臺灣博碩士論文加值系統

睡眠呼吸中止症是一種現代社會常見的疾病，患者的上呼吸道會在睡眠阻塞造
成呼吸暫停進而讓患者從睡眠中驚醒大大的降低睡眠品質、日常作息以及影響身體
健康。此種疾病診斷的方式通常得透過穿戴式設備甚至到睡眠醫療中心進行紀錄。而非接觸式的偵測方法往往無法精確分類出受測者做出的動作為呼吸行為或身體動
作。本研究提出了一種基於Wang et al.技術的改良式檢測呼吸異常以及事先診
斷睡眠呼吸中止症的的全自動紅外線影像監控技術。其中介紹了一種藉由呼吸訊
號來進行動作偵測的模型。只需要電腦以及視訊攝影機就能夠透過不須人為監督
的3D適應型模型來學習每個人的呼吸模式並將其睡覺時做出的動作加以分類為異常
呼吸活動或肢體運動。另外系統能夠先偵測環境亮度和對比度進行歸類以調整本身
的靈敏度和判定門檻值。而當系統開始檢測時，能夠計算出觀測目標的主要呼吸
區域以區分出主要觀測目標以及其背景。本技術讓受測者的受測位置不用受到限
制，不論是背對鏡頭、側睡、使用棉被遮擋、或是背景有外物移動都不會干擾
偵測過程。受測者的強弱呼吸、使用胸部或腹部呼吸和拍攝角度也不會影響系統
判定，達到在熟悉的睡眠環境就能夠進行智慧型監測的效果。當系統發現受測者
呼吸行為發生異常時就能自動辨識並做出警示，讓患者家屬不用時時刻刻都需要
花費心力去注意觀測狀況。本研究於44個測試影片的328個事件中達到93%的正確
率。系統能夠從呼吸中止、微弱呼吸、嘴部呼吸、胸部呼吸、腹部呼吸、改
變睡姿、肢體移動、頭部轉動等事件分辨出該事件屬於異常呼吸行為或是身體動
作 。

This research presents a improved real-time automated infrared video monitoring technique based on Wang et al.'s approach~\cite{kopka} for detection of breathing anomalies, and its application in the diagnosis of Obstructive Sleep Apnoea. We introduce a novel motion model to detect subtle, cyclical
breathing signals from video, a new 3D unsupervised self-adaptive breathing template to learn individuals' normal breathing patterns online, and a robust action classification method to recognize abnormal breathing activities and limb movements. Also the system can detect the environment, classify them, and adjust the environment parameter by itself. When the system start detecting, it can calculate the main breathing region to recognize the subject from the background. This technique avoids imposing positional constraints on the patient, allowing patients to sleep on their back or side, with or without facing the camera, fully or partially occluded by the bed clothes and object moving in the background. Moreover, shallow and abdominal breathing patterns do not adversely affect the performance of the method, and it is insensitive to environmental settings such as infrared lighting levels and camera view angles. The experimental results show that the technique achieves high accuracy (93% for 44 data) in recognizing apnoea episodes and body movements and is robust to various occlusion levels, body
poses, body movements (i.e. minor head movement, limb movement, body rotation and slight torso movement), and breathing behavior (e.g. shallow versus heavy breathing, mouth breathing, chest breathing, abdominal breathing, and respiratory arrest).

摘要 ..................................I
Abstract ..................................II
致謝 ..................................III
1 Introduction..................................1
1.1 Motivation.................................. 2
1.2 Aim and Objectives............................. 3
1.3 Contributions................................ 3
1.4 Thesis Organization............................. 5
2 Relatedworks7
3 Methods11
3.1 Motion Detection for Breathing Analysis................. 13
3.2 State Algorithm for Action Segmentation................. 15
3.3 State Transition Rules........................... 16
3.4 Templates for Normal Breathing Activity................. 17
3.5 Region of Breathing Behavior....................... 20
3.6 Action Recognition by Template Matching................ 23
3.7 Simple Action Recognition Model..................... 25
3.8 Smart Parameter Setup........................... 27
4 Results32
4.1 Experiment Environment.......................... 32
4.1.1 Clinical data with high diffusion infrared............. 33
4.1.2 Simulation data with low diffusion infrared............ 33
4.1.3 Simulation data in pointing infrared................ 34
4.1.4 Simulation data in high diffusion infrared............. 34
4.2 Quantitative Results............................ 38
4.3 Analysis................................... 47
5 Conclusion51
5.1 System Limitation............................. 52
5.2 Future Work................................. 52
References ................................52
Appendix................................58

[1] C. W.Wang,A.Hunter,N.Gravill,and S.Matusiewicz, "Unconstrained Video
Monitoring of Breathing Behavior and Application to Diagnosis of Sleep Apnea",
IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING,VOL.61,NO.2,
FEBRUARY2014.

