孤独症可以应用生物特征信号如核磁共振成像(Magnetic Resonance Imaging, MRI)或脑电波(Electroencephalogram, EEG)等对其进行辅助诊断。然而这些诊断方式并不是最优的,因为需要满足的条件有许多,例如专业的医生或专家、在像医院或者实验室等这种特殊的场所等。因此找到一种客观有效适宜普及的方法就尤为重要。所以本研究从孤独症儿童诊断的便捷性出发,提供了一种通过采取孤独症儿童日常生活的视频,应用深度学习的方法进行辅助诊断,希望能为医疗条件不发达地区的孤独症儿童的早期诊断提供帮助。
Jing Li, Zejin Chen, Yihao Zhong, Hak-Keung Lam, Junxia Han, et al., "Appearance-Based Gaze Estimation for ASD Diagnosis," IEEE Transactions on Cybernetics, pp. 1-14, 2022. (SCI,中科院1区)
Code: https://github.com/Chenzejin/Gaze-Estimation-for-ASD-Diagnosis
Jing Li, Zejin Chen, Gongfa Li, Gaoxiang Ouyang*, and Xiaoli Li, “Automatic Classification of ASD Children Using Appearance-based Features from Videos”, Neurocomputing, vol. 470, pp. 40-50, 2022. (SCI,中科院2区)
Jing Li, Yihao Zhong, Junxia Han, et al., "Classifying ASD children with LSTM based on raw videos," Neurocomputing, vol. 390, pp. 226-238, 2020.(SCI,中科院2区)
Jing Li, Yihao Zhong, Gaoxiang Ouyang, et al., "Identification of ASD Children based on Video Data," International Conference on Pattern Recognition, 2018: 367-372.(EI会议)