**4. Evolutionary pathway for Network Centric Therapy with respect to quantification of reflex response and latency**

The global evolutionary pathway for Network Centric Therapy derives from the Ph.D. Dissertation research conducted by Dr. LeMoyne, which lead to the progressive development of a device known as the Wireless Quantified Reflex Device through the incremental develop of four generations. The preliminary success involved the quantification of reflex response through locally wireless accelerometers. In order to measure the response of the patellar tendon reflex, the wireless accelerometers were mounted proximal to the lateral malleolus, which signified their wearable capability [17, 18, 40].

The original wireless accelerometers were provided through internal UCLA research, and they were referred to as MedNodes. The MedNodes required specialized operation, as they were the scope of graduate-level research at UCLA. These wireless accelerometer nodes that were noted as conveniently wearable were applied to the first and second generations of the Wireless Quantified Reflex Device, and the quantification of the patellar tendon reflex was measured in an accurate and reliable manner. The collected signal data of the wireless accelerometer was transmitted to a locally situated computer for post-processing [55, 56]. Central to all four generations of the Wireless Quantified Reflex Device was the integration of

*Wireless Sensor Networks - Design, Deployment and Applications*

for miniaturization and reliability these inertial sensors have been successfully applied to numerous human movement scenarios, such as the quantification of reflex, gait, and movement disorder. Furthermore, these inertial sensors were noted as functionally wearable with wireless in capability, which ushered the presence of wearable and wireless inertial sensor systems for quantifying human movement [1–12]. A historical and evolutionary perspective leading to the amalgamation of inertial sensors that are functionally wearable with wireless connectivity to Cloud computing resources in conjunction with machine learning classification as an advanced post-processing technique, which is known as

Network Centric Therapy, for the biomedical domain is presented.

between various available ordinal scales [1, 3, 11, 12, 16–30].

**reflex response, gait, and movement disorder**

latency can be derived [17, 18, 39, 40].

sensor signal data may provide a more revealing historical perspective.

**3. Electro-mechanical systems providing signal data for quantifying** 

The acquisition of quantified sensor signal data enables more pertinent clinical acuity regarding the health status with respect to reflex response, gait, and movement disorder [1–12]. With respect to the quantification of reflex response an assortment of electro-mechanical sensor systems have been proposed. These devices generally have consisted of the means for evoking the reflex through a provisional reflex hammer and quantifying the correlated reflex response [17, 18, 31–38]. By temporally synchronizing the input quantification device eliciting the reflex and output quantification sensor of the reflex response a functional reflex

The quantification of the input that commences the reflex has been demonstrated through instrumented provisional reflex hammers and motorized devices. These devices enable measuring of the intensity of the eliciting impact and the time stamp regarding the start of the reflex respective of the neurological pathway. The reflex response, such as deriving from the patellar tendon, can be measured through electromyograms (EMGs), strain gauges, optical motion cameras, force sensors, and wired inertial sensors in addition to the associated time stamp. The temporal

**movement disorder**

**2. Ordinal methodologies for quantifying reflex response, gait, and** 

Prior to the advent of wearable and wireless inertial sensor systems, the diagnosis of a subject's health status was essentially derived from the expert although subjective interpretation of a skilled clinician. The clinician is generally tasked with the responsibility to interpret the health of the patient and apply the observation to an ordinal scale criteria methodology. This ordinal scale process is ubiquitous to the clinical domain, and this approach is relevant to the scope of reflex response, gait, and movement disorder. However, the ordinal scale strategy encompasses contention regarding reliability, and there generally does not exist a means for translating

Further issues with the ordinal scale approach are evident with respect to the imperative need for patient-clinician interaction. From a logistical perspective a patient is required to travel to a clinical appointment, which in the case of a specialized expert may require relatively long-distance travel. Additionally, the clinician is only provided with a short duration of time to interpret the patient's health status, which may be in dispute to the true health condition of the patient. The ordinal scale approach intuitively only provides limited insight of patient health, for which

**292**

a quantified potential energy impact pendulum to consistently evoke the patellar tendon reflex [17, 18, 39, 40, 55, 56].

The third and fourth generations of the Wireless Quantified Reflex Device included a second wireless accelerometer to determine the time of impact for the quantified potential energy impact pendulum. The first wireless accelerometer was mounted to the ankle to quantify reflex response and time of response. Using the temporal offset of the wireless accelerometer mounted on the impact pendulum evoking the patellar tendon reflex and the wireless accelerometer mounted about the lateral malleolus mounted about the ankle to measure reflex response, a functional patellar tendon reflex latency was derived. The third and fourth generations of the Wireless Quantified Reflex Device incorporated the G-link wireless accelerometer developed by Microstrain [17, 18, 39, 40].

The third generation Wireless Quantified Reflex Device utilized streaming signal data to the locally situated portable computer for acquisition of the accelerometer signal data and subsequent post-processing. The third generation Wireless Quantified Reflex Device was the first evolution to feature the ability to derive functional reflex latency through the tandem wireless accelerometers with one wireless accelerometer located on the potential energy impact pendulum that evokes the patellar tendon reflex and the other wireless accelerometer mounted proximal to the lateral malleolus of the ankle for also acquiring reflex response. The research findings demonstrated that patellar tendon reflex response and associated functional latency could be both quantified with considerable accuracy and reliability [39].

