基于多示例学习的颈椎健康评分方法
|
摘要
为解决颈椎健康评分及数据标注困难的问题,提出一种基于多值输入多示例学习的颈椎健康状态在线评分算法,仅需在训练阶段对颈椎运动数据长时序列的简单评分标注,即可估计颈椎短时状态的健康评分.首先将多值输入的多示例学习模型划分为多个二值输入的子分类器并分别进行训练,然后使用高斯模型将各个子分类器训练得到的示例分值融合,最后以一种新的包得分机制计算包的分值并完成颈椎健康状态的实时评分.通过包的分值预测准确率计算、示例可视化显示与分析、包得分曲线显示与分析和实时分值评分分析这些定性和定量实验,说明了该算法评估用户颈椎健康的有效性.
This paper introduces an online scoring method for cervical vertebrae health based on multiple instance learning(MIL) of multiple-valued input, in order to assess cervical vertebrae health score and solve the data labeling difficulty. It is only necessary to simply label the long-term sequence of cervical vertebrae motion data during the training phase to estimate the health score of the cervical short-term state. Firstly, the multiple-valued input is divided into sub-classifiers of multiple binary inputs and trained separately. Then use the Gaussian model to fuse the instance scores trained by each sub-classifier. Finally, the bag score is calculated with a new scoring mechanism and the cervical vertebrae health can be assessed in real-time. Qualitative and quantitative experiments include the bag score prediction accuracy, instance visualization analysis, bag score curve analysis and real-time scoring analysis, which illustrate the effectiveness of the algorithm in assessing the health of the cervical vertebrae.
This paper introduces an online scoring method for cervical vertebrae health based on multiple instance learning(MIL) of multiple-valued input, in order to assess cervical vertebrae health score and solve the data labeling difficulty. It is only necessary to simply label the long-term sequence of cervical vertebrae motion data during the training phase to estimate the health score of the cervical short-term state. Firstly, the multiple-valued input is divided into sub-classifiers of multiple binary inputs and trained separately. Then use the Gaussian model to fuse the instance scores trained by each sub-classifier. Finally, the bag score is calculated with a new scoring mechanism and the cervical vertebrae health can be assessed in real-time. Qualitative and quantitative experiments include the bag score prediction accuracy, instance visualization analysis, bag score curve analysis and real-time scoring analysis, which illustrate the effectiveness of the algorithm in assessing the health of the cervical vertebrae.
引文
[1]Ming Z Y,N?rhi M,Siivola J.Neck and shoulder pain related to computer use[J].Pathophysiology,2004,11(1):51-56
[2]Lv F J,Nevatia R.Recognition and segmentation of 3-D human action using hmm and multi-class AdaBoost[C]//Proceedings of European Conference on Computer Vision.Heidelberg:Springer,2006:359-372
[3]Xia L,Chen C C,Aggarwal J K.View invariant human action recognition using histograms of 3D joints[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.Los Alamitos:IEEEComputer Society Press,2012:20-27
[4]Koppula H S,Gupta R,Saxena A.Learning human activities and object affordances from RGB-D videos[J].The International Journal of Robotics Research,2013,32(8):951-970
[5]Sung J,Ponce C,Selman B,et al.Unstructured human activity detection from RGBD images[C]//Proceedings of the IEEEInternational Conference on Robotics and Automation.Los Alamitos:IEEE Computer Society Press,2012:842-849
[6]Ellis C,Masood S Z,Tappen M F,et al.Exploring the trade-off between accuracy and observational latency in action recognition[J].International Journal of Computer Vision,2013,101(3):420-436
[7]Yao A,Gall J,van Gool L.Coupled action recognition and pose estimation from multiple views[J].International Journal of Computer Vision,2012,100(1):16-37
[8]Weng J W,Weng C Q,Yuan J S.Spatio-Temporal Naive-Bayes Nearest-Neighbor(ST-NBNN)for skeleton-based action recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEEComputer Society Press,2017:445-454
[9]Du Y,Wang W,Wang L.Hierarchical recurrent neural network for skeleton based action recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2015:1110-1118
[10]Huang Z W,Wan C D,Probst T,et al.Deep learning on lie groups for skeleton-based action recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2017:1243-1252
[11]Presti L L,La Cascia M.3D skeleton-based human action classification:a survey[J].Pattern Recognition,2016,53:130-147
[12]Ye M,Zhang Q,Wang L,et al.A survey on human motion analysis from depth data[M]//Time-of-Flight and Depth Imaging.Sensors,Algorithms,and Applications.Heidelberg:Springer,2013:149-187
[13]Villanueva M B G,Jonai H,Sotoyama M,et al.Sitting posture and neck and shoulder muscle activities at different screen height settings of the visual display terminal[J].Industrial Health,1997,35(3):330-336
[14]Kamiya K,Kudo M,Nonaka H,et al.Sitting posture analysis by pressure sensors[C]//Proceedings of the 19th International Conference on Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2008:1-4
