Page 5 - 《中国医疗器械杂志》2026年第1期
P. 5

Chinese Journal of Medical Instrumentation                                         2026年 第50卷 第1期

                                                     医  学  人   工  智  能

              文章编号：1671-7104(2026)01-0001-06

                          基于一维心电二维特征提取的情绪识别研究




             【作     者】 尉思懿 ，安宇坤 ，陈佳雪 ，陈开 ，周平               1
                                1
                                               1
                                                     3
                                       2
                          1 东南大学 生物科学与医学工程学院，南京市，210016
                          2 中国食品药品检定研究院，北京市，102629
                          3 中冶赛迪技术研究中心有限公司，重庆市，400045
             【摘     要】 客观的情绪识别在生理健康、医疗与教育等领域均有着重要的研究意义。从信号获取难易程度、成本与使
                          用者接受度等角度来看，人体的心电信号是实现客观情绪识别的重要生物标志物。针对深度学习网络在一
                          维心电信号时－空特征提取、融合方面的难点，该文首先提出了一种基于小波包分解的1D-2D信号转换方
                          法，将一维心电信号转换为“二维图像”；随后，以此“二维图像”作为输入，以ResNet18作为骨干网，
                          设计了Fusion Block模块，提高了网络的时－空特征提取、融合能力。该文在WESAD与SWELL-KW数据集
                          上进行了情绪识别任务，实验结果表明，与次优方法相比，该文提出的情绪识别方法在准确度与F1分数两
                          个指标上分别高出2.19%与4.48%，为情绪的客观识别提供了技术支持。
             【关   键   词】 情绪识别；1D-2D信号转换；小波包分解；深度学习
             【中图分类号】 R318; TN911.7
             【文献标志码】 A                                                         doi: 10.12455/j.issn.1671-7104.250413
                  Emotion Recognition Based on 2D Feature Extraction from 1D ECG
                                 1
                                           2
                                                                 3
                                                       1
             【   Authors  】 WEI Siyi , AN Yukun , CHEN Jiaxue , CHEN Kai , ZHOU Ping 1
                          1 School of Biological Science and Medical Engineering, Southeast University, Nanjing, 210016
                          2 National Institutes for Food and Drug Control, Beijing, 102629
                          3 CISDI Research & Development Center, Chongqing, 400045
             【  Abstract  】 Objective  emotion  recognition  is  significantly  important  for  some  fields  such  as  physiological  health,
                          healthcare and education. From the perspectives of signal acquisition difficulty, cost and user acceptance,
                          the  electrocardiogram  (ECG)  signals  are  appropriate  biomarkers  for  achieving  objective  emotion
                          recognition.  As  it  is  difficult  for  deep-learning  based  methods  to  extract  and  fuse  the  spatio-temporal
                          features  of  one-dimensional  ECG  signals,  a  1D-2D  signal  transformation  method  based  on  wavelet
                          packet  decomposition  was  proposed  firstly,  which  converted  one-dimensional  ECG  signals  to  "two-
                          dimensional  images".  Subsequently,  the  ResNet18  was  used  as  the  backbone  network  and  the  "two-
                          dimensional images" were used as its input, where a Fusion Block module was designed to improve the
                          network's spatio-temporal feature extraction and fusion capabilities. Finally, extensive experiments were
                          implemented  for  the  emotion  recognition  task  on  the  WESAD  and  SWELL-KW  datasets.  The
                          experimental  results  demonstrated  that  in  comparison  with  the  suboptimal  method,  the  emotion
                          recognition method proposed in this paper improved the performance in both metrics of average accuracy
                          and F1 scores by 2.19 and 4.48 percentage points respectively, which may provide the technical support
                          for objective emotion recognition.
             【Key words】 emotion recognition, 1D-2D signal transformation, wavelet packet decomposition, deep learning



               0    引言                                          长期累积的负面情绪可能引发抑郁症等严重心理问

                                                                题  [1-2] 。在心理健康领域，情绪识别作为情绪障碍
                  情绪对人类的生理、心理状态具有深远影响，
                                                                的早期识别和监测手段，有助于医疗人员评估患者
              收稿日期：2025-06-16                                   情绪状态并及时提供必要的干预和支持                       [3-4] 。目
              基金项目：国家自然科学基金项目（62371121）                         前，常用的情绪识别方法包括基于量表问卷的识别
              作者简介：尉思懿，E-mail: weisiyi@seu.edu.cn
                                                                                                            [6]
                                                                    [5]
              通信作者：周平，E-mail: capzhou@163.com                   方法 ，基于表情、行为等人体表现的识别方法 ，

                                                              1
   1   2   3   4   5   6   7   8   9   10