In order to realize the quantitative assessment of muscle strength in hand function rehabilitation and then formulate scientific and effective rehabilitation training strategies, this paper constructs a multi-scale convolutional neural network (MSCNN) - convolutional block attention module (CBAM) - bidirectional long short-term memory network (BiLSTM) muscle strength prediction model to fully explore the spatial and temporal features of the data and simultaneously suppress useless features, and finally achieve the improvement of the accuracy of the muscle strength prediction model. To verify the effectiveness of the model proposed in this paper, the model in this paper is compared with traditional models such as support vector machine (SVM), random forest (RF), convolutional neural network (CNN), CNN - squeeze excitation network (SENet), MSCNN-CBAM and MSCNN-BiLSTM, and the effect of muscle strength prediction by each model is investigated when the hand force application changes from 40% of the maximum voluntary contraction force (MVC) to 60% of the MVC. The research results show that as the hand force application increases, the effect of the muscle strength prediction model becomes worse. Then the ablation experiment is used to analyze the influence degree of each module on the muscle strength prediction result, and it is found that the CBAM module plays a key role in the model. Therefore, by using the model in this article, the accuracy of muscle strength prediction can be effectively improved, and the characteristics and laws of hand muscle activities can be deeply understood, providing assistance for further exploring the mechanism of hand functions .
Neural Networks, Computer ; Humans ; Hand Strength/physiology* ; Support Vector Machine ; Muscle Strength/physiology* ; Hand/physiology* ; Convolutional Neural Networks

Objective:To explore the role of spinal Reelin protein (RELN) in neuropathic pain and its related mechanisms.Methods:A rat model of neuropathic pain induced by chronic constriction injury (CCI) was established using the sciatic nerve ligation method. The mechanical threshold and thermal threshold of the injured side and contralateral side in the sham-operation group and CCI group were compared. Western blot was used to detect the differences in the expressions of spinal RELN, calcium/calmodulin-dependent protein kinase Ⅱ (CAMKⅡ) and phosphorylated extracellular signal-regulated kinase 1/2 (p-ERK1/2) proteins. On the 7th day after CCI modeling, the CCI model rats were further divided into three groups: CCI group (without any treatment), CCI+ RELN overexpression group (intrathecal injection of 15 μl of 5 μg/μl RELN overexpression plasmid, once a day for 2 consecutive days) and CCI+ PBS group (intrathecal injection of PBS). The mechanical threshold and thermal threshold among the three groups were compared, and Western blot was used again to detect the differences in the expressions of RELN, CaMKⅡ and p-ERK1/2 proteins in the three groups.Results:CCI successfully induced neuropathic pain in rats. On the 7th day after CCI, compared with the contralateral hind paw or the injured hind paw in the sham-operation group, the mechanical threshold and thermal threshold of the injured hind paw in the CCI group were significantly lower, with statistically significant differences (all P<0.001). Western blot results showed that compared with the sham-operation group, the protein expression of RELN in the spinal dorsal horn of the injured side in the CCI group was lower ( P=0.031), the protein expression of CAMKⅡ and the level of p-ERK1/2 were higher (all P<0.05), and there was no statistically significant difference in the level of ERK1/2 among the groups ( P>0.05). The thermal threshold and mechanical threshold of the injured side in the CCI+ RELN overexpression group were significantly higher than those in the CCI group and CCI+ PBS group (all P<0.05). Western blot results showed that 24 hours after the transfection of RELN overexpression plasmid, compared with the CCI+ PBS group, the protein expression of RELN in the CCI+ RELN overexpression group was significantly increased, with a statistically significant difference ( P<0.05), indicating that the transfection of RELN overexpression plasmid was successful. Compared with the CCI+ PBS group, the protein expression of CAMKⅡ and the phosphorylation level of ERK2 in the CCI+ RELN overexpression group were lower (all P<0.05), while there was no statistically significant difference in the phosphorylation level of ERK1 ( P>0.05). Conclusions:The overexpression of the RELN gene in the spinal cord weakens the maintenance of neuropathic pain by inhibiting the activation of the CAMKⅡ/ERK2 pathway, which suggests that RELN may become a new target for pain treatment.

