Abstract
Artificial intelligence is ruling all industrial sectors and has its hand on the medical and healthcare field too. Cough is a symptom of divergent respiratory disorder diseases from a common cold to the current coronavirus disease. Cough is not only extant in humans, but it similarly found to be existing in numerous animals primarily in pigs [1]. Cough is generally a good self-reaction of the body to prevent secretions and its blockages in the upper airway. The frequency, sequence and pattern of the cough reveal the disease along with its severity. Thus, sensing platform and artificial intelligence are used intensively for cough analysis. This chapter is to explore about cough detection and throws light on the various cough detection methodologies, the artificial intelligence algorithms implemented, features involved in cough detection and constraint existent in implementation. In architectural analysis of cough detection; divergent types of the sensors, auxiliary equipment and neural network sustenance instruments deployed are entailed. Cough detection is enacted by voluminous machine and deep learning algorithms using classifiers such as random forest, decision tree, logistic regression, support vector machine, feed forward artificial neural network, convolutional neural network hidden Markov model, multiclass classifier with multilayer perceptron model, and validation is achieved through K-cross validation. The chapter also articulates about the dataset availability of various patterns of cough, the visualizing of sound pattern in frequency and time domain. Further cough is found to have two set of features namely superordinate and subordinate sound features. Superordinate features include Mel-frequency cepstral significant, non-Gaussianity score, Shannon entropy, energy, zero intersection ratio, spectral centroid, spectral bandwidth and spectral roll-off. Subordinate feature covers cough sequence type and duration, bouts occurred in a sequence, cough sequence number in prescribed interval time. The chapter also includes extensive analysis of above feature sets of cough sound. Hence, cough detection using artificial intelligence helps doctors to diagnose early and at ease. At times, it also overcomes the misdiagnosis of the disorders. The chapter also discusses in detail about the various datasets used for cough detection. Finally, includes the constraint of deployment of cough detection that covers the challenges in computational cost, size, budget and ease of deployment with ubiquitous computing.
Access provided by Autonomous University of Puebla. Download chapter PDF
Similar content being viewed by others
Keywords
- Cough detection
- Sensors
- Machine learning
- Deep learning algorithms
- Classifiers
- Superordinate features
- Subordinate features
1 Introduction to Sick Sound—COUGH
One of the most important sickness indicating sounds is cough. Cough occurs due to the presence of disturbance in respiratory track. Based on the presence of liquid, airway passages and lasting time period, the cough can be classified on types, patterns and endurance. The cough along with its acoustic quality sound can be as wet or dry cough types. Wet cough is due to the presence of the disturbance at times due to occurrence of secretions such as mucus and pus. Dry cough is due to inflammations without any fluid secretions [2].
Cough occurring pattern may be obstructive and restrictive based on the nature of airway [3, 4]. In obstructive pattern, airway is widened or narrowed than the normal size. In restrictive pattern, fluid occupied air sacs are present. At time, there is a presence of combined pattern including both the patterns of obstructive and restrictive [5]. Endurance of the cough denotes the time period of existence of the cough as acute, subacute and chronic [6]. Acute coughs are durable for maximum three weeks, and subacute exists more than three weeks up to eight weeks. Chronic cough is serious infections lasting for a longer time period than 8 weeks. In terms of sound signal, the cough possesses two sorts namely airflow and acoustic signal. The airflow signal is plotted as graph between flow in L/sec and time period in second measured in patients mouth. Acoustic cough signal represents sound graph between amplitude and time in seconds generally tested at sternal manubrium. The taxonomy of cough based on types, pattern and endurance is given in Fig. 1 and Table 1.
Cough is a symptom of respiratory medical, non-respiratory medical and environment condition as described in Fig. 2. In case of respiratory medical condition, the reason could be upper respiratory tract infection, lower respiratory tract infect, pneumonia, bronchitis, influenza, asthma, whooping cough, post nasal drip, tuberculosis and corona.