[15]van Niekerk S M,Louw Q,Vaughan C,et al.Photographic measurement of upper-body sitting posture of high school students:a reliability and validity study[J].BMC Musculoskeletal Disorders,2008,9:113
[16]Xu W Y,Huang M C,Amini N,et al.eCushion:a textile pressure sensor array design and calibration for sitting posture analysis[J].IEEE Sensors Journal,2013,13(10):3926-3934
[17]Kim K B,Song D H,Park H J.Computer vision based automatic extraction and thickness measurement of deep cervical flexor from ultrasonic images[J].Computational and Mathematical Methods in Medicine,2016:Article No.5892051
[18]Dietterich T G,Lathrop R H,Lozano-Pérez T.Solving the multiple instance problem with axis-parallel rectangles[J].Artificial intelligence,1997,89(1/2):31-71
[19]Andrews S,Tsochantaridis I,Hofmann T.Support vector machines for multiple-instance learning[C]//Proceedings of the15th International Conference on Neural Information Processing Systems.Cambridge:MIT Press,2002:577-584
[20]Amores J.Multiple instance classification:review,taxonomy and comparative study[J].Artificial Intelligence,2013,201:81-105
[21]Carbonneau M A,Cheplygina V,Granger E,et al.Multiple instance learning:a survey of problem characteristics and applications[J].Pattern Recognition,2018,77:329-353
[22]Maron O,Lozano-Pérez T.A framework for multiple-instance learning[C]//Proceedings of the 10th International Conference on Neural Information Processing Systems.Cambridge:MITPress,1997:570-576
[23]Zhang Q,Goldman S A.EM-DD:an improved multiple-instance learning technique[C]//Proceedings of the 14th International Conference on Neural Information Processing Systems.Cambridge:MIT Press,2001:1073-1080
[24]Wang X G,Zhu Z T,Yao C,et al.Relaxed multiple-instance SVM with application to object discovery[C]//Proceedings of the IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2015:1224-1232
[25]Carbonneau M A,Granger E,Raymond A J,et al.Robust multiple-instance learning ensembles using random subspace instance selection[J].Pattern Recognition,2016,58:83-99
[26]Viola P A,Platt J C,Zhang C.Multiple instance boosting for object detection[C]//Proceedings of the 18th International Conference on Neural Information Processing Systems.Cambridge:MIT Press,2005:1417-1424
[27]Foulds J,Frank E.A review of multi-instance learning assumptions[J].The Knowledge Engineering Review,2010,25(1):1-25
[28]Zhou Z H,Sun Y Y,Li Y F.Multi-instance learning by treating instances as non-I.I.D samples[C]//Proceedings of the 26th Annual International Conference on Machine Learning.New York:ACM Press,2009:1249-1256
[29]Kim M,de la Torre F.Gaussian processes multiple instance learning[C]//Proceedings of the 27th International Conference on Machine Learning.Madison:OmniPress,2010:535-542
[30]Hau?mann M,Hamprecht F A,Kandemir M.Variational Bayesian multiple instance learning with gaussian processes[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2017:810-819
[31]Wei X S,Wu J X,Zhou Z H.Scalable multi-instance learning[C]//Proceedings of the IEEE International Conference on Data Mining.Los Alamitos:IEEE Computer Society Press,2014:1037-1042
[32]Shi M J,Caesar H,Ferrari V.Weakly supervised object localization using things and stuff transfer[C]//Proceedings of the IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2017:3401-3410
[33]Comaniciu D,Meer P.Mean shift:a robust approach toward feature space analysis[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(5):603-619
[34]Li C L,Zhong F,Qin X Y.Accurate 3D head pose estimation with noisy RGBD images[C]//Proceedings of the 33rd Computer Graphics International.New York:ACM Press,2016:37-40
[2]Lv F J,Nevatia R.Recognition and segmentation of 3-D human action using hmm and multi-class AdaBoost[C]//Proceedings of European Conference on Computer Vision.Heidelberg:Springer,2006:359-372
[3]Xia L,Chen C C,Aggarwal J K.View invariant human action recognition using histograms of 3D joints[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.Los Alamitos:IEEEComputer Society Press,2012:20-27
[4]Koppula H S,Gupta R,Saxena A.Learning human activities and object affordances from RGB-D videos[J].The International Journal of Robotics Research,2013,32(8):951-970
[5]Sung J,Ponce C,Selman B,et al.Unstructured human activity detection from RGBD images[C]//Proceedings of the IEEEInternational Conference on Robotics and Automation.Los Alamitos:IEEE Computer Society Press,2012:842-849
[6]Ellis C,Masood S Z,Tappen M F,et al.Exploring the trade-off between accuracy and observational latency in action recognition[J].International Journal of Computer Vision,2013,101(3):420-436
[7]Yao A,Gall J,van Gool L.Coupled action recognition and pose estimation from multiple views[J].International Journal of Computer Vision,2012,100(1):16-37
