Objective To investigate the effect of long non-coding RNA F19 (lncRNA F19) on secondary brain injury following traumatic brain injury (TBI) in mice. Methods (1) A total of 96 C57BL/6J male wild-type mice were divided into sham group, sham+control lentivirus group, sham+F19 lentivirus group, TBI group, TBI+control lentivirus group and TBI+F19 lentivirus group according to the random number table. Each group consisted of two subgroups of 1 day and 3 days after TBI, with eight mice per subgroup. The expression and silence efficiency of lncRNA F19 were detected. ( 2 ) A total of 96 C57BL/6J male wild-type mice were divided into sham group, TBI+control lentivirus group and TBI + F19 lentivirus group according to the random number table. Each group consisted of two subgroups of 1 day and 3 days after TBI, with 16 mice per subgroup. The effect of lncRNA F19 on neuronal apoptosis after TBI was recorded. The mice TBI model was established using the controlled cortical damage method (CCI). The lncRNA F19 lentivirus or control lentivirus were administrated by intracerebroventricular injection 5 days before injury. The expressions of lncRNA F19 ( 2 -ΔΔct ) were detected by real-time quantitative PCR ( qRT-PCR ) at 1 day and 3 days after injury. The Toll-like receptor 4 (TLR4), B lymphocyte tumor-2 (Bcl-2) and Bcl-2 related protein (Bax) expressions were detected by Western blot. The TUNEL was used to detect apoptosis around the traumatic lesions. Results From the first day after injury, both in the sham operation and TBI groups, the control lentivirus had no effect on the level of lncRAN F19 (P >0. 05). One day after injury, compared with sham +control lentivirus group, the levels of lncRNA F19 in sham + F19 lentivirus group were significantly decreased (0. 07 ± 0. 07:0. 93 ± 0. 17);compared with TBI+control lentivirus group, levels of lncRNA F19 in TBI+F19 lentivirus group were significantly decreased (2. 91 ± 1. 18:0. 52 ± 0. 32) (P<0. 05). There were significantly lower protein levels of TLR4 (0. 51 ± 0. 13:0. 66 ± 0. 15), Bax (0. 45 ± 0. 06:0. 67 ± 0. 16), lower TUNEL-positive neurons ratio [(23. 55 ± 6. 85)％ : (31. 58 ± 7. 52)％], but higher protein levels of Bcl-2 (0. 76 ± 0. 16:0. 47 ± 0. 12) in TBI+F19 lentivirus group compared with the TBI+ control lentivirus group (P <0.05). Three days after injury, compared with sham + control lentivirus group, levels of lncRNA F19 in sham+F19 lentivirus group were significantly decreased (0. 11 ± 0. 09:0. 96 ± 0. 09); compared with TBI+control lentivirus group, levels of lncRNA F19 in TBI+F19 lentivirus group were significantly decreased (0. 54 ± 0. 24:3. 39 ± 0. 90) (P <0. 05). There were significantly lower protein levels of TLR4 (0. 60 ± 0. 20):(0. 85 ± 0. 09)], lower Bax (0. 60 ± 0. 12:0. 88 ±0. 21), lower TUNEL-positive neurons ratio [(29. 10 ± 7. 37)％ :(39. 22 ± 10. 64)％], but higher protein levels of Bcl-2 (0. 66 ± 0. 12:0. 35 ± 0. 16) in TBI+F19 lentivirus group compared with the TBI+control lentivirus group (P<0. 05). Conclusion Inhibition of lncRNA F19 can significantly reduce the TLR4-induced neuronal apoptosis in cortex after TBI in mice and alleviate reduce the secondary brain injury.

Background/Aims: The accuracy of endosonographers in diagnosing gastric subepithelial lesions (SELs) using endoscopic ultrasonography (EUS) is influenced by experience and subjectivity. Artificial intelligence (AI) has achieved remarkable development in this field. This study aimed to develop an AI-based EUS diagnostic model for the diagnosis of SELs, and evaluated its efficacy with external validation. Methods: We developed the EUS-AI model with ResNeSt50 using EUS images from two hospitals to predict the histopathology of the gastric SELs originating from muscularis propria. The diagnostic performance of the model was also validated using EUS images obtained from four other hospitals. Results: A total of 2,057 images from 367 patients (375 SELs) were chosen to build the models, and 914 images from 106 patients (108 SELs) were chosen for external validation. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of the model for differentiating gastrointestinal stromal tumors (GISTs) and non-GISTs in the external validation sets by images were 82.01%, 68.22%, 86.77%, 59.86%, and 78.12%, respectively. The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy in the external validation set by tumors were 83.75%, 71.43%, 89.33%, 60.61%, and 80.56%, respectively. The EUS-AI model showed better performance (especially specificity) than some endosonographers.The model helped improve the sensitivity, specificity, and accuracy of certain endosonographers. Conclusions We developed an EUS-AI model to classify gastric SELs originating from muscularis propria into GISTs and non-GISTs with good accuracy. The model may help improve the diagnostic performance of endosonographers. Further work is required to develop a multi-modal EUS-AI system.