Non-respiratory medical conditions are gastroesophageal reflux, heart failure and tumors. Environment conditions like cooking fumes, smoking, air pollutants cause cough.
3 Features of Cough
Generally, in case of clinical trials, cough has found to consist of many features such as Mel-frequency cepstral significant, explosive cough sounds, cough seconds, cough breaths, cough epochs, cough intensity, cough pattern, zero intersection ratio, spectral centroid, spectral bandwidth, and spectral roll-off as shown in Fig. 4 and Table 2.
Mel-frequency cepstral significant (MFCS) feature enables to recognize characteristics of human auditory [8], and hence, it is used in large scale for cough detection [9,10,11]. A 13 dimensional MFCS obtained through the Mel filter is processed through amplitude and zero crossed coupled first-order and second-order differentiator to enhance to 41 dimensional MFCS [12]. Along with cough frequency; explosiveness of cough sounds, the duration seconds, respiration rate inclusive of least cough and number of repeated cough sounds with less interval (epochs) are also important features [13]. Cough intensity is also taken into account by considering peak and mean energy [14, 15]. Along with energy peak cough flow rate, esophageal pressure and gastric pressure are also thrown light for voluntary, induced and spontaneous cough [16, 17]. Cough patterns also form an important end point for disease identification [18]. Zero intersection ratio indirectly implies the frequency of the cough sound generated. Spectral centroid is used to characterize the spectrum, and further, the extent of spectrum data is provided by spectral bandwidth. Spectral roll-off is used to distinguish between voice-related sounds with nonvoice-related sounds [5].
4 Methods and Algorithms Deployed for Cough Detection
Cough detection is performed by many researchers using two main methods namely instinctive cough segmentation and instinctive cough classification as shown in Fig. 5. In instinctive cough segmentation method, the cough-related audio sounds are riven into fragments to identify the features of interest for cough detection.
In case of instinctive cough classification, classifiers are used to analyze and detect the cough sounds. In Instinctive cough segmentation, the fragmenting is done automatically, whereas in instinctive cough classification, manual segmentation is performed before the classification [19].
Many algorithms are deployed by researchers for cough detection animal houses, cough-related diseases and up to detection of COVID-19 through cough. A cough device for recording the ambulatory cough with help of electromyography (EMG), electrocardiogram (ECG) and microphone was devised [20]. A Holter monitor with EMG and audio signal for based ambulatory cough meter [21].
Semi-automated device utilizes audio and EMG signal for cough detection in children [22]. Automated tagging of EMG and audio signal for objective cough monitoring in infants [23]. Accelerometer portable device with no programmed examination software for nocturnal cough and to study the sleep patterns for children with cough [24]. Lifeshirt [25] is programmed scrutiny of an amalgamation of EMG, ECG and plethysmography for quantification of cough frequency in chronic disorder patients.
A spontaneous instrument, hull instinctive cough summer [26], is devised using linear predictive coding significant and predictable neural Web. Hidden Markov model (HMM) is deployed for MFCS characteristic avulsion [27]. Device performing audio recording and physical totaling of sound of interest was performed in chronic obstructive pulmonary disease [28]. Decision tree-based discriminator is used to segment intended coughs and dialogue to extract rate of recurrence and entropy features [29].
Leicester cough monitor (LCM) [30] built using HMM also depends on audio recordings to pre-fragment probable cough events. Cough monitor [31] to observe chronic cough delivers evaluation of respirational sickness. An amalgamation of artificial neural network (ANN) and support vector machine (SVM) classifiers to monitor cough detection of tuberculosis patients by extracting MFCS [32]. Random forest classification [33] identifies cough segments in audio recordings with capability to reconstruct the cough sounds. Simple threshold method [34] differentiates numerous phases of dry and wet coughs and found that dry coughs posed low energy.
