|
Journal of Zhejiang University SCIENCE B
ISSN 1673-1581(Print), 1862-1783(Online), Monthly
2021 Vol.22 No.3 P.233-240
Non-suicidal self-harm is linked to suicidal thoughts in Chinese adolescents with mood disorders: a cross-sectional report
Abstract: Non-suicidal self-injury (NSSI) refers to any intentional, self-inflicted behavior that causes direct damage to body tissues (Kerr et al., 2010), and has emerged as a challenging public health issue worldwide, especially among adolescents. The most common presentations of NSSI include skin-cutting, severe scratching, and burning (Whitlock et al., 2006). In a community-based study conducted in England, 6% of 5506 adolescents reported self-harm in 2014, and 1320 adolescents (74% aged 15-17 years) presented to hospitals due to non-fatal self-harm (Geulayov et al., 2018). In Australia, 8% of adolescents aged less than 20 years reported a history of self-harm (Hiscock et al., 2018). In addition, the prevalence of NSSI has significantly increased in recent years. It was reported by McManus et al. (2019) that the prevalence of self-reported lifetime NSSI increased from 2.4% in 2000 to 6.4% in 2014, with the largest increase in females aged 16–24 years (from 6.5% to 19.7%). Researchers have also proposed an "iceberg" model of self-harm, indicating that the phenomenon of NSSI is often concealed and that most individuals with NSSI do not seek help from others (McMahon et al., 2014; Geulayov et al., 2018).
Key words:
1浙江大学光电科学与工程学院现代光学仪器国家重点实验室,中国杭州市,310027
2之江实验室超级感知中心,中国杭州市,311100
摘要:在内窥镜临床检查中,病灶尺寸精确估计对诊断具有非常重要的意义。本文提出一种基于深度学习的双目内窥镜三维测量方法,可以克服传统双目匹配算法在弱纹理区域鲁棒性较差的缺点。利用三维扫描仪获得的目标三维数据和三维渲染软件仿真的双目相机创建虚拟双目图像数据集,用于训练视差预测模型进行三维测量。实验结果表明,所提方法相比传统双目匹配算法在视差准确度和视差图生成速度上分别提高48.9%和90.5%,能够提供更加准确、可靠的病灶尺寸信息,提高内窥镜诊断效率。
关键词组:
References:
Open peer comments: Debate/Discuss/Question/Opinion
<1>
DOI:
10.1631/jzus.B2000679
CLC number:
Download Full Text:
Downloaded:
1727
Download summary:
<Click Here>Downloaded:
1359Clicked:
3112
Cited:
0
On-line Access:
2024-08-27
Received:
2023-10-17
Revision Accepted:
2024-05-08
Crosschecked:
2021-03-01