SisFall: A Fall and Movement Dataset

Abstract

Research on fall and movement detection with wearable devices has witnessed promising growth. However, there are few publicly available datasets, all recorded with smartphones, which are insufficient for testing new proposals due to their absence of objective population, lack of performed activities, and limited information. Here, we present a dataset of falls and activities of daily living (ADLs) acquired with a self-developed device composed of two types of accelerometer and one gyroscope. It consists of 19 ADLs and 15 fall types performed by 23 young adults, 15 ADL types performed by 14 healthy and independent participants over 62 years old, and data from one participant of 60 years old that performed all ADLs and falls. These activities were selected based on a survey and a literature analysis. We test the dataset with widely used feature extraction and a simple to implement threshold based classification, achieving up to 96% of accuracy in fall detection. An individual activity analysis demonstrates that most errors coincide in a few number of activities where new approaches could be focused. Finally, validation tests with elderly people significantly reduced the fall detection performance of the tested features. This validates findings of other authors and encourages developing new strategies with this new dataset as the benchmark.

Keywords: SisFall; fall detection; mobile health-care; triaxial accelerometer; wearable devices.

Figure 1

Device used for acquisition. The self-developed embedded device included two accelerometers and a gyroscope. It was fixed to the waist of the participants.

Figure 2

Example of processing and classification. The features are computed after the filtering process of the raw data. (a) ADL D11 gives $C_8$ values below threshold $T_1$ (horizontal red line); (b) Feature $C_8$ crosses the threshold when the fall in activity F05 is detected.

Figure 3

Accuracy obtained in validation after a 10-fold cross-validation without (raw data) and with preprocessing (filtered). Features $C_2$ and $C_8$ achieved 95.0% and 96.1% of accuracy when the filter was applied, respectively. However, not all features improved their performance after filtering.

Figure 4

Maximum value per activity obtained with $C_8$. Most $T_1$ threshold crossings (horizontal red line) are contained in activities D04, D18 and F11.