Prediction of pulp exposure risk of carious pulpitis based on deep learning / 华西口腔医学杂志

OBJECTIVES@#This study aims to predict the risk of deep caries exposure in radiographic images based on the convolutional neural network model, compare the prediction results of the network model with those of senior dentists, evaluate the performance of the model for teaching and training stomatological students and young dentists, and assist dentists to clarify treatment plans and conduct good doctor-patient communication before surgery.@*METHODS@#A total of 206 cases of pulpitis caused by deep caries were selected from the Department of Stomatological Hospital of Tianjin Medical University from 2019 to 2022. According to the inclusion and exclusion criteria, 104 cases of pulpitis were exposed during the decaying preparation period and 102 cases of pulpitis were not exposed. The 206 radiographic images collected were randomly divided into three groups according to the proportion: 126 radiographic images in the training set, 40 radiographic images in the validation set, and 40 radiographic images in the test set. Three convolutional neural networks, visual geometry group network (VGG), residual network (ResNet), and dense convolutional network (DenseNet) were selected to analyze the rules of the radiographic images in the training set. The radiographic images of the validation set were used to adjust the super parameters of the network. Finally, 40 radiographic images of the test set were used to evaluate the performance of the three network models. A senior dentist specializing in dental pulp was selected to predict whether the deep caries of 40 radiographic images in the test set were exposed. The gold standard is whether the pulp is exposed after decaying the prepared hole during the clinical operation. The prediction effect of the three network models (VGG, ResNet, and DenseNet) and the senior dentist on the pulp exposure of 40 radiographic images in the test set were compared using receiver operating characteristic (ROC) curve, area under the ROC curve (AUC), accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1 score to select the best network model.@*RESULTS@#The best network model was DenseNet model, with AUC of 0.97. The AUC values of the ResNet model, VGG model, and the senior dentist were 0.89, 0.78, and 0.87, respectively. Accuracy was not statistically different between the senior dentist (0.850) and the DenseNet model (0.850)(P>0.05). Kappa consistency test showed moderate reliability (Kappa=0.6>0.4, P<0.05).@*CONCLUSIONS@#Among the three convolutional neural network models, the DenseNet model has the best predictive effect on whether deep caries are exposed in imaging. The predictive effect of this model is equivalent to the level of senior dentists specializing in dental pulp.

Non-invasive electrical neuromodulation techniques: analgesic effects and neural mechanisms / 生理学报

As non-pharmaceutical interventions, non-invasive electrical neuromodulation techniques are promising in pain management. With many advantages, such as low costs, high usability, and non-invasiveness, they have been exploited to treat multiple types of clinical pain. Proper use of these techniques requires a comprehensive understanding of how they work. In this article, we reviewed recent studies concerning non-invasive electrical peripheral nerve stimulation (transcutaneous electrical nerve stimulation and transcutaneous vagus/vagal nerve stimulation) as well as electrical central nerve stimulation (transcranial direct current stimulation and transcranial alternating current stimulation). Specifically, we discussed their analgesic effects on acute and chronic pain, and the neural mechanisms thereof. We then contrasted the four kinds of nerve stimulation techniques, pointing out limitations of existing studies and proposing directions for future research. With more extensive and in-depth research to overcome these limitations, we shall witness more clinical applications of non-invasive electrical nerve stimulations to alleviate patients' pain and ease the crippling medical and economic burden imposed on patients, their families, and the entire society.

Artificial Intelligence Meets Chinese Medicine / 中国结合医学杂志

As an interdisciplinary subject of medicine and artificial intelligence, intelligent diagnosis and treatment has received extensive attention. The standardization of Chinese medicine (CM) diagnosis has been always a bottleneck in the modernization and globalization of CM. Studying the application technology of artificial intelligence in CM and solving the problems is an urgent need for the development of modern CM in the era of artificial intelligence. Taking the pneumonia with dyspnea and cough in CM as an example, this article gives an overview of intelligent medical technology and application development, brings forward the present technical problems faced and the new advances in intelligent technology on CM diagnosis and treatment.

