Deciding the different robot roles for patient cognitive training

Abstract

Alzheimer's Disease (AD) and Mild Cognitive Impairment (MCI) represent a major challenge for health systems within the aging population. New and better instruments will be crucial to assess the disease severity and progression, as well as to improve its treatment, stimulation, and rehabilitation. With the purpose of detecting, assessing and quantifying cognitive impairments like MCI or AD, several methods are employed by clinical experts. Syndrom Kurztest neuropsychological battery (SKT) is a simple and short test to measure cognitive decline as it assesses memory, attention, and related cognitive functions, taking into account the speed of information processing. In this paper, we present a decision system to embed in robot that can set up a productive interaction with a patient, and can be employed by the caregiver to motivate and support him while performing cognitive exercises as SKT. We propose two different interaction loops. First, the robot interacts with the caregiver in order to set up the mental and physical impairments of the patient and indicate a goal for the exercise. This is used to determine the desired robot behavior (human-centric or robot-centric, and preferred interaction modalities). Second, the robot interacts with the patient and adapts its actions to engage and assist him to complete the exercise. Two batches of experiments were conducted, and the results indicated that the robot can take profit of the initial interaction with the caregiver to provide

Cognitive System Framework for Brain-Training Exercise Based on Human-Robot Interaction

“…In our previous framework [4], we already demonstrated that the HRI problem can be effectively modeled using PDDL, and that an off-the-shelf planner can be used then to manage the interaction between the robot and the patient. However, in this framework, the robot can provide assistance only based on the levels set up by the caregiver and it can switch from one level to the other (more assistive) only for wrong moves of the user.…”

“…It is worth mentioning that although this paper is mainly focused on the loop of interaction between the robot and the patient, in the general approach we rely heavily on the caregiver input (what we call the first loop of interaction) [4]. Thus, the presented work has been designed taking into account that caregivers and patients with dementia agree in receiving assistance from a robot [2].…”

“…We have already presented a general HRI framework [4] to tackle this problem that consists of two different interaction loops: first, the caregiver interacts with the robot to set up the initial desired behaviour of the robot (for example, more helpful or more challenging); second, the robot interacts with the patient while he is playing the sorting tokens exercise.…”

Learning Robot Policies Using a High-Level Abstraction Persona-Behaviour Simulator

“…Our use-case [11] is to develop a robotic system easy to configure from the caregiver (first loop of interaction). The robot can administer cognitive exercises based on the Syndrom-Kurztest (SKT), encouraging and motivating the user through speech and gestures (second loop of interaction).…”

“…In this paper, we build on our previous work [11], where a robot is endowed with the abilities to assist dementia patients during cognitive exercises (see Figure 1). In [11], we define two loops of interaction: the first one in which the caregiver sets the physical and mental user impairment and the initial robot behaviour; the second one in which the robot interacts with a patient using a hand-crafted policy and providing him with adaptive levels of assistance based on his performance, while he is playing a cognitive exercise. In this article, we extend this system by: i) providing an easy tool for the caregiver to set up the patient's profile through the concept of Persona (first loop of interaction) and ii) learning the robot initial policy (second loop of interaction).…”