Objective:To explore the role of spinal Reelin protein (RELN) in neuropathic pain and its related mechanisms.Methods:A rat model of neuropathic pain induced by chronic constriction injury (CCI) was established using the sciatic nerve ligation method. The mechanical threshold and thermal threshold of the injured side and contralateral side in the sham-operation group and CCI group were compared. Western blot was used to detect the differences in the expressions of spinal RELN, calcium/calmodulin-dependent protein kinase Ⅱ (CAMKⅡ) and phosphorylated extracellular signal-regulated kinase 1/2 (p-ERK1/2) proteins. On the 7th day after CCI modeling, the CCI model rats were further divided into three groups: CCI group (without any treatment), CCI+ RELN overexpression group (intrathecal injection of 15 μl of 5 μg/μl RELN overexpression plasmid, once a day for 2 consecutive days) and CCI+ PBS group (intrathecal injection of PBS). The mechanical threshold and thermal threshold among the three groups were compared, and Western blot was used again to detect the differences in the expressions of RELN, CaMKⅡ and p-ERK1/2 proteins in the three groups.Results:CCI successfully induced neuropathic pain in rats. On the 7th day after CCI, compared with the contralateral hind paw or the injured hind paw in the sham-operation group, the mechanical threshold and thermal threshold of the injured hind paw in the CCI group were significantly lower, with statistically significant differences (all P<0.001). Western blot results showed that compared with the sham-operation group, the protein expression of RELN in the spinal dorsal horn of the injured side in the CCI group was lower ( P=0.031), the protein expression of CAMKⅡ and the level of p-ERK1/2 were higher (all P<0.05), and there was no statistically significant difference in the level of ERK1/2 among the groups ( P>0.05). The thermal threshold and mechanical threshold of the injured side in the CCI+ RELN overexpression group were significantly higher than those in the CCI group and CCI+ PBS group (all P<0.05). Western blot results showed that 24 hours after the transfection of RELN overexpression plasmid, compared with the CCI+ PBS group, the protein expression of RELN in the CCI+ RELN overexpression group was significantly increased, with a statistically significant difference ( P<0.05), indicating that the transfection of RELN overexpression plasmid was successful. Compared with the CCI+ PBS group, the protein expression of CAMKⅡ and the phosphorylation level of ERK2 in the CCI+ RELN overexpression group were lower (all P<0.05), while there was no statistically significant difference in the phosphorylation level of ERK1 ( P>0.05). Conclusions:The overexpression of the RELN gene in the spinal cord weakens the maintenance of neuropathic pain by inhibiting the activation of the CAMKⅡ/ERK2 pathway, which suggests that RELN may become a new target for pain treatment.

Objectives:To explore the dynamic changes of right ventricular global longitudinal strain (RVGLS) on echocardiography during the process of right ventricular fibrosis with right ventricular pressure overload.Methods:Animal models with RV pressure overload (n=5) were established by main pulmonary artery (PA) banding in piglets.The ratio of RV pressure to left ventricular (LV) pressure was greater than 60％ as assessed by the pressure measuring needle during the operation.Pre-and post-operative echocardiography were performed.Histopathological examination of RV tissues was performed at the end of the experiment.Results:During the establishment of the animal models,2 piglets died due to hemorrhagic shock,while 3 piglets survived and were included in the analysis.Systolic RV pressure,systolic PA pressure and mean PA pressure(measured at the proximal end of the constriction) were significantly increased immediately after PA banding,mean RV pressure and mean PA pressure were significantly inreased at 14 weeks postopseratively (all P<0.05).Tricuspid annulus plane systolic excursion and peak early diastolic velocity of tricuspid annulus (TV e') decreased immediately after operation (all P<0.05).They returned to the pre-operative level in the following three weeks.The absolute values of RVGLS were reduced after PA banding (all P>0.05).The RVGLS of the inner membrane layer began to recover after 3 weeks,and the RVGLS of the middle and outer membrane layers began to recover after 7 weeks.However,the time to peak RVGLS was significantly prolonged and could not be restored even untill the end of the experiment (14 weeks after surgery)(all P<0.05).At the end of the experiment,pathological examination revealed fibrous tissue hyperplasia in both the myocardial interstitium and endocardium.Conclusions:The measurement of intracardiac pressure and pathological results confirmed the success of the establishment of the animal model of increased right ventricular afterload.Both stress and strain cannot truly represent the intrinsic characteristics of the myocardium under pressure overload.The prolonged time to peak of longitudinal strain might indirectly reflect the progression of myocardial injury more persistently during the process of increased right ventricular afterload and myocardial injury.