Functional role of MicroRNA-19b in acinar cell necrosis in acute necrotizing pancreatitis / 华中科技大学学报（医学）（英德文版）

The expression of microRNA-19b (miR-19b) in acute necrotizing pancreatitis (ANP) and its functional role in acinar cell necrosis of SD rats were investigated. Twelve SD rats were divided into two groups randomly, including control group and ANP group. The rat ANP models were established by intraperitoneal injection of L-arginine (2400 mg/kg body weight), and equal volume of 0.9% NaCl was injected in the control group. MiRNA chip assay was performed to examine the expression of miRNAs in the pancreas in two different groups. Besides, to further explore the role of miR-19b in ANP in vitro, taurolithocholic acid 3-sulfate disodium salt (TLC-S) (200 μmol/L) was administrated to treat the rat pancreatic acinar cell line, AR42J, for establishing the ANP cells model. The quantitative real-time PCR (qRT-PCR) was adopted to measure the miR-19b expression. Moreover, the mimic miRNA, miRNA antisense oligonucleotide (AMO) and control vector were used to transfect AR42J cells, the expression of miR-19b was confirmed by qRT-PCR and the necrotizing rate of AR42J cells was detected with AO/EB method. The expression of miR-19b was significantly higher in ANP group than in control group as displayed by the miRNA chip assay. Furthermore, after inducing necrosis of AR42J cells in vitro, the expression of miR-19b was significantly increased by 2.51±0.14 times in comparison with the control group. As revealed by qRT-PCR assay, the expression of miR-19b was 5.94±0.95 times higher in the mimic miRNA group than in the control vector group, companied with an obviously increased acinar cell necrotizing rate (50.3%±1.5% vs. 39.6%±2.3%, P<0.05). Moreover, the expression of miR-19b in the miRNA AMO group was 0.38±0.15 times lower than in the control vector group, and the cell necrosis rate was much lower accordingly (23.1%±3.3% vs. 39.6%±2.3%, P<0.05). Besides, there was no significant difference between the control vector cells and the cells without treatment (P>0.05). The expression of miR-19b was significantly induced in ANP. In addition, up-regulation of miR-19b could promote the necrosis of pancreatic acinar cells and miR-19b deficiency could decrease the rate of pancreatic acinar cell necrosis.

Functional role of MicroRNA-19b in acinar cell necrosis in acute necrotizing pancreatitis / 华中科技大学学报（医学）（英德文版）

The expression of microRNA-19b (miR-19b) in acute necrotizing pancreatitis (ANP) and its functional role in acinar cell necrosis of SD rats were investigated. Twelve SD rats were divided into two groups randomly, including control group and ANP group. The rat ANP models were established by intraperitoneal injection of L-arginine (2400 mg/kg body weight), and equal volume of 0.9% NaCl was injected in the control group. MiRNA chip assay was performed to examine the expression of miRNAs in the pancreas in two different groups. Besides, to further explore the role of miR-19b in ANP in vitro, taurolithocholic acid 3-sulfate disodium salt (TLC-S) (200 μmol/L) was administrated to treat the rat pancreatic acinar cell line, AR42J, for establishing the ANP cells model. The quantitative real-time PCR (qRT-PCR) was adopted to measure the miR-19b expression. Moreover, the mimic miRNA, miRNA antisense oligonucleotide (AMO) and control vector were used to transfect AR42J cells, the expression of miR-19b was confirmed by qRT-PCR and the necrotizing rate of AR42J cells was detected with AO/EB method. The expression of miR-19b was significantly higher in ANP group than in control group as displayed by the miRNA chip assay. Furthermore, after inducing necrosis of AR42J cells in vitro, the expression of miR-19b was significantly increased by 2.51±0.14 times in comparison with the control group. As revealed by qRT-PCR assay, the expression of miR-19b was 5.94±0.95 times higher in the mimic miRNA group than in the control vector group, companied with an obviously increased acinar cell necrotizing rate (50.3%±1.5% vs. 39.6%±2.3%, P0.05). The expression of miR-19b was significantly induced in ANP. In addition, up-regulation of miR-19b could promote the necrosis of pancreatic acinar cells and miR-19b deficiency could decrease the rate of pancreatic acinar cell necrosis.

