Wearable and Wireless Systems for Healthcare I

The capacity to quantify the movement features of a person undergoing the rehabilitation process enables therapists and clinicians to proactively optimize the therapy strategy. Wearable and wireless systems, such as the smartphone and portable media device, are equipped with accelerometers and gyroscopes that can readily quantify aspects of human movement pertinent to rehabilitation, such as gait and reflex response. The smartphone and portable media device can measure gait and reflex response through their inertial sensors, and the acquired data can be conveyed by wireless transmission to the Internet as an email attachment. This capability enables the experimental site and post-processing resources to be remotely situated. Three phases of the evolution of quantification techniques for the rehabilitation process are observed, which are characterized as a first, second, and third wave. The first wave pertains to the traditional ordinal scale approach used by expert clinicians. The second wave emphasizes the role of quantification systems that are generally constrained to a clinical setting. The third wave envisions the development of Network Centric Therapy through the application of wearable and wireless systems, such as smartphones and portable media devices, for quantifying movement characteristics, such as gait and reflex response. Network Centric Therapy encompasses a quantum leap in rehabilitation capability through Cloud Computing amalgamated with machine learning with patient and therapy team situated remotely anywhere in the world. A summary of each chapter is further presented.

The original technique for quantifying the rehabilitation status of a patient involves the observation by an expert clinician. Based on this expert observation the clinician applies a subjective interpretation to a series of ordinal scale rankings. Examples of scenarios for applying the ordinal scale methodology involve the evaluation of the tendon reflex response and gait. More sophisticated quantification techniques that are derived from the ordinal scale approach pertain to the evaluation of neuro-degenerative diseases, such as Friedreich’s ataxia. Intuitively these ordinal scale techniques are subjective, which causes their reliability to be a subject of controversy. Furthermore, the level of experience of the evaluating clinician can significantly influence the reliability of the evaluation. An alternative solution would be the incorporation of wearable and wireless systems, such smartphones and portable media devices, for quantifying human movement, such as gait and reflex response.

Conventional gait quantification is provided in a highly structured clinical setting. These devices represent a metaphorical second wave encompassing clinically standard quantification techniques. Traditional gait quantification systems, such as force plates, EMG, foot-switches, and motion capture systems are described in the chapter for gait analysis. Their relevance for objectively quantifying the status of a patient’s rehabilitation progress is advocated. Regarding reflex quantification the application of motion capture systems, EMG, and strain/force sensors are covered in the chapter. There are drawbacks of these devices, such as expense, complexity, and limitations to a clinical setting. By contrast, wearable and wireless systems are projected to transcend the capabilities of these traditional quantification systems with expanded autonomy for subject evaluation in the context of Network Centric Therapy.

An inherent aspect of the development of wearable and wireless systems has been the progressive evolution of the inertial measurement unit. Although when preliminarily recommended for quantifying the aspects of human movement, the inertial measurement was not sufficiently developed for application as a wearable and wireless system. With the steady advance from other industries accelerometers became feasible as wearable applications for monitoring activity status and other biomedical and rehabilitation themed scenarios. Eventually wearable accelerometer systems developed from data logger configurations to devices with local wireless connectivity.

With the advent of wireless technology and inertial measurement units, the prevalence of wireless accelerometers is addressed for quantification of gait, reflex response, and reflex latency. Over the course of four generations of research, development, testing, and evaluation the ability to quantify patellar tendon reflex response and latency has been achieved in an accurate, reliable, and reproducible manner. As a transitional phase to the research, development, testing, and evaluation cycle an artificial reflex device was also applied. The central themes to the wireless quantified reflex device are tandem operated wireless accelerometer nodes that are effectively wearable for deriving response and latency and a potential energy impact pendulum for evoking the patellar tendon reflex. The successful application of these wireless accelerometers that are wearable has been further extended toward the quantification of hemiplegic gait, and real-time modification of hemiplegic gait through the quantified feedback of Virtual Proprioception. Other developments regarding the use of wireless accelerometers that are wearable are further addressed.