VITALOJAK—a cough observing structure [35] applied semi-automated recognition by means of physical corroboration. An aural introverted structure [36] based on Artificial Neural Network (ANN) for recognition of cough. Neural network [37] intended for cough recognition with sorts such as MFCS, formant rate of recurrence, kurtosis and B score. Support vector machine (SVM) with Gammatone cepstral coefficient (GMCC) feature for cough signal recognition [38].
Differentiation of wet and dry coughs in pediatric patients with logistic regression model (LRM) classifier with features such as MFCS, formant frequencies, kurtosis, zero crossing and B score. First cough classification for pertussis uses three classifiers namely ANN, random forest and K-nearest neighbor algorithm (KNN) with MFCS feature and energy level extraction. The tool significance for automatic cough detection was reported in 2013 [39].
Voluntary cough detection [40] was achieved with fast Fourier transform (FFT) coefficient using KNN. Automatic childhood pneumonia detection [41] uses LRM classifier with interest on features such as MFCS, wavelet and non-Gaussian. Non-contact pediatric ward cough segmentation deployed with ANN [42].
The acquired data exploited with ANN supporting accurate cough duration and cough detection with mutual information from sensors [43]. Mobi Cough [44] amalgamates Gaussian mixture model and universal background model (GMM-UBM) for forecasting cough sounds. Smart watch [45] records sounds and performs conformal prediction analysis to detect cough or sneezing events. HMM grounded on cough revealing [46] using univariate and multivariate time series cough data. Convolutional neural network (CNN) is deployed for cough detection [47] with mathematical model for sound analysis.
Asthma cough sound detection [48] through GMM-UBM deals with features such as MFCS and constant-Q cepstral coefficients. WheezeD [49] perceives respiration stage and installs CNN with acoustic in 2D spectro temporal image for breathless recognition. Power spectral density of cough sounds in different air quality conditions is tested by recognition algorithm deployed with principal component analysis (PCA) and SVM [50].
AI4COVID-19 [51] is artificial intelligence (AI) deployed for COVID-19 initial symptoms recognition with novel multipronged mediator centered risk averse architecture. FluSense [52] is innovative edge computing technology for crowd behavior and influenza indicators—cough. The exhaustive detail explaining the methods and their purpose is tabulated in Table 3, and important algorithms are highlighted in Fig. 6.
5 Instruments Organized in Cough Detection
In deploying, Internet of things six building blocks such as deployment types, sensor time, sensor types, architecture, application types and data requirements are considered [53]. Generally, IoT is used for automation of various devices for specific applications such as light controls and video surveillance [53, 54]. In this section, the instruments utilized for cough detection is discussed.
Multiparametric cough monitoring system [20] monitors cough along with activity and heart rates. Accelerometer, electrodes and microphone are deployed to record ECG, ECG and cough signal, respectively. Cough monitor [21] achieved with deployment of Holter monitor, a computer-based cough processor with selected filters and tape recorders used overnight. Cystic fibrosis-based cough is monitored with Logan Research (LR) 100 cough recording device. In addition, conventional tape recorder is used in first or second day of hospitalization for the duration of chest physiotherapy period.
Infant cough monitor was through LR100 cough monitor, infrared sound and video recorder [23]. Accelerometer portable device with no programmed examination software for nocturnal cough and to study the sleep patterns for children with cough [24]. Lifeshirt [25] was prepared of a A peripatetic cardiorespiratory observing structure, adapted unimaneuvering, touching base microphone, videoing in pseudo organized circumstance. Hull instinctive cough summer [26] is simulated by a computing device with MATLAB 6.1 version with LS_Toolbox version 2.1.1, signal processing toolbox version 5.1, neural network toolbox version 4.0.1 and Voicebox.
HMM for cough signals in audio recordings [27] deployed digital sound recorder and microphone placed in chest of the patient. Portable digital voice recorder with miniature omnidirectional condenser microphone wrapped in plastic foam for distinction of voluntary coughs [29]. LCM was inserted with unrestricted arena necklace microphone and digital sound recorder [30].