The observations of the third generation Wireless Quantified Reflex Device established opportunity for improvement, such as increasing the sampling rate for the tandem accelerometers. This improvement would implicate better acuity with respect to the derived functional latency of the patellar tendon reflex. An artificial reflex device was applied as intermediary before the development of the fourth generation Wireless Quantified Reflex Device. This intermediary device utilized the data logger of the G-link wireless accelerometers, which permitted augmented sampling rates, while retaining wireless connectivity to a local portable computer for accelerometer signal data downloading and post-processing [57–59].

The fourth generation Wireless Quantified Reflex Device successful applied a longitudinal study for multiple subjects. With the wireless accelerometer set to data logger configuration with subsequent wireless transmission, the Wireless Quantified Reflex Device successfully acquired patellar tendon reflex response and functional latency with considerable accuracy, reliability, and reproducibility [40]. Subsequent evolutions encompass the application of more robust wearable and wireless inertial sensor systems and conjunction with machine learning to distinguish a hemiplegic reflex pair regarding affected patellar tendon reflex and associated unaffected patellar tendon reflex [60–62].

The next improvement incorporated the use of the portable media device and smartphone for the quantification of reflex response as a functional wireless accelerometer platform using the potential energy impact pendulum to evoke the patellar tendon reflex [63, 64]. The portable media device was suited for facilities with local wireless internet zones [63]. For locations requiring broad telecommunication access, the smartphone provides better benefit [64]. Both applications feature a common software application that enables a discrete recording of the accelerometer signal for quantifying the reflex response, and the signal data can be attached to an email for wireless transmission to the Internet for post-processing anywhere in the world [63, 64].

For example, LeMoyne and Mastroianni conducted an experiment to quantify reflex response using a portable media device applying supramaximal stimulation of the patellar tendon reflex in Lhasa, Tibet of China. The signal data was wirelessly transmitted to the Internet as an email attachment, which served as a provisional

**295**

**Figure 1.**

*energy impact pendulum to evoke the reflex [67].*

*An Evolutionary Perspective for Network Centric Therapy through Wearable and Wireless…*

to distinguish between a hemiplegic reflex pair is a matter of contention [21].

Cloud computing resource. The data was later downloaded in Flagstaff, Arizona of the United States of America, which is effectively on the other side of the world, for

Further advancements of the concept of quantifying reflex response, such as the patellar tendon, pertained to using the accelerometer signal, such as through a portable media device, to differentiate between a hemiplegic reflex pair. The hemiplegic affected leg's patellar tendon reflex response is notably more amplified relative to the patellar tendon reflex response of the unaffected leg. By consolidating the respective accelerometer signals to a feature set for machine learning classification using the support vector machine available through the Waikato Environment for Knowledge Analysis (WEKA) considerable machine learning classification accuracy was attained [60]. This achievement is notable, since subjective clinical observations

The gyroscope was eventually incorporated in the inertial sensor package of portable media devices and smartphones. The gyroscope provides a clinical representation for rotational characteristics of a joint, which represents the response of the patellar tendon reflex. Successfully demonstration of the gyroscope to quantify the patellar tendon reflex was demonstrated in the context of the Wireless Quantified Reflex Device through the potential energy impact pendulum [61, 62, 66–68].

Using both the potential energy impact pendulum and supramaximal stimulation to evoke the patellar tendon reflex response multiple machine learning algorithms using WEKA have achieved considerable classification accuracy [60–62, 66, 67]. **Figures 1** and **2** represent the gyroscope signal for the reflex response of the hemiplegic affected leg and unaffected leg, respectively. Machine learning algorithms, such as the J48 decision tree, provide a visualized basis for the most prevalent numeric attributes to establish classification accuracy, such as the time disparity between maximum and minimum angular rate of rotation for the patellar tendon reflex response, as

*The gyroscope signal of the patellar tendon reflex response for the hemiplegic affected leg using the potential* 

*DOI: http://dx.doi.org/10.5772/intechopen.95550*

post-processing [65].

illustrated in **Figure 3** [67].

*An Evolutionary Perspective for Network Centric Therapy through Wearable and Wireless… DOI: http://dx.doi.org/10.5772/intechopen.95550*

Cloud computing resource. The data was later downloaded in Flagstaff, Arizona of the United States of America, which is effectively on the other side of the world, for post-processing [65].

Further advancements of the concept of quantifying reflex response, such as the patellar tendon, pertained to using the accelerometer signal, such as through a portable media device, to differentiate between a hemiplegic reflex pair. The hemiplegic affected leg's patellar tendon reflex response is notably more amplified relative to the patellar tendon reflex response of the unaffected leg. By consolidating the respective accelerometer signals to a feature set for machine learning classification using the support vector machine available through the Waikato Environment for Knowledge Analysis (WEKA) considerable machine learning classification accuracy was attained [60]. This achievement is notable, since subjective clinical observations to distinguish between a hemiplegic reflex pair is a matter of contention [21].

