Collaborative bi-aggregation for directed graph embedding.

Directed graph is able to model asymmetric relationships between nodes and research on directed graph embedding is of great significance in downstream graph analysis and inference. Learning source and target embeddings of nodes separately to preserve edge asymmetry has become the dominant approach, but also poses challenge for learning representations of low or even zero in/out degree nodes that are ubiquitous in sparse graphs. In this paper, a collaborative bi-directional aggregation method (COBA) for directed graph embedding is proposed. Firstly, the source and target embeddings of the central node are learned by aggregating from the counterparts of the source and target neighbors, respectively; Secondly, the source/target embeddings of the zero in/out degree central nodes are enhanced by aggregating the counterparts of opposite-directional neighbors (i.e. target/source neighbors); Finally, source and target embeddings of the same node are correlated to achieve collaborative aggregation. Both the feasibility and rationality of the model are theoretically analyzed. Extensive experiments on real-world datasets demonstrate that COBA comprehensively outperforms state-of-the-art methods on multiple tasks and meanwhile validates the effectiveness of proposed aggregation strategies.

Lateral Lingual Foramina and Canals in the Mandible: CBCT Evaluation of 506 Patients in North China.

OBJECTIVE: To assess the prevalence, location, diameter, course and anastomosis of the lateral lingual foramina (LLF) and canals (LLCs) in a northern Chinese population using CBCT. METHODS: CBCT images of 506 patients (181 male and 325 female, mean age 21.03 ± 8.11 years) were collected. The prevalence, location, diameter, length, angle and anastomosis of the LLF and LLCs were assessed. The measurement variables were analysed by sex and age. Statistical analysis was performed using SPSS (v. 25, IBM, Armonk, NY, USA). RESULTS: A total of 461 LLF were detected in 307 (60.7%) subjects, 175 (57.0%) of whom had unilateral LLF and 132 (43.0%) of whom had bilateral LLF, with each lateral having one to four LLF. The majority of LLF (375/461, 81.3%) were located below the premolars, particularly the first premolar. The mean diameter of the LLF was 0.58 ± 0.20 mm. The mean vertical distance from the LLF to the inferior border and the alveolar crest was 6.68 ± 1.43 mm and 23.65 ± 2.89 mm, respectively. In total, 197 LLCs were visible in the cancellous bone and evaluated. The mean length of LLCs was 6.26 ± 1.29 mm, and the mean angle of LLCs was 140.64° ± 17.29°. The overwhelming majority (93.4%) of LLCs communicated with the mandibular incisive canal and the rest connected with the mandibular canal. CONCLUSION: The prevalence of LLCs was high in the northern Chinese population. The presence of LLCs is a significant predictor of communication with the mandibular incisive canal.

[Relationship between cellular immune response and apoptosis of the brain neurons after brain trauma in rats].

OBJECTIVE: To investigate the relation between CD4(+) and CD8(+) T lymphocyte infiltration and apoptosis of the neurons in the local traumatic brain tissue after brain trauma in rats. METHODS: In rat models of brain trauma, the changes in the number of CD4(+) and CD8(+) T lymphocytes and the apoptosis of neurons in the local traumatic brain tissue were observed by immunohistochemistry at different time points after brain trauma. RESULTS: Twenty-four hours after brain trauma, a significant increase in the number of CD4(+) and CD8(+) T lymphocytes occurred in the injured brain tissue, both reaching the highest levels on day 10, at the point of which the number of CD4(+) cells increased by about 15 folds and that of CD8(+) cells by about 20 folds compared with the control groups. The CD4(+) and CD8(+) T lymphocytes both began to decrease 30 days after the injury. A similar pattern of alterations was found in the apoptosis of neurons in the local brain tissue. Correlation analysis demonstrated a close positive correlation between the changes in CD4(+) and CD8(+) lymphocyte numbers and the number apoptotic neurons in the injured brain tissue. CONCLUSIONS: Brain trauma induces obvious increases in CD4(+) and CD8(+) T lymphocytes and enhanced cellular immune response in the injured brain tissue to mediate neuronal apoptosis and further exacerbate the brain tissue injuries.