Cough sensing deployed with microphone present in mobile phone [33], T-Mobile G1 mobile phone platform was used. VITALOJAK [35] implemented with lapel microphone connected to trained manual cough counter, which compresses the signal with three distinct levels. Karmelsonix system, a commercially obtainable cough counter, used with two microphones namely audio and contact microphone for cough detection [36]. For pediatric patients, couch sideways contactless microphone is implemented for cough recognition [4]. Study of sensor significance [39] exploited the part of ECG, thermal resistor, trunk strap, acceleration sensor, touching base and acoustic microphones in cough finding. Rode NT3 a bed side microphone was placed in two directions for cough detection [41]. Non-contact detection in pediatric ward [42] was deployed with Rode NT3 microphone, preamplifier, A/D converter and mobile pre USB. Sensor-based automatic cough detection system [43] enacted ECG, thermistor, chest belt, accelerometer, oximeter, contact and audio microphones, sensors, analog signal conditioning circuit (front-end), analog-to-digital conversion, communication and storing functional blocks. Mobicough [45] consisted of wireless low-cost microphone connected through Bluetooth to mobile phone. Smart watch [46] for cough detection with low power accelerometer sensor and audio recorder with support of Android app. Asthmatic voluntary sound [48] deployed with computing system supported with MATLAB 2017b and adobe audition CS6. AI4COVID-19 [51] used AI engine for performing cough symptom of COVID-19. FluSense [52] for influenza like illness sensing deployed squat rate microphone, thermal imaging information, Raspberry Pi and Intel Movidius neural engine. The details of instruments utilized for various cough detection work are summarized in Table 4, and important instruments list is highlighted in Fig. 7.
6 Parameters Achieved in Cough Detection Deployment
In cough detection deployment, the main parameters measured as in Fig. 8 are confidence interval, correlation coefficient, positive predictive value or positive foretelling rate, positive rate or optimistic ratio, negative rate, sensitivity or susceptibility, specificity or selectivity, recall, precision, accuracy and F1 score. Confidence interval denotes the sort of value with in which the correct rate lies.
Generally, it is used for comparison between verbal descriptive scores and visual analogue scales [55]. Correlation coefficient gives the strength of relationship with two methods of cough detection mainly video and audio. Positive predictive value indicates the possibility of the focuses with an affirmative screening examination ensuring a sickness. The interlinking connection between positive, negative rate, sensitivity, specificity and predictive value is portrayed in Fig. 9. Supplementary a comprehensive representation about mathematical assessment of parameters is presented in Table 5.
Sensitivity and specificity refer, respectively, to the actual positive cases and actual negative cases predicted appropriately. Positive predictive value and negative predictive value refer the correctness of predicted value is true positive or negative correspondingly.
7 Dataset Details for Cough Detection
For cough detection, numerous data are recorded, created, and existing data bases are used. Data are collected directly from patients [20], and at times, in certain situations, the recordings from subjects were carried out [21]. Mixed sounds from both healthy and ill patients [23], male and female [29] were collected. Sounds from cough due to various illness such as pneumonia, asthma, chronic disorder, tuberculosis and as well as COVID-19 were used [41], [51].
Cough sounds from adults, pediatric [42] and infants [23] are gathered. Already existing YouTube recordings [9], RALE repository [49], environmental sound classification dataset [45], health mode cough dataset [47], sounds from Freesounds.org [45], non-speech audio snippets [52] were utilized. In addition, more group of recordings such as Huawei W1 smart watch recordings [45], sounds from weaners [50] were congregated. Cough sound depending on aerial factors was gathered [50]. Sounds from hospital waiting room in recent for influenza like flu symptom checking were also done [52]. A detailed description of data collection is tabulated in Table 6, and the significant data collection depiction is presented in Fig. 10.
8 Conclusion
In this chapter, an extensive exploration of cough detection methods implemented for various prevailing diseases such as tuberculosis, asthma, wheezing, chronic disorder to newly developed disease coronavirus infection disease in 2019 (COVID-19) is analyzed in a detailed manner. A classic introduction about the famous sick sound is given along with its taxonomy to gather a deep understanding about cough.