[2] Gibson,G.J.:Obstructive sleep apnoea syndrome:under estimated and under treated. British Medical Bulletin 72 (2004) 49-64

[3] Flemons,W.W.,,Littner,M.R.,,Rowley,J.A.,,Gay,P. et al.:Home Diagnosis
of Sleep Apnea : A Systematic Review of the Literature An Evidence Review the
American Thoracic Society CHEST 124(4) (2003) 1543-1579

[4] Hossain, J.,,Shapiro,C.:The prevalence, cost implications, and management of
sleep disorders : An overview.Sleep Medicine Reviews 6(2) (2002) 85-99

[5] Young,T.,,Palta,M.,,Dempsey,J. et al.:Estimation of the clinically diagnosed
proportion of sleep apnea syndrome in middle aged men and women.Sleep 20
(1997) 705-706

[6] Visi: Visi-3DigitalVideoSystem.http:/www.stowood.co.uk/page26.html

[7] Miyake,Y.,Ishihara,K.,Shinmori,H.,Otsuka,H.,Yamashita,K.,Watanabe,M.,
Nakai,H.,Araki,T.:Enhancements of Non-Restrictive Visual Sensing System for
Reliable Monitoring of Respiration Patterns. JSME International Journal : Series
C, Mechanical systems,machine elementsand manufacturing(1999) 42(3) 737-745
53

[8] Wang,C.W.,,Ahmed,A.,,Hunter,A.:Vision analysis in detecting abnormal
breathing activity in application to diagnosis of obstructive sleep apnoea.Proceedings of Annual International Conference of the IEEE Engineering in Medicineand
Biology Society 1 (2006) 4469-4473

[9] Svetlana,I.,,Mammo,H.Y.,,John,W.A.,,Michael,E.H.etal.:A gated
deep inspiration breath holdradiation the rapy technique using a linear position
transducer. Applied Clinical Medical Physics 6(1) (2005) 61-70

[10] Moody,G.B.,,Mark,R.G.,,Bump,M.A.,,Weinstein,J.S.et al.: Clinical Validation of the ECG-Derived Respiration(EDR)Technique.Computers in Cardiology
13 (1986) 507-510

[11] Storck,K.,,Karlsson,M.,,AskP.,,Loyd,D.: Heat Transfer Evaluation of
the Nasal Thermistor Technique. IEEE Transactions on Biomedical Engineering
43(12) (1996) 1187-1191

[12] Hunsaker,D.H.,,Riffenburgh,R.H.:Snoring signiffcance in patients undergoing
home sleep studies Otolaryngology Head and Neck Surgery 134(2006) 756-760

[13] Cheng,C.-M.,,Hsu,Y.-L.,,Young,C.-M.,,Wu,C.-H.:Development o fa
portable device for tele-monitoring of snoring and OSA Ssymptoms.Tellemediiciine
and e-Healltth 14(1) (2008) 55{68

[14] Ng, A.K.,,Wong,K.Y.,,Tan,C.H.,,Koh,T.S.: Bispectral Analysis of Snore
Signals for Obstructive Sleep Apnea Detection.Conference of the IEEEE MBS
(2007) 6195-6198

[15] Randall,D.P.: Remote Respiratory Monitor.Proceedingsofthe8thAnnual
IEEE Symposium on Computer-Based Medical Systems(1995)204{211

[16] Chekmenev,S.Y.,,Rara,H.,,Farag,A.A.:Non-contact, Wavelet based Measurement of Vital Signs using Thermal Imaging. International Journal of Graphics,
Vision and Image Processing 6 (2005) 25-30

[17] Murthy,R.,,PavlidsI.,,Tsiamyrtzis,P.: Touchless Monitoring of Breathing Function. Proceeding of the 26th Annual International Conference of the IEEE EMBS (2004)1196-1199

[18] Zhu,Z.,,Fei,J.,,Pavlidis,I.:Tracking Human Breath in Infrared Imaging. Proceeding of the 5th IEEE Symposiumon Bioinformatics and Bioengineering(2005)227-231

[19] Mostov,K.,Liptsen,E.: Medical applications of short wave FM radar:Remote monitoring of cardiac and respiratory motion.Medical Physics 37(3) (2010) 1332-1338

[20] Li, C.,,Lin,J.:Random Body Movement Cancellation in Doppler Radar Vital
Sign Detection. IEEE Transactionson Microwave Theory and Techniques 56(12)
(2008) 3143-3152

[21] MbataG.,ChukwukaJ.:Obstructive sleep apnea hypopnea syndrome. Annals of
Medical and Health Sciences Research.2012;2(1):74-77.

