留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

头部动态场景下非接触式血氧饱和度测量

刘涛 张亚莉

刘涛, 张亚莉. 头部动态场景下非接触式血氧饱和度测量[J]. 中国光学(中英文). doi: 10.37188/CO.2024-0034
引用本文: 刘涛, 张亚莉. 头部动态场景下非接触式血氧饱和度测量[J]. 中国光学(中英文). doi: 10.37188/CO.2024-0034
LIU Tao, ZHANG Ya-li. Non-contact blood oxygenin saturation measurement dynamic head scenarios[J]. Chinese Optics. doi: 10.37188/CO.2024-0034
Citation: LIU Tao, ZHANG Ya-li. Non-contact blood oxygenin saturation measurement dynamic head scenarios[J]. Chinese Optics. doi: 10.37188/CO.2024-0034

头部动态场景下非接触式血氧饱和度测量

doi: 10.37188/CO.2024-0034
基金项目: 国家重点研发计划项目(2018YFC0808),陕西省重点研发项目(2019SF-260)
详细信息
    作者简介:

    刘 涛(1972—),男,陕西西安人,博士,副教授,硕士生导师,2009 年于西安科技大学获得博士学位,主要从事数字信号处理、物联网系统、网络安全方面的研究。E-mail:liutao@xust.edu.cn

    张亚莉(1996—),女,陕西宝鸡人,硕士研究生,2021 年于陕西理工大学获得学士学位,主要从事医学信号及数字信号处理方面的研究。E-mail:3176207132@qq.com

  • 中图分类号: TN911;TP3391

Non-contact blood oxygenin saturation measurement dynamic head scenarios

Funds: Funded by National Key Research and Development Program of China (2018YFC0808), Shaanxi Province Key Research and Development Project (2019SF-260)
More Information
  • 摘要:

    针对现有非接触式血氧饱和度测量方法在头部动态场景下准确性低的问题,提出一种基于改进的自适应噪声完全集合经验模态分解与小波阈值相结合的去噪方法,用于提取高信噪比的脉搏波信号。首先,为解决自适应噪声完全经验模态分解在分解重构早期产生虚假分量和模态混叠的问题,在分解过程中加入高斯白噪声,使其成为改进的自适应噪声完全集合经验模态分解(ICEEMDAN),从而减少模态分量中残余噪声问题。然后,使用ICEEMDAN对红蓝色通道的脉搏波信号进行模态分解,并使用db8小波基函数对符合血氧频谱范围的分量进行3级分解和重构,将重构后的信号用于后续血氧值的计算。最后,将不同头部动态场景下测量的血氧饱和度结果进行实验对比分析,结果表明:不同头部场景下得到的血氧饱和度平均误差为0.73%,相较于其他算法平均误差降低1.93%。本文提出的去噪方法在不同头部场景下具有较好的稳定性,可满足日常血氧饱和度测量的需求。

     

  • 图 1  Hb和HbO2吸收光谱

    Figure 1.  Hb and HbO2 ofabsorption spectra

    图 2  基于ICEEMDAN-WT的血氧饱和度测量整体设计图

    Figure 2.  Overall design of blood oxygen saturation measurement based on ICEEMDAN-WT

    图 3  检测追踪效果图

    Figure 3.  Detection and tracking effect

    图 4  皮肤分割效果图

    Figure 4.  Skin segmentation effect

    图 5  像素平均后B通道和R通道信号

    Figure 5.  Pixels average after B channel and R channel signals

    图 6  去直流后B通道和R通道信号

    Figure 6.  B channel and R channel signal after DC

    图 7  B通道分解后的信号

    Figure 7.  B channel decomposed signal

    图 8  B通道对应的频谱分量

    Figure 8.  Spectral components of each mode of the B channel

    图 9  R通道分解后的信号

    Figure 9.  R channel decomposed signal

    图 10  R通道对应的频谱分量

    Figure 10.  Spectral components of each mode of the R channel

    图 11  B通道重构后的信号

    Figure 11.  Reconstructed signal of channel B

    图 12  R通道重构后的信号

    Figure 12.  Reconstructed signal of channel R

    图 13  头部运动部分帧

    Figure 13.  Head movement part of frame

    图 14  评价指标对比

    Figure 14.  Evaluation index comparison

    图 15  不同方法的MAE对比

    Figure 15.  Comparison of MAE different methods

    图 16  Bland-Altman散点图

    Figure 16.  Bland-Altman scatter plot

    表  1  不同运动场景之下的SpO2结果

    Table  1.   Blood oxygen results under different exercise scenarios

    实验场景Me(%)MAE(%)RMSE(%)
    静态场景0.570.640.86
    说话场景0.690.831.08
    左右晃动0.890.891.26
    上下晃动0.761.041.29
    下载: 导出CSV

    表  2  不同运动场景下算法性能对比 单位%

    Table  2.   Performance comparison of algorithms in different motion scenarios Unit: %

