Search

Deep brain stimulation offers an advanced means of treating Parkinson’s disease in a patient specific context. However, a considerable challenge is the process of ascertaining an optimal parameter configuration. Imperative for the deep brain stimulation parameter optimization process is the quantification of response feedback. As a significant improvement to traditional ordinal scale techniques is the advent of wearable and wireless systems. Recently conformal wearable and wireless systems with a profile on the order of a bandage have been developed. Previous research endeavors have successfully differentiated between deep brain stimulation “On” and “Off” status through quantification using wearable and wireless inertial sensor systems. However, the opportunity exists to further evolve to an objectively quantified response to an assortment of parameter configurations, such as the variation of amplitude, for the deep brain stimulation system. Multiple deep brain stimulation amplitude settings are considered inclusive of “Off” status as a baseline, 1.0 mA, 2.5 mA, and 4.0 mA. The quantified response of this assortment of amplitude settings is acquired through a conformal wearable and wireless inertial sensor system and consolidated using Python software automation to a feature set amenable for machine learning. Five machine learning algorithms are evaluated: J48 decision tree, K-nearest neighbors, support vector machine, logistic regression, and random forest. The performance of these machine learning algorithms is established based on the classification accuracy to distinguish between the deep brain stimulation amplitude settings and the time to develop the machine learning model. The support vector machine achieves the greatest classification accuracy, which is the primary performance parameter, and K-nearest neighbors achieves considerable classification accuracy with minimal time to develop the machine learning model.

Deep brain stimulation offers the opportunity for patient specific and optimally tuned treatment to ameliorate Parkinson’s disease. Imperative for the efficacy of deep brain stimulation is the quantification of therapeutic response. The implementation of inertial sensor systems has considerably augmented the acuity of quantifying the characteristics of the movement disorder through the respective signal data. For example, these inertial sensors constitute effectively wearable and wireless systems that can be mounted about the dorsum of the hand to objectively quantify tremor. Using machine learning algorithms considerable classification accuracy has been achieved. Conformal wearable and wireless inertial sensor systems have been recently developed that can be conveniently mounted about the dorsum of the hand by an adhesive medium with a profile on the order of a bandage. Additionally, these conformal wearable and wireless inertial sensor systems can have their data wirelessly transmitted to a secure Cloud computing environment. This capability offers the opportunity to ascertain the inertial sensor signal for Parkinson’s disease tremor response in the context of multiple deep brain stimulation parameter configurations, such as an assortment of amplitude settings. The objective of the research was to quantify and distinguish the response of Parkinson’s disease tremor through an assortment of amplitude settings for deep brain stimulation, such as ‘Off’ setting as a baseline, amplitude set to 1.0 mA, amplitude set to 1.75 mA, amplitude set to 2.5 mA, amplitude set to 3.25 mA, and amplitude set to 4.0 mA, using a deep learning convolutional neural network through TensorFlow with the quantification of tremor by a conformal wearable and wireless inertial sensor system. Considerable classification accuracy was attained to distinguish between these specified deep brain stimulation settings for the amelioration of Parkinson’s disease.

The development of wearable and wireless inertial sensor systems enables a highly robust and convenient means for quantifying gait. In particular, the quantification hemiplegic gait can objectively identify disparity, such as with respect to reduced arm swing. Recent developments have produced the BioStamp nPoint, which is a conformal wearable and wireless inertial sensor system. The BioStamp nPoint is conveniently mounted by adhesive to an aspect of the human anatomy, such as the dorsum of the hand, for quantifying reduced arm swing for hemiplegic gait. The acquired inertial sensor data quantifying the characteristics of reduced arm swing for hemiplegic gait is wirelessly transmitted to a secure Cloud computing environment for subsequent post-processing. The pertinent gyroscope signal data is consolidated into a feature set suitable for machine learning classification. Using a multilayer perceptron neural network considerable classification accuracy is attained to distinguish between a hemiplegic affected arm with reduced arm swing and the unaffected arm. These accomplishments imply a pathway for ameliorating reduced arm swing through the application of conformal wearable and wireless inertial sensor systems, such as the BioStamp nPoint.

The advent of the conformal wearable and wireless inertial sensor enables expanded utility for progressive evolution of gait quantification. The acquired signal data, such as through a gyroscope, can be wirelessly transmitted to a secure Cloud computing environment for subsequent post-processing. In particular, the amalgamation of machine learning classification can be applied for the distinction of hemiplegic gait in the context of the hemiplegic affected leg and the unaffected leg. The integration of these technologies enables the realization of Network Centric Therapy, for which the patient can ascertain the benefit of gait quantification through extremely lightweight conformal wearable and wireless inertial sensor systems that have profiles relative to a bandage. Furthermore, the clinical rehabilitation team can remotely instill therapy optimized with the augmented acuity of machine learning. The research endeavor successfully demonstrates considerable classification accuracy through machine learning for the differentiation of a hemiplegic affected leg and unaffected leg during gait.

