Detection of important features and comparison of datasets for fall detection based on wrist-wearable devices

Kim, Jeong-Kyun; Lee, Kangbok; Hong, Sang Gi

doi:10.1016/j.eswa.2023.121034

Cited by 9 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The influence of the training data set on the model’s performance stands out, attributing this to the variability of movements between the sets. In [ 41 ], an FDS was developed using three sets of data from a wrist sensor, resulting in a fall detection accuracy higher than previous studies with the same sets. The robustness of the model was verified through validations between sets and within each set, highlighting the most influential features for detection.…”

Section: Related Workmentioning

confidence: 99%

“…With an accuracy of 78.51 + 4.06%, the diversity and similarity between the types of falls and ADL impacted the validation of the model. The studies in [ 38 , 39 , 40 , 41 ] are similar to this work, being limited by the absence of training methods that combine data sets to increase the variability of movements and for not considering the influence of data imbalance on the results.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Wrist-Based Fall Detection: Towards Generalization across Datasets

Fula,

Moreno

2024

Sensors

View full text Add to dashboard Cite

Increasing age is related to a decrease in independence of movement and with this decrease comes falls, millions of falls occur every year and the most affected people are the older adults. These falls usually have a big impact on health and independence of the older adults, as well as financial impact on the health systems. Thus, many studies have developed fall detectors from several types of sensors. Previous studies related to the creation of fall detection systems models use only one dataset that usually has a small number of samples. Training and testing machine learning models in this small scope: (i) yield overoptimistic classification rates, (ii) do not generalize to real-life situations and (iii) have very high rate of false positives. Given this, the proposal of this research work is the creation of a new dataset that encompasses data from three different datasets, with more than 1300 fall samples and 28 K negative samples. Our new dataset includes a standard way of adding samples, which allow the future addition of other data sources. We evaluate our dataset by using classic cost-sensitive Machine Leaning methods that deal with class imbalance. For the training and validation of this model, a set of temporal and frequency features were extracted from the raw data of an accelerometer and a gyroscope using a sliding window of 2 s with an overlap of 50%. We study the generalization properties of each dataset, by testing on the other datasets and also the performance of our new dataset. The model showed a good ability to distinguish between activities of daily living and falls, achieving a recall of 90.57%, a specificity of 96.91% and an Area Under the Receiver Operating Characteristic curve (AUC-ROC) value of 98.85% against the combination of three datasets.

show abstract

Section: Related Workmentioning

confidence: 99%