Portable Ultrasound-Based Device for Detecting Older Adults' Sit-to-Stand Transitions in Unsupervised 30-Second Chair-Stand Tests

• Published in: Sensors (Basel, Switzerland) Switzerland), 2020-04, Vol.20 (7), p.1975
• Main Authors: Cobo, Antonio, Villalba-Mora, Elena, Hayn, Dieter, Ferre, Xavier, Pérez-Rodríguez, Rodrigo, Sánchez-Sánchez, Alberto, Bernabé-Espiga, Raquel, Sánchez-Sánchez, Juan-Luis, López-Diez-Picazo, Andrea, Moral, Cristian, Rodriguez-Mañas, Leocadio
• Format: Article
• Language: English
• Subjects: 30-s chair stand test; Adaptive filters; Aged; Aged, 80 and over; Aging; Algorithms; Biosensing Techniques; Bluetooth; Confidence intervals; Correlation analysis; Correlation coefficients; Datasets; Disability; Female; Frailty; Frailty - diagnosis; Frailty - diagnostic imaging; Frailty - physiopathology; frailty syndrome; Humans; Lower Extremity - diagnostic imaging; Lower Extremity - physiopathology; Male; Monitoring, Physiologic; Muscle Strength - physiology; Older people; Outliers (statistics); Portable equipment; Sensors; signal processing; Signal Processing, Computer-Assisted; sit-to-stand; Ultrasonic imaging; Ultrasonography; Ultrasound; Velocity
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s20071975

Lower-limb strength is a marker of functional decline in elders. This work studies the feasibility of using the quasi-periodic nature of the distance between a subjects' back and the chair backrest during a 30-s chair-stand test (CST) to carry out unsupervised measurements based on readings from a low-cost ultrasound sensor. The device comprises an ultrasound sensor, an Arduino UNO board, and a Bluetooth module. Sit-to-stand transitions are identified by filtering the signal with a moving minimum filter and comparing the output to an adaptive threshold. An inter-rater reliability (IRR) study was carried out to validate the device ability to count the same number of valid transitions as the gold-standard manual count. A group of elders (age: mean (m) = 80.79 years old, SD = 5.38; gender: 21 female and seven male) were asked to perform a 30-s CST using the device while a trained nurse manually counted valid transitions. Ultimately, a moving minimum filter was necessary to cancel the effect of outliers, likely produced because older people tend to produce more motion artefacts and, thus, noisier signals. While the intra-class correlation coefficient (ICC) for this study was good (ICC = 0.86, 95% confidence interval (CI) = 0.73, 0.93), it is not yet clear whether the results are sufficient to support clinical decision-making.