Objectives:To evaluate the diagnostic value of echocardiography in unroofed coronary sinus syndrome(UCSS)with endocardial cushion defect(ECD). Methods:The echocardiographic data of 43 patients of UCSS with ECD who underwent surgical treatment in our hospital from July 2017 to May 2022 were retrospectively analyzed.The diagnostic accuracy was evaluated by comparing the echocardiographic findings with the intraoperative exploration results. Results:According to Kirklin and Barratt-Boyes classification,there were 26 cases of type Ⅰ(60.5%),9 cases of type Ⅱ(20.9%),2 cases of type Ⅲ(4.7%),4 cases of type Ⅳ(9.3%),and 2 cases of type Ⅲ combined with type Ⅳ(4.7%).There were 26 cases(60.5%)of partial ECD,7 cases(16.3%)of intermediate ECD,10 cases(23.3%)of total ECD.Twenty two cases(51.2%)were associated with single atrium.Twenty seven cases(62.8%)were associated with persistent left superior vena cava(PLSVC).Other coexisting complicated malformations were as follows:2 cases of double outlet of right ventricle,1 case of pulmonary atresia,1 case of total anomalous pulmonary venous connection,and 1 case of aplenia syndrome.The coexisting simple malformations included 4 cases of ostium secundum atrial septal defect,2 cases of ventricular septal defect,3 cases of patent ductus arterial,and 6 cases of patent foramen ovale.Other abnormalities included 5 cases of absence of hepatic segment of inferior vena cava,1 case of hypoplasia of right superior vena cava,1 case of absence of right superior vena cava,3 cases of cor triatriatum,1 case of isolated levocardia,1 case of mirror image dextrocardia,4 cases of right aortic arch.Of the 43 patients,30(69.8%)were correctly diagnosed by preoperative echocardiography and 13(30.2%)by intraoperative exploration.UCSS was misdiagnosed as inferior vena cava type sinus septal defect and PLSVC was missed in 1 case.UCSS was missed in 12 cases,and PLSVC was missed in 3 cases of them. Conclusions:Diagnosis UCSS with ECD by echocardiography is valuable and challenging.It is necessary to strengthen the understanding of the embryonic development and pathological anatomy characteristics of these malformations to improve the diagnostic accuracy.

Objective To investigate the prenatal diagnosis and perinatal outcomes between anterior pla-centa accreta and non-anterior placenta.Methods A retrospective analysis was done for 560 pregnant women who were diagnosed with placenta accreta and delivered in the Third Affiliated Hospital of Guangzhou Medical Uni-versity.According to the location of the placenta,the group was dividing into anterior placenta accreta group(319 cases)and non-anterior placenta accreta group(241 cases).The general characteristics,maternal and infant out-comes of the two groups were analyzed.The non-anterior placenta accrete group(241 cases)then were dividing into two groups according to the time of clear diagnosis.Those who were firstly diagnosed with placenta accrete dur-ing or after the operation was the intrapartum diagnosis group(missed diagnosis)(70 cases),and those who were diagnosed with clear placenta accreta before the delivery was prenatal diagnosis group(171 cases).The general characteristics,maternal and infant outcomes of the two groups were also analyzed.Results There were statisti-cally significant differences in the parity,history of cesarean section,delivery mode,degree of placenta accreta,missed diagnosis rate,neonatal birth weight,and hysterectomy rate between the non-anterior placenta accrete group and the anterior placenta accreta group.In the case of prenatal diagnosis of different degrees of placenta accreta,the prenatal diagnosis rate of placental adhesion in the non-anterior placenta accreta group was lower than that of the anterior placenta accreta group,which was statistically significant.In the non-anterior placenta accrete group,there were statistically significant differences in the age,cesarean section history,placenta previa status,mode of delivery,degree of implantation,24-hour bleeding volume,blood transfusions,NICU transfer rate,uterine loss rate between the intrapartum diagnosis group(missed diagnosis)and the prenatal diagnosis group.Conclusions The high-risk factors of patients with non-anterior placenta accreta are different from those of patients with anterior placenta accreta.Multiple births and a history of cesarean section are high-risk factors for anterior placenta accreta patients.Non-anterior placenta accreta are more likely to be missed diagnosed,especially the placental adhesion.For pregnant women with non-anterior placenta accreta missed diagnosis,there is a high rate of adverse birth out-comes,especially in the rate of neonatal transfer to the NICU.