The gyroscope was eventually incorporated in the inertial sensor package of portable media devices and smartphones. The gyroscope provides a clinical representation for rotational characteristics of a joint, which represents the response of the patellar tendon reflex. Successfully demonstration of the gyroscope to quantify the patellar tendon reflex was demonstrated in the context of the Wireless Quantified Reflex Device through the potential energy impact pendulum [61, 62, 66–68].

Using both the potential energy impact pendulum and supramaximal stimulation to evoke the patellar tendon reflex response multiple machine learning algorithms using WEKA have achieved considerable classification accuracy [60–62, 66, 67]. **Figures 1** and **2** represent the gyroscope signal for the reflex response of the hemiplegic affected leg and unaffected leg, respectively. Machine learning algorithms, such as the J48 decision tree, provide a visualized basis for the most prevalent numeric attributes to establish classification accuracy, such as the time disparity between maximum and minimum angular rate of rotation for the patellar tendon reflex response, as illustrated in **Figure 3** [67].

#### **Figure 1.**

*Wireless Sensor Networks - Design, Deployment and Applications*

ometer developed by Microstrain [17, 18, 39, 40].

tendon reflex [17, 18, 39, 40, 55, 56].

a quantified potential energy impact pendulum to consistently evoke the patellar

The third and fourth generations of the Wireless Quantified Reflex Device included a second wireless accelerometer to determine the time of impact for the quantified potential energy impact pendulum. The first wireless accelerometer was mounted to the ankle to quantify reflex response and time of response. Using the temporal offset of the wireless accelerometer mounted on the impact pendulum evoking the patellar tendon reflex and the wireless accelerometer mounted about the lateral malleolus mounted about the ankle to measure reflex response, a functional patellar tendon reflex latency was derived. The third and fourth generations of the Wireless Quantified Reflex Device incorporated the G-link wireless acceler-

The third generation Wireless Quantified Reflex Device utilized streaming signal data to the locally situated portable computer for acquisition of the accelerometer signal data and subsequent post-processing. The third generation Wireless Quantified Reflex Device was the first evolution to feature the ability to derive functional reflex latency through the tandem wireless accelerometers with one wireless accelerometer located on the potential energy impact pendulum that evokes the patellar tendon reflex and the other wireless accelerometer mounted proximal to the lateral malleolus of the ankle for also acquiring reflex response. The research findings demonstrated that patellar tendon reflex response and associated functional latency could be both quantified with considerable accuracy and reliability [39]. The observations of the third generation Wireless Quantified Reflex Device established opportunity for improvement, such as increasing the sampling rate for the tandem accelerometers. This improvement would implicate better acuity with respect to the derived functional latency of the patellar tendon reflex. An artificial reflex device was applied as intermediary before the development of the fourth generation Wireless Quantified Reflex Device. This intermediary device utilized the data logger of the G-link wireless accelerometers, which permitted augmented sampling rates, while retaining wireless connectivity to a local portable computer

for accelerometer signal data downloading and post-processing [57–59].

associated unaffected patellar tendon reflex [60–62].

The fourth generation Wireless Quantified Reflex Device successful applied a longitudinal study for multiple subjects. With the wireless accelerometer set to data logger configuration with subsequent wireless transmission, the Wireless Quantified Reflex Device successfully acquired patellar tendon reflex response and functional latency with considerable accuracy, reliability, and reproducibility [40]. Subsequent evolutions encompass the application of more robust wearable and wireless inertial sensor systems and conjunction with machine learning to distinguish a hemiplegic reflex pair regarding affected patellar tendon reflex and

The next improvement incorporated the use of the portable media device and smartphone for the quantification of reflex response as a functional wireless accelerometer platform using the potential energy impact pendulum to evoke the patellar tendon reflex [63, 64]. The portable media device was suited for facilities with local wireless internet zones [63]. For locations requiring broad telecommunication access, the smartphone provides better benefit [64]. Both applications feature a common software application that enables a discrete recording of the accelerometer signal for quantifying the reflex response, and the signal data can be attached to an email for wireless trans-

For example, LeMoyne and Mastroianni conducted an experiment to quantify reflex response using a portable media device applying supramaximal stimulation of the patellar tendon reflex in Lhasa, Tibet of China. The signal data was wirelessly transmitted to the Internet as an email attachment, which served as a provisional

mission to the Internet for post-processing anywhere in the world [63, 64].

**294**

*The gyroscope signal of the patellar tendon reflex response for the hemiplegic affected leg using the potential energy impact pendulum to evoke the reflex [67].*

**Figure 2.**

*The gyroscope signal of the patellar tendon reflex response for the unaffected leg using the potential energy impact pendulum to evoke the reflex [67].*

#### **Figure 3.**

*The J48 decision tree to distinguish between a hemiplegic affected leg and unaffected leg, for which the time disparity between maximum and minimum angular rate of rotation for the patellar tendon reflex response numeric attribute is illustrated as the most prevalent for establishing classification accuracy [67].*
