When a Step Is Not a Step! Specificity Analysis of Five Physical Activity Monitors

O’Connell, Sandra; ÓLaighin, Gearóid; Quinlan, Leo R.

doi:10.1371/journal.pone.0169616

Cited by 41 publications

(40 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, this is error prone, especially if we consider the amount of data collected. There are commercial step counting solutions that perform particularly well when the user walks ideally (with the mobile in the palm of his/her hand) but which are susceptible to false positives when the movement or position of the mobile is far from the ideal one [59]. There are also some other solutions, described in medical literature, but which we have not used since they involve sensorized environments that constrain the freedom of movement of the user [60].…”

Section: Ground Truthmentioning

confidence: 99%

Walking Recognition in Mobile Devices

Casado

Rodríguez²,

Iglesias

et al. 2020

Sensors

View full text Add to dashboard Cite

Presently, smartphones are used more and more for purposes that have nothing to do with phone calls or simple data transfers. One example is the recognition of human activity, which is relevant information for many applications in the domains of medical diagnosis, elderly assistance, indoor localization, and navigation. The information captured by the inertial sensors of the phone (accelerometer, gyroscope, and magnetometer) can be analyzed to determine the activity performed by the person who is carrying the device, in particular in the activity of walking. Nevertheless, the development of a standalone application able to detect the walking activity starting only from the data provided by these inertial sensors is a complex task. This complexity lies in the hardware disparity, noise on data, and mostly the many movements that the smartphone can experience and which have nothing to do with the physical displacement of the owner. In this work, we explore and compare several approaches for identifying the walking activity. We categorize them into two main groups: the first one uses features extracted from the inertial data, whereas the second one analyzes the characteristic shape of the time series made up of the sensors readings. Due to the lack of public datasets of inertial data from smartphones for the recognition of human activity under no constraints, we collected data from 77 different people who were not connected to this research. Using this dataset, which we published online, we performed an extensive experimental validation and comparison of our proposals.

show abstract

Section: Ground Truthmentioning

confidence: 99%

Walking Recognition in Mobile Devices

Casado

Rodríguez²,

Iglesias

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Physical and mental health conditions benefit from physical activity. A healthy lifestyle improves both physical and mental health aspects [ 8 ]. In order to monitor the physical activity of a particular user, the counting of steps is one of the most used measures, and therefore it is important that activity-monitoring devices are both specific and sensitive in estimating the actual number of steps walked (discriminating real steps from non-stepping body movements) [ 8 ].…”

Section: Related Workmentioning

confidence: 99%

“…A healthy lifestyle improves both physical and mental health aspects [ 8 ]. In order to monitor the physical activity of a particular user, the counting of steps is one of the most used measures, and therefore it is important that activity-monitoring devices are both specific and sensitive in estimating the actual number of steps walked (discriminating real steps from non-stepping body movements) [ 8 ]. Many off-the-shelf devices register a significant number of false positive steps per minute when executing non-walking activities [ 8 ] and fail to count steps while walking at slow speeds [ 7 ].…”

Section: Related Workmentioning

confidence: 99%

“…In order to monitor the physical activity of a particular user, the counting of steps is one of the most used measures, and therefore it is important that activity-monitoring devices are both specific and sensitive in estimating the actual number of steps walked (discriminating real steps from non-stepping body movements) [ 8 ]. Many off-the-shelf devices register a significant number of false positive steps per minute when executing non-walking activities [ 8 ] and fail to count steps while walking at slow speeds [ 7 ]. Some previous studies have reviewed the accuracy of accelerometers for gait analysis in slow walking, such as [ 6 , 7 , 9 , 10 ], finding that off-the-shelf sensors tend to underestimate strides (significantly in some cases) at slower walking speeds (more specifically for patients of severe medical conditions which affect the gait) [ 11 ].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Detecting Steps Walking at very Low Speeds Combining Outlier Detection, Transition Matrices and Autoencoders from Acceleration Patterns

Muñoz-Organero

Blázquez

2017

Sensors

View full text Add to dashboard Cite

In this paper, we develop and validate a new algorithm to detect steps while walking at speeds between 30 and 40 steps per minute based on the data sensed from a single tri-axial accelerometer. The algorithm concatenates three consecutive phases. First, an outlier detection is performed on the sensed data based on the Mahalanobis distance to pre-detect candidate points in the acceleration time series that may contain a ground contact segment of data while walking. Second, the acceleration segment around the pre-detected point is used to calculate the transition matrix in order to capture the time dependencies. Finally, autoencoders, trained with data segments containing ground contact transition matrices from acceleration series from labeled steps are used to reconstruct the computed transition matrices at each pre-detected point. A similarity index is used to assess if the pre-selected point contains a true step in the 30–40 steps per minute speed range. Our experimental results, based on a database from three different participants performing similar activities to the target one, are able to achieve a recall = 0.88 with precision = 0.50 improving the results when directly applying the autoencoders to acceleration patterns (recall = 0.77 with precision = 0.50).

show abstract

“…Although previous systematic reviews have been informative, several limitations exist within these reviews. First, two of the three systematic reviews (18,22) have included studies with very small sample sizes or compared Fitbits to questionable criterion measures, such as other wearable devices (i.e., accelerometers), instead of ground-truth or "gold-standard" measures of energy expenditure (23), heart rate (24) or steps (25) [e.g., indirect calorimetry, electrocardiogram and steps count video recorded]. Second, all previous reviews have included older versions of the Fitbit that do not use photoplethysmography combined with accelerometry, which are likely to result in more bias than more recent Fitbits (21).…”

Section: Introductionmentioning

confidence: 99%