Each year, approximately 1 million people suffer a Cerebral Vascular Accident (CVA) in Europe alone, making stroke on the biggest burdens to health and social care. Improving the quality and efficiency of rehabilitation programs is one of the fundamental strategies for a better and cost-effective healthcare system.
ARCADE proposes leveraging interactive and digital technologies to create context-aware workspaces to improve physical rehabilitation practices. The project explores the use of interactive rehabilitation equipment and context- aware environments that support both physiotherapists' and patients' activities. ARCADE research hypothesis is that smart spaces enables therapists to improve work performance while providing personalized, informed, and meaningful physical rehabilitation to stroke patients.
Following a multidisciplinary approach, the proposed research blends computer science, computer engineering, computational biomechanics, physiotherapy, interaction design, and pervasive computing. Based on an active participation of stakeholders (CVA patients, physiotherapists, careers, technologists, and educators), ARCADE combines commodity markerless tracking technologies, interactive objects, and large-scale screens to create smart spaces that recognize and adapt to individual behaviours and support therapists and patients in clinical rehabilitation settings.
ARCADE's key contributions include: extension of commodity depth sensors with accurate biomechanical algorithms; retrofit of rehabilitation equipment to track kinematic performance; and a context-aware environment reactive to physiotherapists' and patients' actions in rehabilitation settings.
ARCADE is aligned with EU societal challenge of health and wellbeing. The project will have an impact on the rehabilitation system, encouraging clinicians to use pervasive computing in clinical settings, creating new ways to better diagnostics and more effective therapy, beyond the use of limited static records.
ARCADE is composed by experts in mixed reality environments and visualization (INESC-ID), experimental biomechanics (IDMEC), and experienced healthcare professionals and educators (CiiEM). Creating more natural forms of interaction in context-aware environments, particularly for rehabilitation, is an area abundantly researched at INESC-ID. From numerous previous projects, they have devised innovative multimodal applications, immersive environments, and several multimodal frameworks that include motion capture, depth cameras, full-body interaction, interactive surfaces, powerwalls, and wearables. Together with the considerable Experimental Biomechanics experience of the personnel at IDMEC, synergies will be conducted to develop and validate novel tracking algorithms. The rehabilitation scenarios will be proposed and validated by the clinical partner (CiiEM). They will also play an important role as an education institution, offering an unique practice-teaching environment that can be leveraged in novel educational applications to train the next generation of physiotherapists, thus enhancing ARCADE's impact.
Although physiotherapists see the potential of intelligent and interactive environment to improve rehabilitation practices, current systems are restricted to static digital records, which do not match the dynamic nature of rehabilitation facilities. In the context of Requirements Analysis, we will apply Human-Computer Interaction Task Analysis techniques to gather information about users, rehabilitation protocols, and practices, in close cooperation with physical medicine clinicians and physiotherapists. This task will identify user scenarios, needs, and activities in the rehabilitation field considering their workflow, as well as how they use kinematic data during the rehabilitation process (diagnostic, treatment, evaluation), and how they can take advantage of context-aware environments.
Analysis of human body kinematics have proven to be very useful for biomedical-related applications, namely in rehabilitation. However, there is a lack of public datasets for rehabilitation movements, especially when considering markerless technologies and the target user group. In this task, the rehabilitation movements identified in (Task 1) will be acquired. Several trials will be captured simultaneously with Kinect One sensors and high-end marker-based motion capture system. Kinematic data from the latter system will be considered as ground truth due to its sub-millimetric tracking precision. Several cameras will be used to track reflective markers placed on the subject's body. Movement from healthy and impaired populations will be acquired to train the machine learning algorithm in (Task 3).The experimental protocol is based on the rigid body reference frames reported in the ISB guidelines for joint coordinate systems.
In this task, we aim at augmenting current markerless sensor technology by proposing a novel skeletal tracking algorithm to accurately estimate the relative orientations of body segments in real-time. The only input are the joints and extremities positions of the stick figure captured by the Kinect sensor.
During rehabilitation sessions, CVA patients use a series of equipment, such as therapy bars of various weights, balls of various sizes, poles, weights, and pedals. This task has the main goal of retrofitting current rehabilitation equipment with monitoring and interactivity capabilities. By knowing which equipment patients are using and how they are using them, ARCADE can propose novel interactive applications to engage patients with exercises and support physiotherapists performance analysis.
In this task, a set of tools for rehabilitation support will be developed, including tools to aid physiotherapists daily tasks, tools to support CVA patients during treatment, and tools to improve therapists' training. We will follow a user-centred design process to fit the tools capabilities to users' needs and context of use.
Rehabilitation facilities are dynamic environments that can greatly benefit from context-aware computing. This task focus on the development of an intelligent environment that monitors contextual variables to estimate users' activities. We will leverage the sensors previously developed in Tasks 3 and 4 to estimate users' information needs in a given situation.
This task focuses on evaluating project results. We will start by defining a usability testing strategy for the various developed prototypes. This protocol will describe tasks and examine criteria from user requirements to define a set of benchmarks that will allow us to measure application performance and user experience. The evaluation will have two main stages: usability studies and field pilots.
This task concerns the main outputs of the project. Through consolidation of tools, synthesis of research results, and exploitation, the outcomes will reach a broader audience outside the project partners.
This task deals with management related activities within the project, including the coordination of the three research teams, logistic and contractual tasks, and the production of reports resulting from partner activities for the duration of the project. This task will produce a progress report in 12th, 24th, and 36th month of the project.
Morada
INESC-ID - IST Taguspark - Room 2N7.17, Av. Prof. Dr. Cavaco Silva, 2744-016 Porto Salvo, Portugal
Telefone
(+351) 21 423 3508
Sandra Sá
sandra.sa@inesc-id.pt