Cough-associated medical, non-medical and environment condition are presented. Important features of cough sound such as Mel-frequency cepstral significant, explosive cough sounds, cough seconds, cough breaths, cough epochs, cough intensity, cough pattern, zero intersection ratio, spectral centroid, spectral bandwidth and spectral roll-off are discussed.
An extensive study of classification and segmentation algorithms such as random forest, KNN, principal component analysis, ANN, SVM, HMM, decision tree, discriminator, accelerometer portable device, universal background model, cough monitor, automatic tagging, cough counter, LRM and GMM deployed is highlighted.
A scrutiny of instruments such as accelerometer, thermistor, cough monitoring systems, microphone, sound recorders, cough counters, smart phones, chest belt, computers, mini processors and AI neural engine utilized in cough detection from conventional stage to the latest edge computing stage is investigated.
The important parameters like correlation coefficient, confidence interval, positive predictive value, precision, recall, sensitivity, specificity, true positive rate, true negative rate, accuracy and F1 score that plays a major role in cough detection are analyzed. Further, dataset involved and created in cough detection such as direct collection from subjects, occasion-based recordings, mixed healthy and ill recordings, cough footages, cough segments, smoking subject collections, environmental sound classification dataset, RALE repository, health mode cough sets, sneezing sounds, non-speech audio snippets, aerial factor cough, weaner cough sounds and Huawei cough sounds is also studied.
References
Jans, P., Guarino, M., Costa, A., Aerts, J.M., Berckmans, D.: Evaluation of an algorithm for cough detection in pig houses. IFAC Proc. Volumes 38(1), 92–96 (2005)
Mannino, D.M., Ford, E.S., Redd, S.C.: Obstructive and restrictive lung disease and markers of inflammation: data from the third national health and nutrition examination. Am. J. Med. 114(9), 758–62 (2003). https://doi.org/10.1016/s0002-9343(03)00185-2
Smith, J.A., Ashurst, H.L., Jack, S., Woodcock, A.A., Earis, J.E.: The description of cough sounds by healthcare professionals. Cough. 2(1), (2006). https://doi.org/10.1186/1745-9974-2-1
Swarnkar, V., Abeyratne, U.R., Chang, A.B., Amrulloh, Y.A., Setyati, A., Triasih, R.: Automatic identification of wet and dry cough in pediatric patients with respiratory diseases. Ann. Biomed. Eng. 41(5), 1016–1028 (2013)
Rudraraju, G., Palreddy, S., Mamidgi, B., Sripada, N.R., Sai, Y.P., Vodnala, N.K., Haranath, S.P.: Cough sound analysis and objective correlation with spirometry and clinical diagnosis. Inf. Med. Unlocked. 100319 (2020)
De Blasio, F., Virchow, J.C., Polverino, M., et al.: Cough management: a practical approach. Cough 7, 7 (2011). https://doi.org/10.1186/1745-9974-7-7
Shi, Y., Liu, H., Wang, Y., Cai, M., Xu, W.: Theory and application of audio-based assessment of cough. J. Sensors. (2018)
Tiwari, V.: MFCS and its applications in speaker recognition. Int. J. Emerg. Technol. 1, 19–22 (2010). https://www.mendeley.com/research-papers/MFCS-applications-speaker-recognition/
Parker, D., Picone, J., Harati, A., Lu, S., Jenkyns, M. H., Polgreen, P.M.: Detecting paroxysmal coughing from pertussis cases using voice recognition technology. PloS One. 8, 12 (2013)
