# Neural Network-Based Granular Activity Recognition from Accelerometers: Assessing Generalizability Across Diverse Mobility Profiles

## TL;DR

This paper shows that neural networks can accurately recognize physical activities from accelerometer data and work well across different age groups.

## Contribution

A neural network approach for activity recognition that generalizes well across age groups using wearable sensor data.

## Key findings

- Neural networks outperformed fixed sliding-window methods for activity recognition.
- The method achieved high accuracy across both young and old cohorts.
- Combining datasets improved performance on older age groups.

## Abstract

What are the main findings?
Outperformed fixed sliding-window methods for frame-level activity recognition.Length of data window did not affect the performance.Satisfactory cross-cohort performance.

Outperformed fixed sliding-window methods for frame-level activity recognition.

Length of data window did not affect the performance.

Satisfactory cross-cohort performance.

What are the implications of the main findings?
Granular recognition for capturing transitions.Handling of long and arbitrary length real-world wearable sensor data.

Granular recognition for capturing transitions.

Handling of long and arbitrary length real-world wearable sensor data.

Human activity recognition (HAR) lies at the core of digital healthcare applications that monitor different types of physical activity. Traditional HAR methods often struggle to adapt to variable-length, real-world activity data and to generalise across cohorts (e.g., from young to old cohorts). Thus, the aim of this study was to investigate HAR using wearable sensor data, with a particular focus on cross-cohort evaluation. Each dataset included two accelerometers (right thigh and lower back) sampling at 50 Hz, capturing a range of daily-life activities that were annotated using video recordings from chest-mounted cameras synchronised with the accelerometers. Neural networks were trained on young cohorts’ data and tested on old cohorts’ data. The effects of network architecture, sampling frequency and sensor location on classification performance were investigated. Network performance was evaluated using accuracy, recall, precision, F1-score and confusion matrices. The gated recurrent unit architecture achieved the best performance when trained solely on young cohorts’ data, with weighted F1-score of 0.95 ± 0.05 and 0.93 ± 0.05 for young and old cohorts, respectively, resulting in a highly generalizable method. Classification performance across multiple sampling frequencies was comparable. The thigh-mounted sensor consistently achieved higher performance than the lower back sensor across activities except lying. Furthermore, combining datasets significantly improved performance on the old cohort (weighted F1-score: 0.97 ± 0.02) due to increased variability in the training data. This study highlights the importance of network architecture and dataset composition in HAR and demonstrates the potential of neural networks for robust, real-world activity recognition across age-defined cohorts, specifically between young and old cohorts.