[22] Xia, J.,Siochi,R.A.: A real-time respiratory motion monitoring system using
KINECT: proof of concept.Medical Physics 39(5) (2012) 2682-2685

[23]Alnowami,M.,Alnowami,B.,Tahavori,F.,Copland,M.,Wells,K.:A quantitative assessment of using Kinect for Xbox360 for respiratory surface motion tracking.
Proc.SPIE, Medical Imaging 8316 (2012)

[24] Wixson,L.:Detecting salient motion by accumulating directionary-consistent
ow. IEEE Trans.Pattern Anal.Machine Intell 22 (2000) 774-780

[25] Ran,Y.,,Weiss,I.,,Zheng,Q.,,Davis,L.S.:Pedestrian Detection via Periodic
Motion Analysis. International Journal of Computer Vision 71(2) (2007) 143-160

[26] Lipton,A.:Local Application of Optic Flow to Analyse Rigidversus Non-Rigid
Motion. ICCV Workshop on Frame-Rate Vision(1999)

[27] Efros,A.A.,,Berg,A.C.,,Mori,G.,,Malik,J.:Recognizing Actionat a Distance.Proceedings of International Conference on Computer Vision(2003)726-733

[28] Bobick,A.F.,,Davis,J.W.:The recognition of human movement using temporal
templates. IEEET ranson Pattern Analysis and Machine Intelligence 23(3) (2001)257-267

[29] Albu,A.B.,,Beugeling,T.:A Three-Dimensional Spatiotemporal Template for
Interative Human Motion Analysis. Journal of Multimedia 2(4) (2007) 45-54

[30] Valstar,M.,,Pantic,M.,,Patras,I.:Motion History for FacialActionDetection
in Video. Proceedings of IEEE International Conference on Systems, Man and Cybernetics(2004)635-640

[31]Gorelick,L.,,Blank,M.,,Shechtman,E.,,Irani,M.,,Basri,R.:Actionsas Space-Time Shapes. IEEE Transations on Pattern Analysis and MachineIntelligence 29(12) (2007) 2247-2253

[32]Dollar,P.,,Rabaud,V.,,Cottrell,G.,,Belongie,S.:Behavior recognition via sparse spatio-temporal features. Proceedings of Visual Surveillance and Performance Evaluation of Tracking and Surveillance(2005)65-72

[33] Niebles,J.C.,,Wang,H.,,Li,F.-F.:UnsupervisedLearningofHumanAction
Categories Using Spatial-Temporal Words.Internation Journal of Computer Vision
79 (2008) 299-318

[34] Hofmann,T.:Probabilistic Latent Semantic Analysis.Proceedingsofinternational ACMSIGIR conference on research and development in information retrieval
(1999) 50-57

[35] Blei,D.M.,,Ng,A.Y.,,Jordan,M.I.:Latent Dirichl et Al location. Journal of
Machine Learning Research 3 (2003) 993-1022

[36] NareshM.P.:The Epidemiology of Adult Obstructive Sleep Apnea.Proceedings
of the American Thoracic Society,Vol.5,No.2(2008),pp.136-143.

[37] Makarov,A.:Comparison of Background Extraction Based Intrusion Detection
Algorithms. Proceedings of International Conference on Image Processing(1996)
521-524

[38] Sibel,N.T.,,Maybank,S.J.: Fusion of multiple tracking algorithms for robust
people tracking.Proceedings of European Conference on Computer Vision 4 (2002)
373-387

[39] Gavrila01,D.:The visual analysis of human movement : Asurvey. Comput. Vis.
Image Understanding 73(1) (1999) 82-98

[40] Gonzalez,R.C.,,Woods,R.E.: Digital Image Processing.Massachusetts
Addison-Wesley(1992)585

[41]Baudrier,E,,Millon,G.,,Nicolier,F.,,Ruan,S.:ANewSimilarity Measure Using Hausdorff Distance Map. Proceedings of International Conference on Image
Processing(2004)669-672

[42]Huttenlocher,D.P.,,Klanderman,G.A.,,Rucklidge,W.J.:ComparingImages Using Hausdor Distance.IEEE Transactions on Pattern Analysis and Machine
Intelligence 15(9) (1993) 850-863

[43] Ballard,D.H.,,Swain,M.J.:Color indexing. International Journal of Computer
Vision 7(1) (1991) 11-32

[44] Nillius,P.,,Eklundh,J.-O.:Fast Block Matching with NormalizedCrossCorrelation using Walsh Transforms. Report ISRNKTHNAP-0211-SE(2002)

[45] Kohavi,R.,,Provost,F.:Special Issue on Applications of Machine Learning and
the Knowledge Discovery Process.Machine Learning 30 (1998) 271-274