Objective:To investigate the correlation between 25-hydroxyvitamin D3 (25 (OH) D3), insulin-like growth factor 1 (IGF-1) and glycolipid metabolism in patients with diabetes 2 mellitus (T2DM), as well as their predictive value in retinopathy.Methods:A total of 120 T2DM patients admitted to Linyi Central Hospital of Shandong Province from Oct. 2020 to Oct.r 2023 were selected as the study objects (defined as the study group). Another 120 healthy volunteers who underwent physical examination in our hospital during the same period were selected as the control group. Serum 25 (OH) D3, IGF-1, fasting blood glucose (FBG), 2 h plasma glucose (2 hPG) ) and lipid levels (triglycerides) were compared between the two groups. The levels of TG, total cholesterol (TC) and serum 25 (OH) D3 and IGF-1 were analyzed by Pearson correlation analysis. At the same time, the patients in the study group were divided into diabetic group with retinopathy (DR Group, 40 cases) and non-retinopathy group (NDR group, 80 cases) according to the status of retinopathy. Multivariate analysis was used to analyze the risk factors affecting the occurrence of retinopathy in T2DM patients, and receiver operating characteristic (ROC) curve was drawn to analyze the predictive value of serum 25 (OH) D3 and IGF-1 levels in the occurrence of retinopathy in T2DM patients.Results:The level of serum 25 (OH) D3 was (36.15±4.25) nmol/L in the study group, lower than that in the control group (51.24±5.32) nmol/L ( P<0.05), and the level of IGF-1 was (30.26±4.52) mg/L was in the study group, higher than that in the control group ( P<0.05). The levels of FBG, 2 hPG, TG and TC in the study group were (8.67±2.52) mmol/L, (11.36±2.43) mmol/L, (2.05±0.72) mmol/L, (5.05±1.54) mmol/L respectively, higher than those in the control group [ (5.02±0.42) mmol/L, (6.32±0.54) mmol/L, (1.21±0.32) mmol/L, (3.42±0.68) mmol/L] ( P<0.05). Pearson correlation analysis showed that serum 25 (OH) D3 levels were negatively correlated with FBG, 2 hPG, TG and TC levels in T2DM patients ( r=-0.762, -0.782, -0.736, -0.721, P<0.05). Serum IGF-1 levels were positively correlated with the levels of FBG, 2 hPG, TG and TC in T2DM patients ( r=0.741, 0.756, 0.715, 0.698, P<0.05). Family history of diabetes in DR group, FBG, 2 hPG, TG, TC, IGF-1 levels was 35.00%, (9.31±2.49) mmol/L, (12.52±2.34) mmol/L, (2.76±0.61) mmol/L, (5.92±1.42) mmol/L, (37.89±4.41) mg/L respectively, higher than those in NDR group [16.25%, (8.35±2.15) mmol/L, (10.78±1.75) mmol/L, (1.69±0.52) mmol/L, (4.62±1.31) mmol/L, (26.45±4.06) mg/L] ( P<0.05). 25 (OH) D3 in DR group was (30.21±3.51) nmol/L, lower than that in NDR group (39.12±3.85) nmol/L ( P<0.05). Logistic regression analysis showed that family history of diabetes mellitus, duration of diabetes mellitus, 25 (OH) D3, IGF-1, FBG, 2hPG, TG and TC levels were all risk factors for the occurrence of retinopathy in elderly T2DM patients ( P<0.05). ROC curve analysis showed that AUC and sensitivity of 25 (OH) D3 and IGF-1 combined to predict retinopathy in elderly T2DM patients were 0.854 and 92.50%, respectively, higher than that of 25 (OH) D3 and IGF-1 alone ( P<0.05) . Conclusion:Serum 25 (OH) D3 and IGF-1 levels are abnormally expressed in elderly patients with T2DM, and there is a close relationship between glucose and lipid metabolism in elderly patients with T2DM, and the combined detection of these indicators has a higher predictive value for the occurrence of DR In elderly patients with T2DM.

