Objective To explore the approach to establish a temporal lobe epilepsy model via intraperitoneal pilocarpine injection in C57BL/6J mice,and to summarize behavioral indicators predicting successful modeling during the acute phase of epileptic seizures after pilocarpine administration,aiming to offer a practical mice model for future epilepsy research.Methods Thirty C57BL/6J substrain mice(primary subjects)and forty C57BL/6N substrain mice(control subjects)were selected to establish a temporal lobe epilepsy model by inducing seizures through a single intraperitoneal injection of pilocarpine.The mice from the two substrains were each divided into 3 groups,and were injected intraperitoneally with 300 mg/kg,330 mg/kg,or 360 mg/kg of pilocarpine,respectively.Motor seizure behaviors were observed and compared between the two substrains of C57BL/6 mice post pilocarpine injection,and the spontaneous recurrent seizures(SRS)were continuously monitored from the 7th day after injection.On the 28th day post-injection,the mice were euthanized and the histopathological changes in their hippocampi were examined.Results After pilocarpine administration,C57BL/6N mice displayed characteristic motor seizures followed by the onset of status epilepticus(SE).Conversely,C57BL/6J mice showed fewer instances of typical motor seizure behavior and the subsequent SE.Instead,they more often exhibited systemic tremors lasting several seconds to tens of seconds following limb twitching.This behavior is classified as"severe seizure(SS)"in current study.Following intraperitoneal injection of 330 mg/kg and 360 mg/kg pilocarpine,C57BL/6J mice displaying SS during the acute phase of seizure might exhibit SRS after a latency period.The percentage of spontaneous seizures observed in C57BL/6J mice post-modeling(70% )was comparable to that seen in C57BL/6N mice(75% )which developed SRS subsequent to SE.C57BL/6J mice displayed characteristic pathological alterations associated with temporal lobe epilepsy in the hippocampi after 28 d following pilocarpine injection,including increased mossy fiber sprouting and neuronal death.Conclusions When inducing an epilepsy model via intraperitoneal pilocarpine injection in C57BL/6J mice,the behavioral criteria to predict the successful establishment of the model could be either the occurrence of SE or the manifestation of SS.

Objective::To determine whether deep learning algorithms are suitable for predicting preterm birth.Methods::A retrospective study was conducted at Peking University Third Hospital from January 2018 to June 2023. Birth data were divided into two parts based on the date of delivery: the first part was used for model training and validation, while real world viability was evaluated using the second part. Four machine learning algorithms (logistic regression, random forest, support vector machine, and transformer) were employed to predict preterm birth. Receiver operating characteristic curves were plotted, and the area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated.Results::This research included data on 30,965 births, where 24,770 comprised the first part, and included 3164 (12.77%) in the preterm birth group, with 6195 in the second part, including 795 (12.83%) in the preterm birth group. Significant differences in various factors were observed between the preterm and full-term birth groups. The transformer model (AUC = 79.20%, sensitivity = 73.67%, specificity = 72.48%, PPV = 28.21%, NPV = 94.95%, and accuracy = 72.61 % in the test dataset) demonstrated superior performance relative to logistic regression (AUC = 77.96% in the test dataset), support vector machine (AUC = 71.70% in the test dataset), and random forest (AUC = 75.09% in the test dataset) approaches.Conclusion::This study highlights the promise of deep learning algorithms, specifically the transformer algorithm, for predicting preterm birth.

Objective::To determine whether deep learning algorithms are suitable for predicting preterm birth.Methods::A retrospective study was conducted at Peking University Third Hospital from January 2018 to June 2023. Birth data were divided into two parts based on the date of delivery: the first part was used for model training and validation, while real world viability was evaluated using the second part. Four machine learning algorithms (logistic regression, random forest, support vector machine, and transformer) were employed to predict preterm birth. Receiver operating characteristic curves were plotted, and the area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated.Results::This research included data on 30,965 births, where 24,770 comprised the first part, and included 3164 (12.77%) in the preterm birth group, with 6195 in the second part, including 795 (12.83%) in the preterm birth group. Significant differences in various factors were observed between the preterm and full-term birth groups. The transformer model (AUC = 79.20%, sensitivity = 73.67%, specificity = 72.48%, PPV = 28.21%, NPV = 94.95%, and accuracy = 72.61 % in the test dataset) demonstrated superior performance relative to logistic regression (AUC = 77.96% in the test dataset), support vector machine (AUC = 71.70% in the test dataset), and random forest (AUC = 75.09% in the test dataset) approaches.Conclusion::This study highlights the promise of deep learning algorithms, specifically the transformer algorithm, for predicting preterm birth.

Objective: To investigate the effect of a modified puncture cannula on prevention of bone cement leakage in percutaneous vertebroplasty (PVP). Methods: From January 2014 to February 2018, 243 patients with single-segmental osteoporotic vertebral fracture were treated with PVP at Department of Orthopedics, Shanghai Ninth People's Hospital. Their clinical data were retrospectively analyzed. Of them, a common puncture cannula was used in 169 cases (control group) and a modified puncture cannula in 74 (modified group). In the control group, there were 41 men and 128 women with an age of 71.6±9.5 years, and the fracture was distributed from T5 to T10 in 7 cases, from T11 to L2 in 132 and from L3 to L5 in 30. In the modified group, there were 20 men and 54 women with an age of 73.6±9.3 years, and the fracture was distributed from T5 to T10 in 3 cases, from T11 to L2 in 63 and from L3 to L5 in 8. The 2 groups were compared in terms of postoperative recovery of vertebral height, reduction in visual analogue scale(VAS) and bone cement leakage. Results: There were no significant differences between the 2 groups in age, gender, distribution of fractured vertebral bodies, compression degree, condition of vertebral posterior wall, or bone cement volume injected (P>0.05). There were no significant differences either between the control and modified groups in the postoperative recovery of vertebral height (7.43%±7.82% versus 6.20%±7.84%) or reduction in VAS score (5.83±0.99 versus 5.81±0.89) (P>0.05). Bone cement leakage occurred in 93 cases (55.0%) in the control group but in 26 cases (35.1%) in the modified group, showing a significant difference (P<0.05). The incidences of bone cement leakage in the paravertebral vessels [13.5% (10/74)], paravertebral soft tissue [9.5%(7/74)] and spinal canal [4.1%(3/74)] in the modified group were all significantly lower than those in the control group [25.4%(43/169), 20.1%(34/169) and 15.4%(26/169)](P<0.05). Conclusion Application of the modified end-to-side puncture cannula is an optional scheme to prevent bone cement leakage in PVP, because it can reduce the incidence of bone cement leakage without compromising postoperative short-term outcomes, especially in the spinal canal, paraspinal vessels and paraspinal soft tissue.