Evaluation of an Automated Swallow-Detection Algorithm Using Visual Biofeedback in Healthy Adults and Head and Neck Cancer Survivors
Mobile health (mHealth) technologies may offer an opportunity to address longstanding clinical challenges, such as access and adherence to swallowing therapy. Mobili-T® is an mHealth device that uses surface electromyography (sEMG) to provide biofeedback on submental muscles activity during exercise. An automated swallow-detection algorithm was developed for Mobili-T®. This study evaluated the performance of the swallow-detection algorithm. Ten healthy participants and 10 head and neck cancer (HNC) patients were fitted with the device. Signal was acquired during regular, effortful, and Mendelsohn maneuver saliva swallows, as well as lip presses, tongue, and head movements. Signals of interest were tagged during data acquisition and used to evaluate algorithm performance. Sensitivity and positive predictive values (PPV) were calculated for each participant. Saliva swallows were compared between HNC and controls in the four sEMG-based parameters used in the algorithm: duration, peak amplitude ratio, median frequency, and 15th percentile of the power spectrum density. In healthy participants, sensitivity and PPV were 92.3 and 83.9%, respectively. In HNC patients, sensitivity was 92.7% and PPV was 72.2%. In saliva swallows, HNC patients had longer event durations (U = 1925.5, p < 0.001), lower median frequency (U = 2674.0, p < 0.001), and lower 15th percentile of the power spectrum density [t(176.9) = 2.07, p < 0.001] than healthy participants. The automated swallow-detection algorithm performed well with healthy participants and retained a high sensitivity, but had lowered PPV with HNC patients. With respect to Mobili-T®, the algorithm will next be evaluated using the mHealth system.
KeywordsDeglutition Deglutition disorders Dysphagia Head and neck cancer Mobile health Surface electromyography Swallow recognition Visual biofeedback
This work was supported by Alberta Cancer Foundation Transformative Program Grant (26355) and Alberta Innovates (AI) Clinician Fellowship (201400350). The team also would like to thank Mr. Fraaz Kamal for developing the custom software used in data acquisition.
Compliance with Ethical Standards
Conflict of interest
Gabriela Constantinescu and Jana Rieger are inventors listed on a patent for the mobile swallowing therapy device. The patent application was made through TEC Edmonton Office, University of Alberta (File Number 2014015. No commercial interest has been shown at this stage).
- 1.Dwivedi RC, Chisholm E, Kanwar N, Komorowski A, Kazi R. Epidemiology, aetiology and natural history of head and neck cancer. In: Staffieri A, Sebastian P, Kapre M, Kazi R, Varghese BT, editors. Essentials of head and neck cancer. Delhi: Byword Books Private Limited; 2012.Google Scholar
- 4.Robbins J, Butler SG, Daniels SK, Gross RD, Langmore S, Lazarus CL, Martin-Harris B, McCabe D, Musson N, Rosenbek JC. Swallowing and dysphagia rehabilitation: translating principles of neural plasticity into clinically oriented evidence. J Speech Lang Hear Res. 2008;51:S276–300.CrossRefPubMedGoogle Scholar
- 5.Wheeler-Hegland K, Ashford J, Frymark T, McCabe D, Mullen R, Musson N, Smith Hammond C, Schooling T. Evidence-based systematic review: oropharyngeal dysphagia behavioral treatments. Part II—impact of dysphagia treatment on normal swallow function. J Rehabil Res Dev. 2009;46:185–94.CrossRefPubMedGoogle Scholar
- 7.WHO. In: WHO, editor. mHealth: new horizons for health through mobile technologies. Geneva: World Health Organization; 2011.Google Scholar
- 9.Steele C: Treating dysphagia with sEMG biofeedback. The ASHA Leader, 2004.Google Scholar
- 11.Davis J, Goadrich M: The relationship between precision–recall and ROC curves. In: Proceedings of the 23rd international conference on machine learning, 2006.Google Scholar
- 12.Eklund JM, Fontana N, Pugh E, McGregor C, Yielder P, James A, Keyzers M, Hahn C, McNamara P: Automated sleep-wake detection in neonates from cerebral function monitor signals. In: IEEE 27th international symposium on computer-based medical systems, New York, USA, 2014, p. 22–7.Google Scholar
- 13.Wang XL, Eklund JM, McGregor C: Parametric power spectrum analysis of ECG signals for obstructive sleep apnoea classification. In: IEEE 27th international symposium on computer-based medical systems, New York, USA, 2014, p. 8–13.Google Scholar
- 14.Zhu W, Zeng N, Wang N: Sensitivity, specificity, accuracy, associated confidence interval and ROC analysis with practical SAS® implementations. In: NESUG proceedings: health care and life sciences, 2010.Google Scholar
- 18.Dysphagia Section OCSGMAoSCiCISoOO, Raber-Durlacher JE, Brennan MT, Verdonck-de Leeuw IM, Gibson RJ, Eilers JG, Waltimo T, Bots CP, Michelet M, Sollecito TP, Rouleau TS, Sewnaik A, Bensadoun RJ, Fliedner MC, Silverman S Jr, Spijkervet FK. Swallowing dysfunction in cancer patients, Support Care Cancer Off J Multinatl Assoc Support Care Cancer. 2012;20:433–43.Google Scholar
- 20.Phinyomark A, Thongpanja S, Hu H, Phukpattaranont P, Limsakul C. The usefulness of mean and median frequencies in electromyography analysis. J Comput. 2012;1:71–80.Google Scholar
- 23.Mangalagiri AS, Razvi MRA. Variations in the anterior belly of diagastric. Int J Health Sci. 2009;3:257–62.Google Scholar
- 24.Altman DG: Some common problems in medical research. In: Practical statistics for medical research. London: Chapman and Hall; 1991.Google Scholar
- 26.Kim HR, Lee SA, Kim K, Leigh JH, Han TR, Oh BM. Submental muscle activity is delayed and shortened during swallowing following stroke. Am Acad Phys Med Rehabil. 2015;7:938–45.Google Scholar