Detection of non‐suicidal self‐injury based on spatiotemporal features of indoor activities

Abstract

Non-suicide self-injury (NSSI) can be dangerous and difficult for guardians or caregivers to detect in time. NSSI refers to when people hurt themselves even though they have no wish to cause critical or long-lasting hurt. To timely identify and effectively prevent NSSI in order to reduce the suicide rates of patients with a potential suicide risk, the detection of NSSI based on the spatiotemporal features of indoor activities is proposed. Firstly, an NSSI behaviour dataset is provided, and it includes four categories that can be used for scientific research on NSSI evaluation. Secondly, an NSSI detection algorithm based on the spatiotemporal features of indoor activities (NssiDetection) is proposed. NssiDetection calculates the human bounding box by using an object detection model and employs a behaviour detection model to extract the temporal and spatial features of NSSI behaviour. Thirdly, the optimal combination schemes of NssiDetection is investigated by checking its performance with different behaviour detection methods and training strategies. Lastly, a case study is performed by implementing an NSSI behaviour detection prototype system. The prototype system has a recognition accuracy of 84.18% for NSSI actions with new backgrounds, persons, or camera angles. K E Y W O R D Sbehavioural sciences computing, computer vision, convolutional neural nets, feature extraction, object detection | INTRODUCTIONNon-suicidal self-injury (NSSI) [1] refers to the direct and repetitive intentional destruction of one's body in the absence of suicidal intent, and its typical behaviours include cutting, severe scratching, burning, biting, climbing and beating. More than 20% of adolescents and over 5% of adults commit self-injury [2,3]. NSSI is one of the emotional outlets for patients with depression, borderline personality disorder and other common mental diseases [4]. Particularly, patients with depression, accounting for 3.8% of the global population [5], are likely to commit NSSI according to the Global Health Data Exchange Platform.Usually, suicidal intent is absent in NSSI, but NSSI behaviours are meant for self-punishment or signal personal pain to the outside world, especially to important persons. NSSI is associated with potential suicide attempts, and this association implies that NSSI behaviours and their related psychology may already be on the same danger trajectory [6]. In addition, evidence shows that over time, raw emotions, such as anxiety, anger and sadness, along with additional feelings of guilt and shame will continue to exist and may even worsen [7].In conclusion, suicide and NSSI are not the same, but NSSI can be dangerous. Guardians or caregivers may have difficulty in determining when a person is hurting himself evenThis is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

Multi‐innovation gradient‐based iterative identification methods for feedback nonlinear systems by using the decomposition technique

“…The MIGI algorithm and the D‐MIGI algorithm in this article can be extended to polynomial nonlinear models, rational models, switching models, and exponential autoregressive models. The proposed multi‐innovation gradient‐based iterative identification methods for feedback nonlinear systems by using the decomposition technique in this paper can combine other identification idea and methods 108‐117 for develop new identification algorithms of dynamical stochastic linear and nonlinear systems 118‐126 such as chemical process control systems, robot control systems, information processing systems 127‐135 and so on.…”

Hierarchical recursive least squares parameter estimation methods for multiple‐input multiple‐output systems by using the auxiliary models

“…Moreover, the convergence analysis proclaims that the AM-HLS parameter estimates converge to their true values. The algorithm proposed in this article can combine with other identification methods and be applied to MIMO systems with colored noise and other field [99][100][101][102][103][104] such as process control systems [105][106][107][108][109][110][111] and information processing systems.…”

Hierarchical gradient‐ and least‐squares‐based iterative estimation algorithms for input‐nonlinear output‐error systems from measurement information by using the over‐parameterization

“…$$ In summary, Equations (28)–(43) consist of the O‐AM‐HGI algorithm based on the mean value. The O‐AM‐HGI algorithms can combine some mathematical tools 72‐75 and identification approaches 76‐82 to explore parameter estimation methods of different dynamic stochastic systems 83‐89 and can be applied to signal processing and chemical process control. The computing steps of the O‐AM‐HGI algorithm in (28)–(43) are summarized as follows:…”