Caspase-2, a highly conserved member of the caspase family, is considered an initiator caspase that triggers apoptosis in response to some cellular stresses. Previous studies suggest that an intracellular multi-protein complex PIDDosome, induced by genotoxic stress, serves as a platform for caspase-2 activation. Due to caspase-2's inability to process effector caspases, however, the mechanism underlying caspase-2-mediated cell death upon PIDDosome activation remains unclear. Here, we conducted an unbiased genome-wide genetic screen and identified that the Bcl2 family protein BID is required for PIDDosome-induced, caspase-2-mediated apoptosis. PIDDosome-activated caspase-2 directly and functionally processes BID to signal the mitochondrial pathway for apoptosis induction. In addition, a designed chemical screen identified a compound, HUHS015, which specifically activates caspase-2-mediated apoptosis. HUHS015-stimulated apoptosis also requires BID but is independent of the PIDDosome. Through extensive structure-activity relationship efforts, we identified a derivative with a potency of ~60 nmol/L in activating caspase-2-mediated apoptosis. The HUHS015-series of compounds act as efficient agonists that directly target the interdomain linker in caspase-2, representing a new mode of initiator caspase activation. Human and mouse caspase-2 differ in two crucial residues in the linker, rendering a selectivity of the agonists for human caspase-2. The caspase-2 agonists are valuable tools to explore the physiological roles of caspase-2-mediated cell death and a base for developing small-molecule drugs for relevant diseases.
Caspase 2/genetics* ; Humans ; BH3 Interacting Domain Death Agonist Protein/metabolism* ; Apoptosis/drug effects* ; Death Domain Receptor Signaling Adaptor Proteins/metabolism* ; Animals ; Mice ; Cysteine Endopeptidases

Objectives:To explore the dynamic changes of right ventricular global longitudinal strain (RVGLS) on echocardiography during the process of right ventricular fibrosis with right ventricular pressure overload.Methods:Animal models with RV pressure overload (n=5) were established by main pulmonary artery (PA) banding in piglets.The ratio of RV pressure to left ventricular (LV) pressure was greater than 60％ as assessed by the pressure measuring needle during the operation.Pre-and post-operative echocardiography were performed.Histopathological examination of RV tissues was performed at the end of the experiment.Results:During the establishment of the animal models,2 piglets died due to hemorrhagic shock,while 3 piglets survived and were included in the analysis.Systolic RV pressure,systolic PA pressure and mean PA pressure(measured at the proximal end of the constriction) were significantly increased immediately after PA banding,mean RV pressure and mean PA pressure were significantly inreased at 14 weeks postopseratively (all P<0.05).Tricuspid annulus plane systolic excursion and peak early diastolic velocity of tricuspid annulus (TV e') decreased immediately after operation (all P<0.05).They returned to the pre-operative level in the following three weeks.The absolute values of RVGLS were reduced after PA banding (all P>0.05).The RVGLS of the inner membrane layer began to recover after 3 weeks,and the RVGLS of the middle and outer membrane layers began to recover after 7 weeks.However,the time to peak RVGLS was significantly prolonged and could not be restored even untill the end of the experiment (14 weeks after surgery)(all P<0.05).At the end of the experiment,pathological examination revealed fibrous tissue hyperplasia in both the myocardial interstitium and endocardium.Conclusions:The measurement of intracardiac pressure and pathological results confirmed the success of the establishment of the animal model of increased right ventricular afterload.Both stress and strain cannot truly represent the intrinsic characteristics of the myocardium under pressure overload.The prolonged time to peak of longitudinal strain might indirectly reflect the progression of myocardial injury more persistently during the process of increased right ventricular afterload and myocardial injury.

Objective:To evaluate the diagnostic accuracy of the convolutional neural network（CNN） in diagnosing nasopharyngeal carcinoma using endoscopic narrowband imaging. Methods:A total of 834 cases with nasopharyngeal lesions were collected from the People's Hospital of Guangxi Zhuang Autonomous Region between 2014 and 2016. We trained the DenseNet201 model to classify the endoscopic images, evaluated its performance using the test dataset, and compared the results with those of two independent endoscopic experts. Results:The area under the ROC curve of the CNN in diagnosing nasopharyngeal carcinoma was 0.98. The sensitivity and specificity of the CNN were 91.90% and 94.69%, respectively. The sensitivity of the two expert-based assessment was 92.08% and 91.06%, respectively, and the specificity was 95.58% and 92.79%, respectively. There was no significant difference between the diagnostic accuracy of CNN and the expert-based assessment （P=0.282, P=0.085）. Moreover, there was no significant difference in the accuracy in discriminating early-stage and late-stage nasopharyngeal carcinoma（P=0.382）. The CNN model could rapidly distinguish nasopharyngeal carcinoma from benign lesions, with an image recognition time of 0.1 s/piece. Conclusion:The CNN model can quickly distinguish nasopharyngeal carcinoma from benign nasopharyngeal lesions, which can aid endoscopists in diagnosing nasopharyngeal lesions and reduce the rate of nasopharyngeal biopsy.
Humans ; Nasopharyngeal Carcinoma ; Narrow Band Imaging ; China ; Neural Networks, Computer ; Nasopharyngeal Neoplasms/diagnostic imaging*