Yin, Y., Mo, H.: The identification method of cough signals using Mel-Frequency cepstrum coefficient. Inf. Technol. pp. 86–91 (2012)
Chung, Y., Oh, S., Lee, J., Park, D., Chang, H.-H., Kim, S.: Automatic detection and recognition of pig wasting diseases using sound data in audio surveillance systems. Sensors (Basel) 13, 12929–12942 (2013). https://doi.org/10.3390/s131012929
Du, B., Gao, J., Cao, C.: Objective Recognition of Cough as a Non-invasive biomarker for exposure to cooking oil fumes. Proc. Eng. 205, 3497–3502 (2017)
Shi, Y., Zhang, B., Cai, M., Zhang, X.D.: Numerical simulation of volume-controlled mechanical ventilated respiratory system with 2 different lungs. Int. J. Numer. Methods Biomed. Eng. 33(9), e2852 (2017)
Pavesi, L., Subburaj, S., Porter-Shaw, K.: Application and validation of a computerized cough acquisition system for objective monitoring of acute cough: a meta-analysis. Chest. 120(4), 1121–1128 (2001)
Lee, K.K., Matos, S., Ward, K., Raywood, E., Evans, D.H., Moxham, J., Birring, S.S.: P158 cough sound intensity: the development of a novel measure of cough severity. Thorax 67(Suppl 2), A130–A131 (2012)
Gibson, G.J., Whitelaw, W., Siafakas, N., Supinski, G.S., Fitting, J.W., Bellemare, F. et al.: American Thoracic Society. ATS/ERS statement on respiratory muscle testing. Am. J. Respir. Crit. Care. Med. 166, 518–624 (2002)
Shi, Y., Zhang, B., Cai, M., Xu, W.: Coupling effect of double lungs on a VCV ventilator with automatic secretion clearance function. IEEE/ACM Trans. Comput. Biol. Bioinform. (2017)
Abaza, A.A., Day, J.B., Reynolds, J.S., Mahmoud, A.M., Goldsmith, W.T., McKinney, W.G., Frazer, D.G. et al.: Classification of voluntary cough sound and airflow patterns for detecting abnormal pulmonary function. Cough. 5(1), 8 (2009)
Pramono, R.X.A., Imtiaz, S.A., RodriguezVillegas, E.: A cough-based algorithm for automatic diagnosis of pertussis. PLoS ONE 11(9), e0162128 (2016). https://doi.org/10.1371/journal.pone.0162128
Munyard, P., Busst, C., Logan-Sinclair, R., Bush, A.: A new device for ambulatory cough recording. Pediatr. Pulmonol. 18(3), 178–186 (1994)
Chang, A.B., Newman, R.G., Phelan, P.D., Robertson, C.F.: A new use for an old Holter monitor: an ambulatory cough meter. Eur. Respir. J. 10(7), 1637–1639 (1997)
Hamutcu, R., Francis, J., Karakoc, F., Bush, A.: Objective monitoring of cough in children with cystic fibrosis. Pediatr. Pulmonol. 34(5), 331–335 (2002)
Corrigan, D.L., Paton, J.Y.: Pilot study of objective cough monitoring in infants. Pediatr. Pulmonol. 35(5), 350–357 (2003)
Paul, I.M., Yoder, K.E., Crowell, K.R., Shaffer, M.L., McMillan, H.S., Carlson, L.C., Berlin, C.M. et al.: Effect of dextromethorphan, diphenhydramine, and placebo on nocturnal cough and sleep quality for coughing children and their parents. Pediatrics. 114(1), e85–e90 (2004)
Coyle, M.A., Keenan, D.B., Henderson, L.S., Watkins, M.L., Haumann, B.K., Mayleben, D.W., Wilson, M.G.: Evaluation of an ambulatory system for the quantification of cough frequency in patients with chronic obstructive pulmonary disease. Cough. 1(1), 3 (2005)
Barry, S.J., Dane, A.D., Morice, A.H., Walmsley, A.D.: The automatic recognition and counting of cough. Cough. 2(1), 8 (2006)
Matos, S., Birring, S.S., Pavord, I.D., Evans, H.: Detection of cough signals in continuous audio recordings using hidden Markov models. IEEE Trans. Biomed. Eng. 53(6), 1078–1083 (2006)
