Background： Age-related macular degeneration （AMD） is a major cause of irreversible blindness, and awareness of this disease is important in the prevention of blindness. However, lack of public awareness of AMD was...Background： Age-related macular degeneration （AMD） is a major cause of irreversible blindness, and awareness of this disease is important in the prevention of blindness. However, lack of public awareness of AMD was sbown in previous studies, and there was no report of AMD awareness in the Mainland of China. Therefore, the aim of our study was to assess the awareness of AMD and its risk factors among Beijing residents in China. Methods： A cross-sectional, computer-assisted, telephone investigation was conducted to measure the awareness of AMD among Beijing residents. All the contacts of potential respondents were randomly generated by computer. Only those above 18 years of age and willing to participate in the study were included. The questionnaire for the study was moditied from the AMD Alliance International Global Report. Pearson＇s Chi-square test and binary logistic regression analysis were used to identify the factors that affected the knowledge of AMD. Results： Among 385 Beijing residents who agreed to participale, the awareness of AMD was 6.8%, far below than that of cataract and glaucoma. Participants who were above 30 years of age （odds ratio [OR] 6.17, confidence interval [el] 1.44-26.57）, with experience of health-related work （OR 8.11, CI3.25-20.27）, and whose relatives/friends or themselves suffering from AMD （OR 32.18, CI11.29-91.68） had better AMD awareness. Among those familiar with AMD, only 35% of them identified smoking as a risk factor, and only 23.1% of the residents believed that smoking could lead to blindness. Conclusions： The sample of Chinese population had limited knowledge of AMD. Educational programs need to be carried out to raise public awareness of AMD.展开更多
细粒度表情识别任务因其包含更丰富真实的人类情感而备受关注.现有面部表情识别算法通过提取局部关键区域等方式学习更优的图像表征.然而,这些方法忽略了图像数据集内在的结构关系,且没有充分利用标签间的语义关联度以及图像和标签间的...细粒度表情识别任务因其包含更丰富真实的人类情感而备受关注.现有面部表情识别算法通过提取局部关键区域等方式学习更优的图像表征.然而,这些方法忽略了图像数据集内在的结构关系,且没有充分利用标签间的语义关联度以及图像和标签间的相关性,导致所学特征带来的性能提升有限.其次,现有细粒度表情识别方法并未有效利用和挖掘粗细粒度的层级关系,因而限制了模型的识别性能.此外,现有细粒度表情识别算法忽略了由于标注主观性和情感复杂性导致的标签歧义性问题,极大影响了模型的识别性能.针对上述问题,本文提出一种基于关系感知和标签消歧的细粒度面部表情识别算法(fine-grained facial expression recognition algorithm based on Relationship-Awareness and Label Disambiguation,RALD).该算法通过构建层级感知的图像特征增强网络,充分挖掘图像之间、层级标签之间以及图像和标签之间的依赖关系,以获得更具辨别性的图像特征.针对标签歧义性问题,算法设计了基于近邻样本的标签分布学习模块,通过整合邻域信息进行标签消歧,进一步提升模型识别性能.在细粒度表情识别数据集FG-Emotions上算法的准确度达到97.34%,在粗粒度表情识别数据集RAF-DB上比现有主流表情分类方法提高了0.80%~4.55%.展开更多
文摘Background： Age-related macular degeneration （AMD） is a major cause of irreversible blindness, and awareness of this disease is important in the prevention of blindness. However, lack of public awareness of AMD was sbown in previous studies, and there was no report of AMD awareness in the Mainland of China. Therefore, the aim of our study was to assess the awareness of AMD and its risk factors among Beijing residents in China. Methods： A cross-sectional, computer-assisted, telephone investigation was conducted to measure the awareness of AMD among Beijing residents. All the contacts of potential respondents were randomly generated by computer. Only those above 18 years of age and willing to participate in the study were included. The questionnaire for the study was moditied from the AMD Alliance International Global Report. Pearson＇s Chi-square test and binary logistic regression analysis were used to identify the factors that affected the knowledge of AMD. Results： Among 385 Beijing residents who agreed to participale, the awareness of AMD was 6.8%, far below than that of cataract and glaucoma. Participants who were above 30 years of age （odds ratio [OR] 6.17, confidence interval [el] 1.44-26.57）, with experience of health-related work （OR 8.11, CI3.25-20.27）, and whose relatives/friends or themselves suffering from AMD （OR 32.18, CI11.29-91.68） had better AMD awareness. Among those familiar with AMD, only 35% of them identified smoking as a risk factor, and only 23.1% of the residents believed that smoking could lead to blindness. Conclusions： The sample of Chinese population had limited knowledge of AMD. Educational programs need to be carried out to raise public awareness of AMD.
文摘细粒度表情识别任务因其包含更丰富真实的人类情感而备受关注.现有面部表情识别算法通过提取局部关键区域等方式学习更优的图像表征.然而,这些方法忽略了图像数据集内在的结构关系,且没有充分利用标签间的语义关联度以及图像和标签间的相关性,导致所学特征带来的性能提升有限.其次,现有细粒度表情识别方法并未有效利用和挖掘粗细粒度的层级关系,因而限制了模型的识别性能.此外,现有细粒度表情识别算法忽略了由于标注主观性和情感复杂性导致的标签歧义性问题,极大影响了模型的识别性能.针对上述问题,本文提出一种基于关系感知和标签消歧的细粒度面部表情识别算法(fine-grained facial expression recognition algorithm based on Relationship-Awareness and Label Disambiguation,RALD).该算法通过构建层级感知的图像特征增强网络,充分挖掘图像之间、层级标签之间以及图像和标签之间的依赖关系,以获得更具辨别性的图像特征.针对标签歧义性问题,算法设计了基于近邻样本的标签分布学习模块,通过整合邻域信息进行标签消歧,进一步提升模型识别性能.在细粒度表情识别数据集FG-Emotions上算法的准确度达到97.34%,在粗粒度表情识别数据集RAF-DB上比现有主流表情分类方法提高了0.80%~4.55%.
