Digital Biomarkers: Using Smartwatch Data For Clinically Relevant Outcomes

Bhavsar, Karan; Singhal, Shubham; Chandel, Vivek; Samal, Avijit; Khandelwal, Sundeep; Ahmed, N.; Ghose, Avik

doi:10.1109/percomworkshops51409.2021.9431000

Cited by 4 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ubiquity of smartphones in modern life [82] has significantly contributed to digital health by providing accessible technology with wireless communication and embedded inertial sensors and camera(s) [83][84][85]. The inclusion of those sensing modalities has enabled smartphones to provide a range of physiological measurement and data entry/collection capabilities, via apps, in an IoT context [86]. Peripheral wearable technologies broaden data collection with notable examples, including popular commercial devices such as Apple Watch, FitBit, and Garmin.…”

Section: Smartphonesmentioning

confidence: 99%

“…Specifically, the use of Bluetooth-enabled wearables could augment more tailored assessments at different locations of the body. For example, a bespoke, Bluetoothenabled, embedded IMU-based smartwatch [86] or smart sock [22] could transmit inertial data from the wrist and/or foot/feet to a smartphone for processing to provide more tailored outcomes. However, with relatively low-energy devices such as smartphones exhibiting low computational power, their use as (big) data processing apparatus may be limited for complex assessments that rely on high computational power, such as the use of artificial intelligence [102].…”

Section: Data Capture: Commercial Ubiquitymentioning

confidence: 99%

“…Rather, the use of a smartphone as the conjugate device (Figure 2i) for data capture via Bluetooth from a peripheral inertial sensing wearable and onward transmission to a higher-powered (i.e., more computationally equipped) cloud instance may be the optimal approach. Subsequent storage and/or further processing of gathered data [22,86,87] would produce specific gait outcomes. For example, an IoTenabled, wearable, foot-mounted IMU [22] has been shown to successfully transmit 1 min bouts of running data to an iOS/Android smartphone before the latter packages the data and communicates with a cloud-based feature extraction and deep learning instance.…”

Section: Data Capture: Commercial Ubiquitymentioning

confidence: 99%

See 2 more Smart Citations

IoT-Enabled Gait Assessment: The Next Step for Habitual Monitoring

Young

Mason

Morris

et al. 2023

Sensors

View full text Add to dashboard Cite

Walking/gait quality is a useful clinical tool to assess general health and is now broadly described as the sixth vital sign. This has been mediated by advances in sensing technology, including instrumented walkways and three-dimensional motion capture. However, it is wearable technology innovation that has spawned the highest growth in instrumented gait assessment due to the capabilities for monitoring within and beyond the laboratory. Specifically, instrumented gait assessment with wearable inertial measurement units (IMUs) has provided more readily deployable devices for use in any environment. Contemporary IMU-based gait assessment research has shown evidence of the robust quantifying of important clinical gait outcomes in, e.g., neurological disorders to gather more insightful habitual data in the home and community, given the relatively low cost and portability of IMUs. The aim of this narrative review is to describe the ongoing research regarding the need to move gait assessment out of bespoke settings into habitual environments and to consider the shortcomings and inefficiencies that are common within the field. Accordingly, we broadly explore how the Internet of Things (IoT) could better enable routine gait assessment beyond bespoke settings. As IMU-based wearables and algorithms mature in their corroboration with alternate technologies, such as computer vision, edge computing, and pose estimation, the role of IoT communication will enable new opportunities for remote gait assessment.

show abstract

Section: Smartphonesmentioning

confidence: 99%

Section: Data Capture: Commercial Ubiquitymentioning

confidence: 99%

Section: Data Capture: Commercial Ubiquitymentioning

confidence: 99%

See 1 more Smart Citation

IoT-Enabled Gait Assessment: The Next Step for Habitual Monitoring

Young

Mason

Morris

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Recently, wearable devices, or wearables, have been widely used with the advancement of smart devices and embedded sensors. The most commonly used wearables are wrist-worn smartwatches which can collect the user’s physiological data while performing the basic functions of smartphones [ 8 , 9 , 10 ]. Wearable sensors can also be attached to various body positions of the torso (chest, waist, and hip) [ 11 , 12 ], lower limbs (legs and feet) [ 13 ], upper limbs (forearm and finger) [ 14 ], or head (scalp and ear) [ 15 , 16 , 17 ], depending on their purpose.…”

Section: Introductionmentioning

confidence: 99%

A Comparative Study on the Influence of Undersampling and Oversampling Techniques for the Classification of Physical Activities Using an Imbalanced Accelerometer Dataset

et al. 2022

View full text Add to dashboard Cite

Accelerometer data collected from wearable devices have recently been used to monitor physical activities (PAs) in daily life. While the intensity of PAs can be distinguished with a cut-off approach, it is important to discriminate different behaviors with similar accelerometry patterns to estimate energy expenditure. We aim to overcome the data imbalance problem that negatively affects machine learning-based PA classification by extracting well-defined features and applying undersampling and oversampling methods. We extracted various temporal, spectral, and nonlinear features from wrist-, hip-, and ankle-worn accelerometer data. Then, the influences of undersampilng and oversampling were compared using various ML and DL approaches. Among various ML and DL models, ensemble methods including random forest (RF) and adaptive boosting (AdaBoost) exhibited great performance in differentiating sedentary behavior (driving) and three walking types (walking on level ground, ascending stairs, and descending stairs) even in a cross-subject paradigm. The undersampling approach, which has a low computational cost, exhibited classification results unbiased to the majority class. In addition, we found that RF could automatically select relevant features for PA classification depending on the sensor location by examining the importance of each node in multiple decision trees (DTs). This study proposes that ensemble learning using well-defined feature sets combined with the undersampling approach is robust for imbalanced datasets in PA classification. This approach will be useful for PA classification in the free-living situation, where data imbalance problems between classes are common.

show abstract