Smith, J., Owen, E., Earis, J., Woodcock, A.: Effect of codeine on objective measurement of cough in chronic obstructive pulmonary disease. J. Allergy. Clin. Immunol. 117(4), 831–835 (2006)
Martinek, J., Tatar, M., Javorka, M.: Distinction between voluntary cough sound and speech in volunteers by spectral and complexity analysis. J. Physiol. Pharmacol. 59(6), 433–440 (2008)
Birring, S.S., Fleming, T., Matos, S., Raj, A.A., Evans, D.H., Pavord, I.D.: The Leicester Cough Monitor: preliminary validation of an automated cough detection system in chronic cough. Eur. Respir. J. 31(5), 1013–1018 (2008)
Smith, J.: Monitoring chronic cough: current and future techniques. Expert Rev. Respir. Med. 4(5), 673–683 (2010)
Tracey, B.H., Comina, G., Larson, S., Bravard, M., López, J.W., Gilman, R.H.: Cough detection algorithm for monitoring patient recovery from pulmonary tuberculosis. In: 2011 Annual international conference of the IEEE engineering in medicine and biology society (pp. 6017–6020). IEEE (2011)
Larson, E.C., Lee, T., Liu, S., Rosenfeld, M., Patel, S.N.: Accurate and privacy preserving cough sensing using a low-cost microphone. In: Proceedings of the 13th International Conference on Ubiquitous Computing, pp. 375–384, (2011)
Chatrzarrin, H., Arcelus, A., Goubran, R., Knoefel, F.: Feature extraction for the differentiation of dry and wet cough sounds. In: 2011 IEEE International Symposium on Medical Measurements and Applications (pp. 162–166). IEEE, May 2011
McGuinness, K., Holt, K., Dockry, R., Smith, J.: P159 Validation of the VitaloJAK™ 24 hour ambulatory cough monitor. Thorax 67(Suppl 2), A131–A131 (2012)
Drugman, T., Urbain, J., Bauwens, N., Chessini, R., Aubriot, A. S., Lebecque, P., Dutoit, T.: Audio and contact microphones for cough detection. In: Thirteenth Annual Conference of the International Speech Communication Association (2012)
Swarnkar, V., Abeyratne, U. R., Amrulloh, Y., Hukins, C., Triasih, R., Setyati, A. (2013, July). Neural network based algorithm for automatic identification of cough sounds. In: 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (pp. 1764–1767). IEEE
Liu, J.M., You, M., Li, G.Z., Wang, Z., Xu, X., Qiu, Z., Chen, S et al.: Cough signal recognition with gammatone cepstral coefficients. In: 2013 IEEE China Summit and International Conference on Signal and Information Processing (pp. 160–164). IEEE, July 2013
Drugman, T., Urbain, J., Bauwens, N., Chessini, R., Valderrama, C., Lebecque, P., Dutoit, T.: Objective study of sensor relevance for automatic cough detection. IEEE J. Biomed. Health. Inform. 17(3), 699–707 (2013)
Lücio, C., Teixeira, C., Henriques, J., de Carvalho, P., Paiva, R. P.: Voluntary cough detection by internal sound analysis. In: 2014 7th International Conference on Biomedical Engineering and Informatics (pp. 405–409). IEEE
Kosasih, K., Abeyratne, U.R., Swarnkar, V., Triasih, R.: Wavelet augmented cough analysis for rapid childhood pneumonia diagnosis. IEEE Trans. Biomed. Eng. 62(4), 1185–1194 (2014)
Amrulloh, Y.A., Abeyratne, U.R., Swarnkar, V., Triasih, R., Setyati, A.: Automatic cough segmentation from non-contact sound recordings in pediatric wards. Biomed. Sig. Process. Control. 21, 126–136 (2015)
Mahmoudi, S.A., Possa, P.D.C., Ravet, T., Drugman, T., Chessini, R., Dutoit, T., Valderrama, C.: Sensor-based system for automatic cough detection and classification. In: ICT Innovations Conference (2016)
