A ccording to the National Stroke Association (http://www.stroke.org), in the US alone, there are more than 750,000 people experiencing a new or recurrent stroke each year, leading to motor disabilities. There is considerable evidence which directly links functional recovery from a stroke to the duration, frequency, regularity, and intensity of physical (motion and force) interac- tions in a rehabilitation regimen. 1,2 To this end, we wanted to develop the archi- tecture and algorithms for an inexpensive haptic telerehabilitation framework that extends the individualized interactive nature of traditional rehabilitation therapies to patients’ homes. We implemented this in the form of the haptic Virtual Driving Environment (hVDE), with an intended audience of patients with upper limb (UL) dysfunction (secondary to a cerebrovascular event, such as a stroke or physical injury). For a brief background on important considerations as well as the state of work in the field, see the side- bar, “Home-Based Rehabilitation Programs.” Architecture The hVDE, shown in Figure 1 (on page 34), serves as an illustrative example of an individual- ized interactive haptic telerehabilitation frame- work by allowing us to integrate multiple aspects of our research. At the same time, it lets us identi- fy the issues with development, implementation, and deployment of a flexible, reconfigurable, inexpensive, portable telerehabilitation tool, suit- able for setup in patients’ homes and outpatient clinics. Finally, the development of such a reha- bilitation tool in the context of driving, one of the higher activities of daily living, can serve to enhance the motivation and compliance aspects of a therapeutic regimen. However, we note that this hVDE is intended to serve as a network-based tool for assessment and rehabilitation of UL physical motor dysfunction and not as a driving simulator for cognitive assessment. (See for com- parison Systems Technology’s STISIM Drive sim- ulator at http://www.systemstech.com/index. php?pid=22.) Patient interface The hVDE consists of a patient interface (ulti- mately intended to be home based) and a thera- pist interface (ultimately intended to be at a remote central hospital location) that are con- nected through the Internet. The patient interface serves both as the data-acquisition framework as well as the exercise-deployment framework. It consists of force-feedback kinesthetic-interface devices coupled with a variety of exercise scenar- ios implemented in the form of immersive dri- ving activities within a haptics-enabled virtual environment. Kinesthetic interface devices A careful selection of the kinesthetic interface is important because it serves to stimulate the sense of touch and movement, while creating quantitatively measurable and customizable pat- terns of user motions and forces. We focused on selecting and validating the use of low-cost, mass-produced devices (such as the Microsoft Sidewinder force-feedback steering wheel and hobby-rate gyros) with simplified PC interfaces (Universal Serial Bus-based versus explicit data acquisition). Additionally, such commercial, off- the-shelf gaming devices employ standard soft- ware interfaces making them easy to control from a PC environment. For example, we took advantage of the extensive DirectX libraries of force-feedback (FFB) effects (available free at http://www.microsoft.com/windows/directx/ default.aspx). These can be composed, within our specially developed Matlab/DirectX FFB interface 32 1070-986X/06/$20.00 © 2006 IEEE Published by the IEEE Computer Society Individualized Interactive Home-Based Haptic Telerehabilitation Chetan Jadhav, Pravin Nair, and Venkat Krovi State University of New York at Buffalo Haptic User Interfaces for Multimedia Systems We present a haptic telerehabilitation framework for patients with upper- limb dysfunction that is well-suited for deployment in patients’ homes. Specifically, a commercial-off-the- shelf (COTS) haptic force-feedback driving wheel interfaces with a PC to create a haptic Virtual Driving Environment (hVDE). Coupling this framework with parametric exercise/ movement protocols— structured as driving exercises along paths of varying complexity—is the key to the creation of an inexpensive, immersive, and yet individualized personal-movement trainer.