聚焦于利用多模态数据及机器学习与大语言模型构建情感障碍预测模型，旨在通过这些模型提升情感障碍的筛选效率。本次报告将分享三项研究成果：第一项研究采用连接组预测模型，整合全脑静息态功能连接与白质结构连接数据，准确预测了脆弱人群的自杀风险；第二项研究基于行为特质与经验问卷数据，结合机器学习方法构建抑郁症预测模型并筛选关键影响因素，以区分高、低抑郁风险人群；第三项研究通过大语言模型自动生成大量模拟精神病医生与患者的对话数据，训练出名为EmoScan的新型AI系统，为抑郁症的精准评估提供重要的理论依据和实践工具。

Millisecond pulsars (MSPs) sit at the crossroads of extreme magnetospheric physics and a long-standing Galactic Center mystery: the GeV Excess (GCE), for which MSPs remain the leading astrophysical explanation. In this talk, I present a multi-messenger investigation that links gamma-ray phenomenology to gravitational-wave prospects. First, using 15 years of Fermi-LAT observations, we uncover a robust phase-resolved correlation between gamma-ray luminosity and spectral cutoff energy in our MSP sample. This trend provides new evidence favoring particle acceleration and curvature radiation associated with the Equatorial Current Sheet, and it highlights tensions with standard Outer Gap interpretations. Second, we turn to the GCE debate. If the excess is powered by thousands of unresolved bulge MSPs, can we hear them through continuous gravitational waves? We systematically model the expected continuous-wave emission from such a population under representative deformation scenarios and realistic search considerations. While the predicted signals are below the reach of current detectors, next-generation interferometers, such as the Einstein Telescope and Cosmic Explorer, could probe a detectable fraction of the population, offering an independent multi-messenger test of the MSP origin of the Galactic Center Excess.

教学课件作为课堂教学的核心载体，是知识传递、思维引导与师生互动的重要媒介。优质的课件能够通过清晰的逻辑架构、多模态的内容呈现与互动设计，显著提升学生的课堂参与度与知识内化效率。本次讲座通过对教学PPT基本特点、核心构架、优化要点的探讨，帮助参训教师明确教学PPT的核心设计目标、构建教学PPT基本的设计体系、熟悉教学PPT的实用操作技巧。

教学是教师的教和学生的学所组成的一种人类特有的人才培养活动。教师作为教育质量的核心要素，其教学能力直接影响人才培养成效。 本讲座将从教师如何上好一堂课谈起，以教师的教学能力为主线，和教师一起理清教学设计的思路、探讨教学实施的策略、交流育人的方法和技巧，共同促进本科教学水平和质量的进一步提高。