Prediction of larynx function using multichannel surface EMG classification.

Total laryngectomy (TL) affects critical functions such as swallowing, coughing and speaking. An artificial, bio-engineered larynx (ABL), operated via myoelectric signals, may improve quality of life for TL patients. To evaluate the efficacy of using surface electromyography (sEMG) as a control signal to predict instances of swallowing, coughing and speaking, sEMG was recorded from submental, intercostal and diaphragm muscles. The cohort included TL and control participants. Swallowing, coughing, speaking and movement actions were recorded, and a range of classifiers were investigated for prediction of these actions. Our algorithm achieved F1-scores of 76.0 ± 4.4 % (swallows), 93.8 ± 2.8 % (coughs) and 70.5 ± 5.4 % (speech) for controls, and 67.7 ± 4.4 % (swallows), 71.0 ± 9.1 % (coughs) and 78.0 ± 3.8 % (speech) for TLs, using a random forest (RF) classifier. 75.1 ± 6.9 % of swallows were detected within 500 ms of onset in the controls, and 63.1 ± 6.1 % in TLs. sEMG can be used to predict critical larynx movements, although a viable ABL requires improvements. Results are particularly encouraging as they encompass a TL cohort. An ABL could alleviate many challenges faced by laryngectomees. This study represents a promising step toward realising such a device.

Automatic Audio-Based Classification of Patient Inhaler Use: A Pharmacy Based Study.

Chronic respiratory diseases may be controlled through the delivery of medication to the airways and lungs using an inhaler. However, adherence to correct inhaler technique is poor, which impedes patients from receiving maximum clinical benefit from their medication. In this study, the Inhaler Compliance Assessment device was employed to record audio of patients using a Diskus dry powder inhaler. An algorithm that classifies inhaler sounds (blister, inhalation, interference) was developed to automatically assess patient adherence from these inhaler audio recordings. The presented algorithm employed audio-based signal processing methods and statistical modeling in the form of quadratic discriminant analysis (QDA). A total of 350 audio recordings were obtained from 70 patients. The acquired audio dataset was split evenly for training and testing. A total accuracy of 85.35% was obtained (testing dataset) for this 3-class classification system. A sensitivity of 89.22% and 70% was obtained for inhalation and blister detection respectively. This approach may have significant clinical impact by providing healthcare professionals with an efficient, objective method of monitoring patient adherence to inhaler treatment.