Modeling an Ambient Agent to Support Depression Relapse Prevention

Abstract

One of the challenges for the patients with a history of unipolar depression is to stay healthy throughout their lifetime. In principle, with more prior onset cases, it escalates the risk of the patients to fall into a relapse. In this paper, an ambient agent based model to support patients from relapse is presented. Theories and related works in depression relapse prevention provide a foundation for the formalization of the temporal properties to describe the model. This model was analyzed under several scenarios using simulation and automated verification.

Verification analysis of an agent based model in behaviour change process

“…For instance, if the behaviours of the system of interest are linear, then those linear behaviours must be reflected in the formal specification underlying the model. To address this question, properties of the models with important characteristics will be evaluated as reported in literature [7]. Figure 2 summarizes the verification process that was involved in this model.…”

An integrative ambient agent model for unipolar depression relapse prevention

“…The field of pervasive and wearable technologies has recently witnessed an explosion of new inexpensive and distributed at all scales throughout everyday life which utilize personal-cues and information. These devices can be used in any combination of three main system environments either in a physical world, human-centred environments, or distributed computing environments [4]. Those devices are either loosely-bound to users, or can be more tightlybound to users.…”

“…By integrating the domain model, the ambient agent will be able to reason about the human and environmental processes. It is important to have such capabilities, since an ambient agent should be aware of human behaviours and states [4,32]. Through this mechanism, the agent will use this knowledge to provide related actions related to the predicted state of the human and the environment.…”

“…Examples can be seen in communication and cooperation between agents to manage patient care, information agents that retrieve medical information from the Internet, and multi-agent systems that assist the doctors in the tasks of monitoring and diagnosis [3]. With the advent of wearable devices, and mobile applications, new ways are created for agents to obtain more human-awareness, by human-related information gathered from sensors and based on this enhanced human-awareness will interact and react in a more knowledgeable manner [4]. Such humanaware ambient agents will be able to contribute towards the development of personal care and human wellbeing applications by harnessing vital information from human itself [32].…”

Markov Chain Model to Explain the Dynamics of Human Depression

“…It is probably the best time to visit a psychiatrist. There are several existing models, to monitor a patients' state and behaviour with environmental sensors [13][14][15]. In our present model, we have used a continuous time Markov chain transitional probability model, to describe the depressive behaviour in patients.…”