Role of vascular cell adhesion molecule-1 in the mouse model of hepatic ischemia/reperfusion and the hematogenic metastasis / 中国医学科学院学报

<p><b>OBJECTIVE</b>To investigate the effect of ischemia/reperfusion (I/R) on tumor metastasis in a experimental mouse model of hematogenous metastasis after I/R and to quantify expression of vascular cell adhesion molecule-1 (VCAM-1) during I/R.</p><p><b>METHODS</b>An experimental mouse model of metastasis after partial hepatic I/R was designed to determine the effects of I/R on tumor metastasis to liver. Tumor loads were valued 14 days after operation. In addition, the expressions of alanine transaminase (ALT), aspartate transaminase (AST), and VCAM-1 were detected.</p><p><b>RESULTS</b>Two hours after hepatic reperfusion, ALT and AST levels in ischemia 45-minute group and ischemia 30-minute group were significantly higher than in the sham group (all P < 0.05). Also, the changes of ALT and AST were more obvious in the ischemia 45-minute group than in ischemia 30-minute group (all P < 0.05). In the sham group, both ALT and AST slightly and transiently increased. ALT and AST in the ischemia 45-minute group and ischemia 30-minute group at 8 hours were both significantly higher than those at 2 hours reperfusion (P<0.05). The tumor load (valued by hepatic replacement area) and the expression of VCAM-1 in ischemic lobe were significantly larger in the ischemia 45-minute group than in the ischemia 30-minute group and sham group (P = 0.013, P = 0.007). However, there was no statistical difference on tumor load between the right lobe of sham operated mice and the right lobe (nonischemic lobes) of mice subjected to I/R (P = 0.089). Mouse survivals were significantly longer in the sham group than in the ischemia 30-minute group (P = 0.041) but were not significantly different between the ischemia 45-minute group and ischemia 30-minute group (P = 0.055). VCAM-1 expression in ischemia 45-minute group was significantly higher than in ischemia 30-minute group and sham group(P = 0.003, P < 0.001), and it was positively correlated with the hepatic replacement area (r = 0.491, P = 0.045).</p><p><b>CONCLUSION</b>Hepatic I/R promotes liver hematogenic metastasis of hepatocellular carcinoma in mice and at least in part, through the induction of VCAM-1 expression.</p>

Whole-liver perfusion imaging by multi-slice spiral computed tomography based on Couinaud segments for evaluation of the blood flow state in patients with hepatitis B cirrhosis / 中华肝脏病杂志

<p><b>OBJECTIVE</b>To investigate the blood flow characteristics of hepatitis B cirrhosis based on Couinaud's hepatic segments and to analyze the correlation between perfusion parameters of liver cirrhosis in the hepatic segments and between the different Child-Pugh classifications by using the pattern of whole liver perfusion detected by multi-slice spiral computed tomography (MSCT).</p><p><b>METHODS</b>Whole-liver perfusion enhanced CT imaging scan was performed for 51 patients with hepatitis B cirrhosis (including 24 classified as Child-Pugh A, 19 as Child-Pugh B, and 8 as Child-Pugh C) and 20 patients without any liver abnormalities (who served as the control group).The perfusion parameters of Couinaud's segments were measured in order to compare the blood perfusion differences among the Couinaud's hepatic segments in liver cirrhosis. The blood flow characteristics were analyzed for each lobe and compared between groups, and the time to peak and peak of enhancement values of the aorta and portal veins and spleen were recorded and compared between groups.The F-test was used for statistical analysis.</p><p><b>RESULTS</b>The liver cirrhosis group showed higher time to peak for aorta and portal veins and spleen but lower peak of enhancement values than the control group. The hepatic arterial perfusion (HAP) was significantly higher in segment 3 than in segment 7 were (11.40+/- 5.72 vs.9.46+/-5.18 mL/min/100 mL; P less than 0.05). In addition, the differences of the portal venous perfusion (PVP) and total hepatic perfusion (THP) were significant between the liver lobes in various groups (P less than 0.05). The HAP and hepatic perfusion index (HPI) were not significantly different between the groups, but the HAP and HPI in every lobe were higher in the Child C subgroup than in either the Child A or Child B subgroups.</p><p><b>CONCLUSION</b>The cirrhotic liver has different blood flow in the various Couinaud's segments. Whole-liver perfusion MSCT can reflect the hemodynamic changes of liver cirrhosis and Child-Pugh classification, and as such may be helpful for distinguishing the normal liver from the cirrhotic liver.</p>