The smartphone and portable media device are equipped with inertial sensors, such as an accelerometer and gyroscope. With the proper software application they can function as wireless accelerometer and gyroscope platforms. This capability enables the smartphone and portable media device to function as wearable and wireless systems for gait and reflex response. The experimental trial data can be conveyed through wireless connectivity to the Internet as an email attachment for post-processing. The signal data can be further consolidated into a feature set for machine learning classification. Many experimental scenarios pertaining to quantifying the domains of gait and reflex response are presented. The smartphone and portable media device present an insightful perspective of the significant potential of Network Centric Therapy.

Bluetooth wireless enables localized connectivity to a smartphone, portable media device, and tablet. Rather than using these devices as wearable and wireless systems alone, the nature of Bluetooth wireless enables locally situated inertial sensors to be mounted to a subject for quantified evaluation of gait. The smartphone, portable media device, and tablet can then wirelessly transmit the data to a Cloud Computing resource for post-processing. Preliminary demonstration is presented regarding the machine learning classification of gait for Friedreich’s ataxia. A perspective of the application of Bluetooth wireless for reflex quantification is presented. Themes, such as sensor fusion and the Internet of Things, are further discussed. The prevalence of Bluetooth wireless further establishes the realization of Network Centric Therapy.

Wearable and wireless systems equipped with the ability to mutually record the accelerometer and gyroscope signal can be applied to sensor fusion. Sensor fusion can provide the location of the inertial sensor with trajectory information, such as displacement, velocity, and acceleration as a function of time. In order to achieve the results of sensor fusion multiple subjects must be applied, such as the use of quaternion mathematics and orientation filtering. A traditional orientation filter is the Kalman filter; however, the gradient descent orientation filter offers a more computationally robust alternative that is suitable for wearable and wireless systems. The result information provided by sensor fusion is particularly useful for the assessment of gait trajectory. Sensor fusion is anticipated to enhance Network Centric Therapy with improved visualization of patient status.

Over the span of the past decade machine learning has been applied to distinguishing between disparate health status scenarios with considerable classification accuracy. Recent examples pertain to notable classification accuracy with regards to gait and reflex response disparity, especially in the context of a hemiplegic affected leg and unaffected leg. Machine learning classification serves as an instrumental post-processing methodology for the signal acquired through a wearable and wireless accelerometer or gyroscope. A summary of machine learning platforms is presented. The application and demonstration of machine learning as a diagnostic tool is described within the scope of gait, reflex response, and associated subjects. The amalgamation of machine learning and wearable and wireless systems is anticipated to further evolve Network Centric Therapy with capabilities, such as prognostic assessment of rehabilitation, objective consideration of therapy efficacy, therapy optimization, and diagnosis of appropriate transitional phases of therapy strategy.

The context of smartphones and portable media devices as wearable and wireless systems can logically be extrapolated to homebound therapy, especially with regards to of a rehabilitation for hemiparesis from traumatic brain injury and stroke. Four applications are addressed. The portable media device operating as a functionally wireless accelerometer platform can be mounted to a cane for machine learning classification to distinguish appropriate and inappropriate use. An ankle rehabilitation system can apply a smartphone as a wireless gyroscope to differentiate between a hemiplegic ankle and unaffected ankle. Further applications using a portable media device as a wireless gyroscope platform involve the use of a wobble board with machine learning also classifying between a hemiplegic ankle and unaffected ankle. Another scenario applies the smartphone as a wireless gyroscope for Virtual Proprioception as feedback for eccentric training while applying machine learning to classify between Virtual Proprioception feedback and without Virtual Proprioception feedback for eccentric training. These preliminary systems are capable of providing essentially autonomous homebound therapy amendable for Network Centric Therapy.

The previous ten chapters demonstrate the vast utility of wearable and wireless systems for the quantification of reflex and gait. These evolutionary trends are envisioned to facilitate the development of Network Centric Therapy. A perspective from the authors on the role of Network Centric Therapy and associated opportunities is briefly presented.