Pham, C.: MobiCough: real-time cough detection and monitoring using low-cost mobile devices. In: Asian Conference on Intelligent Information and Database Systems, pp. 300–309. Springer, Berlin, Heidelberg (2016)
Nguyen, K.A., Luo, Z.: Cover your cough: detection of respiratory events with confidence using a smartwatch. In: Conformal and Probabilistic Prediction and Applications (pp. 114–131) (2018)
Teyhouee, A., Osgood, N.D.: Cough detection using hidden markov models. In: International Conference on Social Computing, Behavioral-Cultural Modeling and Prediction and Behavior Representation in Modeling and Simulation, pp. 266–276. Springer, Cham, 2019
Kvapilova, L., Boza, V., Dubec, P., Majernik, M., Bogar, J., Jamison, J., Karlin, D.R. et al: Continuous sound collection using smartphones and machine learning to measure cough. Dig. Biomarkers. 3(3), 166–175 (2019)
Hee, H.I., Balamurali, B.T., Karunakaran, A., Herremans, D., Teoh, O.H., Lee, K.P., Chen, J. M. et al.: Development of machine learning for asthmatic and healthy voluntary cough sounds: a proof of concept study. Appl. Sci. 9(14), 2833 (2019)
Chatterjee, S., Rahman, M.M., Nemanti, E., Kuang, J.: WheezeD: Respiration phase based wheeze detection using acoustic data from pulmonary patients under attack. In: 13th EAI International Conference on Pervasive Computing Technologies for Healthcare-Demos and Posters. European Alliance for Innovation (EAI) (2019)
Wang, X., Zhao, X., He, Y., Wang, K.: Cough sound analysis to assess air quality in commercial weaner barns. Comput. Electron. Agric. 160, 8–13 (2019)
Imran, A., Posokhova, I., Qureshi, H.N., Masood, U., Riaz, S., Ali, K., Nabeel, M.: AI4COVID-19: AI enabled preliminary diagnosis for COVID-19 from cough samples via an App. arXiv preprint arXiv:2004.01275 (2020)
Al Hossain, F., Lover, A.A., Corey, G.A., Reich, N.G., Rahman, T.: FluSense: a contactless syndromic surveillance platform for influenza-like illness in hospital waiting areas. Proc. ACM Interact. Mobile. Wearable. Ubiquit. Technol. 4(1), 1–28 (2020)
Raju, P.S., Mahalingam, M., Rajendran, R.A.: Review of intellectual video surveillance through internet of things. In: The Cognitive Approach in Cloud Computing and Internet of Things Technologies for Surveillance Tracking Systems, pp. 141–155. Academic Press (2020)
Sambandam Raju, P., Mahalingam, M., Arumugam Rajendran, R.: Design, implementation and power analysis of pervasive adaptive resourceful smart lighting and alerting devices in developing countries supporting incandescent and led light bulbs. Sensors. 19(9), 2032 (2019)
Leconte, S., Ferrant, D., Dory, V., Degryse, J.: Validated methods of cough assessment: a systematic review of the literature. Respiration. 81(2), 161–174 (2011)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
Preethi, S.R., Revathi, A.R., Murugan, M. (2020). Exploration of Cough Recognition Technologies Grounded on Sensors and Artificial Intelligence. In: Chakraborty, C., Banerjee, A., Garg, L., Rodrigues, J.J.P.C. (eds) Internet of Medical Things for Smart Healthcare. Studies in Big Data, vol 80. Springer, Singapore. https://doi.org/10.1007/978-981-15-8097-0_8
Download citation
DOI: https://doi.org/10.1007/978-981-15-8097-0_8
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-15-8096-3
Online ISBN: 978-981-15-8097-0
eBook Packages: Computer ScienceComputer Science (R0)