    方法静态场景说话场景上下晃动场景左右晃动场景
    MeMAERMSEMeMAERMSEMeMAERMSEMeMAERMSE
    文献[7]1.552.272.742.231.822.883.282.814.323.562.64.41
    文献[8]0.701.101.301.131.301.721.331.401.931.701.802.23
    文献[10]0.511.121.230.630.931.121.111.361.751.271.441.92
    本文方法0.570.640.860.690.831.080.890.891.260.761.041.29
    下载: 导出CSV
  • [1] STRUYF T, DEEKS J J, DINNES J, et al. Signs and symptoms to determine if a patient presenting in primary care or hospital outpatient settings has COVID-19 disease[J]. Cochrane Database of Systematic Reviews, 2020, 7(7): CD013665.
    [2] MORO E, PRIORI A, BEGHI E, et al. The international European Academy of Neurology survey on neurological symptoms in patients with COVID-19 infection[J]. European Journal of Neurology, 2020, 27(9): 1727-1737. doi: 10.1111/ene.14407
    [3] TAMURA T. Current progress of photoplethysmography and SPO2 for health monitoring[J]. Biomedical Engineering Letters, 2019, 9(1): 21-36. doi: 10.1007/s13534-019-00097-w
    [4] ALHARBI S, HU S, MULVANEY D, et al. Oxygen saturation measurements from green and orange illuminations of multi-wavelength optoelectronic patch sensors[J]. Sensors, 2019, 19(1): 118.
    [5] BAL U. Non-contact estimation of heart rate and oxygen saturation using ambient light[J]. Biomedical Optics Express, 2015, 6(1): 86-97. doi: 10.1364/BOE.6.000086
    [6] 荣猛, 范强, 李凯扬. 基于IPPG非接触式生理参数测量算法的研究[J]. 生物医学工程研究,2018,37(1):27-31,35.

    RONG M, FAN Q, LI K Y. Study on the measurement algorithm of contactless physiological parameter based on imaging photoplenthysmography[J]. Journal of Biomedical Engineering Research, 2018, 37(1): 27-31,35. (in Chinese).
    [7] AL-NAJI A, KHALID G A, MAHDI J F, et al. Non-Contact SpO2 prediction system based on a digital camera[J]. Applied Sciences, 2021, 11(9): 4255. doi: 10.3390/app11094255
    [8] WEI B, WU X P, ZHANG CH, et al. Analysis and improvement of non-contact SpO2 extraction using an RGB webcam[J]. Biomedical Optics Express, 2021, 12(8): 5227-5245. doi: 10.1364/BOE.423508
    [9] 嵇晓强, 刘振瑶, 李炳霖, 等. 面部视频非接触式生理参数感知[J]. 中国光学,2022,15(2):276-285. doi: 10.37188/CO.2021-0157

    JI X Q, LIU ZH Y, LI B L, et al. Non-contact perception of physiological parameters from videos of faces[J]. Chinese Optics, 2022, 15(2): 276-285. doi: 10.37188/CO.2021-0157
    [10] PIRZADA P, MORRISON D, DOHERTY G, et al. Automated remote pulse oximetry system (ARPOS)[J]. Sensors, 2022, 22(13): 4974. doi: 10.3390/s22134974
    [11] HU M, WU X, WANG X H, et al. Contactless blood oxygen estimation from face videos: A multi-model fusion method based on deep learning[J]. Biomedical Signal Processing and Control, 2023, 81: 104487. doi: 10.1016/j.bspc.2022.104487
    [12] KONG L Q, ZHAO Y J, DONG L Q, et al. Non-contact detection of oxygen saturation based on visible light imaging device using ambient light[J]. Optics Express, 2013, 21(15): 17464-17471. doi: 10.1364/OE.21.017464
    [13] ALHARBI S, HU S, MULVANEY D, et al. Oxygen saturation measurements from green and orange illuminations of multi-wavelength optoelectronic patch sensors[J]. Sensors, 2019, 19(1): 118. (查阅网上资料, 本条文献与第4条文献重复, 请确认) .
    [14] VIOLA P, JONES M J, SNOW D. Detecting pedestrians using patterns of motion and appearance[J]. International Journal of Computer Vision, 2005, 63(2): 153-161. doi: 10.1007/s11263-005-6644-8
    [15] MSTAFA R J, ELLEITHY K M. A video steganography algorithm based on Kanade-Lucas-Tomasi tracking algorithm and error correcting codes[J]. Multimedia Tools and Applications, 2016, 75(17): 10311-10333. doi: 10.1007/s11042-015-3060-0
    [16] KHANAM F T Z, AL-NAJI A, CHAHL J. Remote monitoring of vital signs in diverse non-clinical and clinical scenarios using computer vision systems: A review[J]. Applied Sciences, 2019, 9(20): 4474. doi: 10.3390/app9204474
    [17] KONG L Q, ZHAO Y J, DONG L Q, et al. Non-contact detection of oxygen saturation based on visible light imaging device using ambient light[J]. Optics Express, 2013, 21(15): 17464-17471. (查阅网上资料, 本条文献与第12条文献重复, 请确认) .
    [18] NIU X S, HU H, SHAN SH G, et al. VIPL-HR: A multi-modal database for pulse estimation from less-constrained face video[C]. Proceedings of the 14th Asian Conference on Computer Vision, Springer, 2018: 562-576.
    [19] NIU X S, SHAN SH G, HAN H, et al. RhythmNet: End-to-end heart rate estimation from face via spatial-temporal representation[J]. IEEE Transactions on Image Processing, 2020, 29: 2409-2423. doi: 10.1109/TIP.2019.2947204
  • 加载中
图(16) / 表(2)
计量
  • 文章访问数:  63
  • HTML全文浏览量:  28
  • PDF下载量:  6
  • 被引次数: 0
出版历程
  • 收稿日期:  2024-02-07
  • 录用日期:  2024-04-22
  • 网络出版日期:  2024-05-10

目录

    /

    返回文章
    返回

    重要通知

    2024年2月16日科睿唯安通过Blog宣布,2024年将要发布的JCR2023中,229个自然科学和社会科学学科将SCI/SSCI和ESCI期刊一起进行排名!《中国光学(中英文)》作为ESCI期刊将与全球SCI期刊共同排名!