Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Depth-Weighted Detection of Behaviours of Risk in People with Dementia using Cameras

Pratik K. Mishra, Irene Ballester, Andrea Iaboni, Bing Ye, Kristine Newman, Alex Mihailidis, Shehroz S. Khan

TL;DR

The paper tackles automatic detection of dangerous behaviours in people with dementia (PwD) using in-unit cameras. It introduces a depth-weighted loss to achieve depth-invariant reconstruction errors and uses training outliers to set a robust anomaly threshold, integrating these ideas into three depth-variant detectors (depCAE, depCTBAE, depVAE). Evaluated on data from nine participants across three cameras, the approach improves MCC, F1-score, and AUROC, with best AUROC values of $0.852$, $0.81$, and $0.768$ for Cam1, Cam2, and Cam3 respectively, while reducing false alarms. The method is easily adaptable to existing anomaly detectors and shows potential for deployment in real-world care settings, with future work exploring cross-camera generalization, privacy-preserving sensing, and deployment in LTC facilities with varying incident rates.

Abstract

The behavioural and psychological symptoms of dementia, such as agitation and aggression, present a significant health and safety risk in residential care settings. Many care facilities have video cameras in place for digital monitoring of public spaces, which can be leveraged to develop an automated behaviours of risk detection system that can alert the staff to enable timely intervention and prevent the situation from escalating. However, one of the challenges in our previous study was the presence of false alarms due to disparate importance of events based on distance. To address this issue, we proposed a novel depth-weighted loss to enforce equivalent importance to the events happening both near and far from the cameras; thus, helping to reduce false alarms. We further propose to utilize the training outliers to determine the anomaly threshold. The data from nine dementia participants across three cameras in a specialized dementia unit were used for training. The proposed approach obtained the best area under receiver operating characteristic curve performance of 0.852, 0.81 and 0.768, respectively, for the three cameras. Ablation analysis was conducted for the individual components of the proposed approach and effect of frame size and frame rate. The performance of the proposed approach was investigated for cross-camera, participant-specific and sex-specific behaviours of risk detection. The proposed approach performed reasonably well in reducing false alarms. This motivates further research to make the system more suitable for deployment in care facilities.

Depth-Weighted Detection of Behaviours of Risk in People with Dementia using Cameras

TL;DR

The paper tackles automatic detection of dangerous behaviours in people with dementia (PwD) using in-unit cameras. It introduces a depth-weighted loss to achieve depth-invariant reconstruction errors and uses training outliers to set a robust anomaly threshold, integrating these ideas into three depth-variant detectors (depCAE, depCTBAE, depVAE). Evaluated on data from nine participants across three cameras, the approach improves MCC, F1-score, and AUROC, with best AUROC values of , , and for Cam1, Cam2, and Cam3 respectively, while reducing false alarms. The method is easily adaptable to existing anomaly detectors and shows potential for deployment in real-world care settings, with future work exploring cross-camera generalization, privacy-preserving sensing, and deployment in LTC facilities with varying incident rates.

Abstract

The behavioural and psychological symptoms of dementia, such as agitation and aggression, present a significant health and safety risk in residential care settings. Many care facilities have video cameras in place for digital monitoring of public spaces, which can be leveraged to develop an automated behaviours of risk detection system that can alert the staff to enable timely intervention and prevent the situation from escalating. However, one of the challenges in our previous study was the presence of false alarms due to disparate importance of events based on distance. To address this issue, we proposed a novel depth-weighted loss to enforce equivalent importance to the events happening both near and far from the cameras; thus, helping to reduce false alarms. We further propose to utilize the training outliers to determine the anomaly threshold. The data from nine dementia participants across three cameras in a specialized dementia unit were used for training. The proposed approach obtained the best area under receiver operating characteristic curve performance of 0.852, 0.81 and 0.768, respectively, for the three cameras. Ablation analysis was conducted for the individual components of the proposed approach and effect of frame size and frame rate. The performance of the proposed approach was investigated for cross-camera, participant-specific and sex-specific behaviours of risk detection. The proposed approach performed reasonably well in reducing false alarms. This motivates further research to make the system more suitable for deployment in care facilities.
Paper Structure (9 sections, 8 equations, 3 figures, 4 tables)

This paper contains 9 sections, 8 equations, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Methods
  4. Data Collection
  5. Data Annotation
  6. Data Preprocessing
  7. Behaviours of Risk Detection
  8. Results and Discussion
  9. Conclusion and Future Work

Figures (3)

  • Figure 1: Three different installed camera views.
  • Figure 2: (a) Illustration of the geometric relationship between a 3D prism (side length $L$) and the area of its projection in the image plane. (b) Two identical 3D prisms placed at different distances from camera position $C$ produce different projected areas. In this example, if $Z_a > Z_b$, then $A_a < A_b$.
  • Figure 3: depCAE architecture to detect behaviours of risk.