameter to evaluate the performance of the model. In the binary classification task to identify ONFH, the best accuracy was 95.80%. As for the multi-classification task for classification of ONFH and DDH from other hip diseases, the best accuracy was 91.40%. The plateau of the model was observed in each task after 50 000 times of training. The mean accuracy in plateaus was 95.35% (95% CI: 95.33%, 95.37%), and 90.85% (95% CI: 90.82%, 90.87%), respectively. Conclusion:The present study proves the encouraging performance of a deep transfer learning method for the first-visit classification of ONFH, DDH, and other hip diseases using the convenient and economical anteroposterior hip radiographs.

ObjectiveTo analyze the reliability and validity of Chinese version of Dysarthria Impact Profile (DIP) in assessment of the psychosocial impact of dysarthria in Parkinson's disease (PD). MethodsFrom May, 2021 to March, 2022, 43 PD patients from Department of Rehabilitation Medicine, the First Affiliated Hospital of Sun Yat-sen University were selected, and 43 age matched healthy controls were enrolled. The process of translation and adaptation was used to develop the Chinese version of DIP, and two groups were evaluated. The internal consistency reliability and intra-rater reliability were analyzed as well as the correlation between each item and its subscale, DIP scores to the Voice Handicap Index (VHI) and 36-item Short Form Health Survey (SF-36). DIP scores of two groups were compared. ResultsThe Cronbach's α was 0.732 to 0.942. The intra-rater correlation coefficient of subsection four was the highest (r = 0.670, P < 0.001). The correlation coefficients were 0.315 to 0.871, which were correlated (P < 0.05), except items 1, 6, 11 of subsection three and item 11 of subsection four. The correlation coefficient between DIP and VHI was -0.821 (P < 0.01), and it was 0.684 (P < 0.01) between DIP and SF-36. DIP scores were significant different between PD patients and the control group (P < 0.01). ConclusionThe Chinese version of DIP shows good reliability and validity, and can be used as a tool to measure the psychosocial impact of dysarthria in PD patients.

Objective To study the effect of gender and maneuvers on anterior cruciate ligament (ACL) injury risk factors for volleyball players. Methods Sports biomechanics data of volleyball players during stop-jump, drop landing and sidestep cutting were collected. The ACL injury rate and biomechanical parameters of simulated injured jumps were obtained with Monte Carlo simulation. The influence of gender and maneuvers on ACL injury risk factors was validated by 2×3 mixed designed two-way ANOVA. Results Sidestep cutting was the highest risk maneuver of ACL injury for both genders (P＜0.001). Compared with male players, female players had a greater risk of ACL injury during sidestep cutting and stop-jump (P＜0.001), while male players were more prone to have ACL injury than female players during drop landing (P＜0.001). The risk factors of ACL injury obtained by simulation were significantly influenced by gender and maneuvers (P＜0.001). Conclusions Male players were more likely to increase ACL load due to smaller knee flexion, forward leg tilt and heel landing than female players during sidestep cutting, while female players owned larger ground reaction force (GRF) and knee extension moment. Smaller knee flexion angle during stop jump was the major risk factor for both genders, however more characteristics contributed to the males. Female players with large GRF, knee valgus and extension moment, and heel-landing were likely to have ACL injury, while the small knee flexion angle was the key risk factor for male players. The results can provide evidences for evaluation of volleyball players’ ACL injury risk, individualized injury prevention protocols, and clinical treatment and rehabilitation directions.

Objective To compare rotational techniques used by Chinese and world elite men’s shot put athletes, so as to provide scientific references for Chinese male shot putters to improve their sports performance and results in international competitions. Methods Three-dimensional (3D) kinematics data from Chinese male shot putters in actual competitions were obtained, and kinematic parameters of Chinese and world elite men’s shot put athletes were calculated and compared. Namely, shot velocities and hip-shoulder separation angles at five critical instants of rotational techniques: right foot off, left foot off, right foot touchdown, left foot touchdown and release, as well as phase durations and shot travel distances during four critical phases: first single support, air-born flight, second single support and final delivery. ResultsCompared with world elite athletes, Chinese male shot putters using rotational techniques had significantly lower vertical release velocity, longer air-borne duration, smaller hip-shoulder separation angles at left foot off and touchdown instant, and longer shot travel distance during air-borne flight, but shorter shot travel distance during final shot delivery. Conclusions The difference in lower extremity strength is a primary casue leading to different sports performance between Chinese and world elite male shot putters. The technique differences in Chinese and world elite male shot putters mainly lie in different phase timing and shot travel distances during different techinique phases.

Objective: To compare the long-term outcomes in ST-elevation myocardial infarction (STEMI) patients who underwent early or late delayed percutaneous coronary intervention (PCI) using drug-eluting stents (DES). Methods: This study was a retrospective, observational and single-center study. Consecutive STEMI patients (n=977), who admitted to Fuwai Hospital in 2013 and underwent successful selective PCI using drug-eluting stents (DES) within 3 to 35 days after symptom onset were enrolled and divided into the early delayed PCI (3-14 d) group (n=495) and the late delayed PCI (15-35 d) group (n=482). General clinical data of the patients and related data of coronary angiography and interventional therapy were collected, and the endpoint events were followed up. The primary endpoint was 2-year major adverse cardiac and cerebrovascular events (MACCE) including cardiac death, recurrent myocardial infarction, definite or probable stent thrombosis and ischemic stroke. The secondary endpoint was 2-year ischemia-driven target vessel revascularization. The incidence of endpoint events of the two groups was compared, and it was compared again after the primary baseline characteristics such as age and gender were matched by the propensity scoring method at a 1∶1 ratio. Results: A total of 910 (93.1%) patients who underwent delayed PCI were transferred from other hospitals, and 292 (29.9%) patients received thrombolysis before PCI. The time interval before PCI was 14 (10, 20) days. The incidence of 2-year MACCE (3.0%(15/495) vs. 2.3%(11/482), P=0.468) and ischemia-driven target vessel revascularization (3.8%(19/495) vs. 5.0%(24/482), P=0.385) were similar between the two groups. The incidence of 2-year MACCE (3.3%(15/453 vs. 2.4%(11/453), P=0.426) and ischemia-driven target vessel revascularization (4.2% (19/453) vs. 4.9%(22/453), P=0.632) were also similar between the two groups after matching propensity score. Conclusion The long-term clinical outcomes after early delayed PCI using DES is statistically equivalent to those of late delayed PCI using DES for STEMI patients who missed the time window for emergency PCI.

Objective To build a method for calculating the optimal length of hamstring muscles in vivo, and make comparison with other indirect parameters which represent the optimal length. Methods By synchronously recording knee flexion torque and kinematic data, the musculoskeletal model of lower limbs was built to obtain hamstring strength and muscle length, and to further calculate the optimal length of hamstring muscles. Results Flexion angle at peak knee flexion torque was significantly greater than that at peak hamstring strength and their correlation coefficient was 0.741. The optimal lengths of each bi-articulated hamstring muscles were significantly greater than the corresponding muscle lengths during standing and their correlation coefficients was low. Conclusions The established estimating method for optimal length of hamstring muscles provided references for future studies on injury mechanism and risk factors. Flexion angle at peak knee flexion torque could partly represent the optimal length of hamstring. It is not suggested that hamstring muscle length during standing should be used as an approximation of hamstring optimal length.