[8]Weng J W,Weng C Q,Yuan J S.Spatio-Temporal Naive-Bayes Nearest-Neighbor(ST-NBNN)for skeleton-based action recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEEComputer Society Press,2017:445-454
[9]Du Y,Wang W,Wang L.Hierarchical recurrent neural network for skeleton based action recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2015:1110-1118
[10]Huang Z W,Wan C D,Probst T,et al.Deep learning on lie groups for skeleton-based action recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2017:1243-1252
[11]Presti L L,La Cascia M.3D skeleton-based human action classification:a survey[J].Pattern Recognition,2016,53:130-147
[12]Ye M,Zhang Q,Wang L,et al.A survey on human motion analysis from depth data[M]//Time-of-Flight and Depth Imaging.Sensors,Algorithms,and Applications.Heidelberg:Springer,2013:149-187
[13]Villanueva M B G,Jonai H,Sotoyama M,et al.Sitting posture and neck and shoulder muscle activities at different screen height settings of the visual display terminal[J].Industrial Health,1997,35(3):330-336
[14]Kamiya K,Kudo M,Nonaka H,et al.Sitting posture analysis by pressure sensors[C]//Proceedings of the 19th International Conference on Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2008:1-4
[15]van Niekerk S M,Louw Q,Vaughan C,et al.Photographic measurement of upper-body sitting posture of high school students:a reliability and validity study[J].BMC Musculoskeletal Disorders,2008,9:113
[16]Xu W Y,Huang M C,Amini N,et al.eCushion:a textile pressure sensor array design and calibration for sitting posture analysis[J].IEEE Sensors Journal,2013,13(10):3926-3934
[17]Kim K B,Song D H,Park H J.Computer vision based automatic extraction and thickness measurement of deep cervical flexor from ultrasonic images[J].Computational and Mathematical Methods in Medicine,2016:Article No.5892051
[18]Dietterich T G,Lathrop R H,Lozano-Pérez T.Solving the multiple instance problem with axis-parallel rectangles[J].Artificial intelligence,1997,89(1/2):31-71
[19]Andrews S,Tsochantaridis I,Hofmann T.Support vector machines for multiple-instance learning[C]//Proceedings of the15th International Conference on Neural Information Processing Systems.Cambridge:MIT Press,2002:577-584
[20]Amores J.Multiple instance classification:review,taxonomy and comparative study[J].Artificial Intelligence,2013,201:81-105
[21]Carbonneau M A,Cheplygina V,Granger E,et al.Multiple instance learning:a survey of problem characteristics and applications[J].Pattern Recognition,2018,77:329-353
[22]Maron O,Lozano-Pérez T.A framework for multiple-instance learning[C]//Proceedings of the 10th International Conference on Neural Information Processing Systems.Cambridge:MITPress,1997:570-576
[23]Zhang Q,Goldman S A.EM-DD:an improved multiple-instance learning technique[C]//Proceedings of the 14th International Conference on Neural Information Processing Systems.Cambridge:MIT Press,2001:1073-1080
[24]Wang X G,Zhu Z T,Yao C,et al.Relaxed multiple-instance SVM with application to object discovery[C]//Proceedings of the IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2015:1224-1232
[25]Carbonneau M A,Granger E,Raymond A J,et al.Robust multiple-instance learning ensembles using random subspace instance selection[J].Pattern Recognition,2016,58:83-99
[26]Viola P A,Platt J C,Zhang C.Multiple instance boosting for object detection[C]//Proceedings of the 18th International Conference on Neural Information Processing Systems.Cambridge:MIT Press,2005:1417-1424
[27]Foulds J,Frank E.A review of multi-instance learning assumptions[J].The Knowledge Engineering Review,2010,25(1):1-25
[28]Zhou Z H,Sun Y Y,Li Y F.Multi-instance learning by treating instances as non-I.I.D samples[C]//Proceedings of the 26th Annual International Conference on Machine Learning.New York:ACM Press,2009:1249-1256
[29]Kim M,de la Torre F.Gaussian processes multiple instance learning[C]//Proceedings of the 27th International Conference on Machine Learning.Madison:OmniPress,2010:535-542
[30]Hau?mann M,Hamprecht F A,Kandemir M.Variational Bayesian multiple instance learning with gaussian processes[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Los Alamitos:IEEE Computer Society Press,2017:810-819
[31]Wei X S,Wu J X,Zhou Z H.Scalable multi-instance learning[C]//Proceedings of the IEEE International Conference on Data Mining.Los Alamitos:IEEE Computer Society Press,2014:1037-1042
[32]Shi M J,Caesar H,Ferrari V.Weakly supervised object localization using things and stuff transfer[C]//Proceedings of the IEEE International Conference on Computer Vision.Los Alamitos:IEEE Computer Society Press,2017:3401-3410
[33]Comaniciu D,Meer P.Mean shift:a robust approach toward feature space analysis[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(5):603-619
[34]Li C L,Zhong F,Qin X Y.Accurate 3D head pose estimation with noisy RGBD images[C]//Proceedings of the 33rd Computer Graphics International.New York:ACM Press,2016:37-40