As one of the main contributions of Miguel’s doctoral thesis, this scientific article recently published in IEEE Journal of Biomedical and Health Informatics presents an mHealth system that seeks to automate the Timed Up and Go (TUG) test using a smartphone and a smartwatch, allowing self-administration of the test in everyday environments. The system is based on the analysis of data from inertial sensors to detect user activities and calculate the duration of the test and its subphases.

The main conclusions of the study can be summarized in the following points:

• The mHealth system demonstrated high accuracy and reliability in measuring the total duration of the TUG test, with results comparable to those obtained by manual reference methods. This suggests that self-administration of the text without expert supervision is feasible.
• The system configuration using the smartwatch as a detection device (C1) obtained superior performance compared to the smartphone-based configuration (C2). This means that smartwatches presented a higher success rate in detecting test duration and its subphases than smartphones.
• Measurement precision for TUG subphases was lower compared to total duration, especially in the C2 configuration. This suggests that smartwatches (C1) are more reliable than smartphones when it comes to recognise finer human activities as those in the TUG test.

How important are the results? Well, for clinicians, the system demonstrates that the use of commercial devices provides equal or better results than the use of specialized ad hoc sensors for TUG testing. On other words, our system makes the TUG test more accessible for a variety of applications and medical cases. For the patient, self-administration of the TUG test opens the possibility of increasing the accessibility and frequency of mobility assessments in everyday settings, such as at home or the workplace, rather than in the doctor’s office.

Read the full story at:

Miguel Matey-Sanz, Alberto González-Pérez, Sven Casteleyn, Carlos Granell. Implementing and evaluating the Timed Up and Go test automation using smartphones and smartwatches. IEEE Journal of Biomedical and Health Informatics, 28(11): 6594-6605, 2024, ISSN 2168-2194.