A standard aspect of the neurological examination involves the assessment of diadochokinesia. Diadochokinesia pertains the other swift alternation of an agonist and antagonist muscle pair, such as pronation and supination with respect to the forearm. For example, hemiparesis can manifest perceptible distinction for diadochokinesia with respect to the hemiplegic affected forearm and unaffected forearm. Standard techniques involve ordinal scales, which are inherently subjective, and the development of quantified methods has been recommended. The advent of conformal wearable and wireless inertial sensor systems, such as the BioStamp nPoint with a lightweight and bandage-like profile, enables quantification of diadochokinesia with wireless access to a Cloud computing environment for post-processing. The gyroscope signal data quantifying diadochokinesia can be consolidated to a feature set for machine learning classification to differentiate between a hemiplegic affected forearm and unaffected forearm. Considerable machine learning classification accuracy is attained for distinguishing between a hemiplegic affected forearm and unaffected forearm based on the quantified gyroscope signal data acquired from a conformal wearable and wireless inertial sensor system.

Deep brain stimulation enables highly specified patient-unique therapeutic intervention ameliorating the symptoms of Parkinson’s disease. Inherent to the efficacy of deep brain stimulation is the acquisition of an optimal parameter configuration. Using conventional methods, the optimization process for tuning the deep brain stimulation system parameters can intrinsically induce strain on clinical resources. An advanced means of quantifying Parkinson’s hand tremor and distinguishing between parameter settings would be highly beneficial. The conformal wearable and wireless inertial sensor system, such as the BioStamp nPoint, has a volumetric profile on the order of a bandage that readily enables convenient quantification of Parkinson’s disease hand tremor. Furthermore, the BioStamp nPoint has been certified by the FDA as a 510(k) medical device for acquisition of medical grade data. Parametric variation of the amplitude parameter for deep brain stimulation can be quantified through the BioStamp nPoint conformal wearable and wireless inertial sensor system mounted to the dorsum of the hand. The acquired inertial sensor signal data can be wirelessly transmitted to a secure Cloud computing environment for post-processing. The quantified inertial sensor data for the parametric study of the effects of varying amplitude can be distinguished through machine learning classification. Software automation through Python can consolidate the inertial sensor data into a suitable feature set format. Using the multilayer perceptron neural network considerable machine learning classification accuracy is attained to distinguish multiple parametric settings of amplitude for deep brain stimulation, such as 4.0 mA, 2.5 mA, 1.0 mA, and ‘Off’ status representing a baseline. These findings constitute an advance toward the pathway of attaining real-time closed loop automated parameter configuration tuning for treatment of Parkinson’s disease using deep brain stimulation.

Essential tremor is a prevalent neurodegenerative movement disorder. Deep brain stimulation represents a highly effective means of treatment, especially for scenarios for which traditional medical intervention is no longer feasible. One of the major post-operative challenges is the determination of an optimal set of tuning parameters. Optimizing the deep brain stimulation parameters can impart a time-intensive task to the clinician. The smartphone in the context of a wearable and wireless inertial sensor system offers the capability to objectively quantify the characteristics of the tremor. Machine learning in conjunction with a wearable and wireless inertial sensor system, such as a smartphone, can distinguish between disparate states, such as deep brain stimulation in ‘On’ and ‘Off’ status. Multiple machine learning classification techniques are available, such as the multilayer perceptron neural network, support vector machine, K-nearest neighbors, logistic regression, J-48 decision tree, and random forest. The objective of this research endeavor is to evaluate these six machine learning classification algorithms for classification of deep brain stimulation regarding ‘On’ and ‘Off’ status for Essential tremor during a reach and grasp task. The reach and grasp task is quantified through the smartphone as wearable and wireless inertial sensor system mounted to the dorsum of the hand and secured by latex glove. Multiple feature set scenarios are considered, such as recordings from both the accelerometer and gyroscope, accelerometer, and gyroscope. These scenarios facilitate the determination of the most robust machine learning algorithms. The multilayer perceptron neural network, support vector machine, K-nearest neighbors, and logistic regression achieve the highest classification accuracy for three feature set scenarios derived by recordings from both accelerometer and gyroscope, accelerometer, and gyroscope.

Recently, machine learning has augmented the capability of the amalgamation of wearable and wireless systems for deep brain stimulation systems. Machine learning platforms have been applied to attain considerable classification accuracy for distinguishing between deep brain stimulation set to “On” and “Off” modes for Essential tremor and Parkinson’s disease. Other movement disorders, such as hemiplegic affected and unaffected limb pairs, have been successfully differentiated through machine learning classification. Central to these machine learning endeavors has been the application of wearable and wireless systems using inertial sensors, such as the accelerometer and gyroscope, to consolidate signal data into feature sets for machine learning classification. An assortment of prevalent machine learning platforms is discussed, such as J48 decision tree, K-nearest neighbor, logistic regression, support vector machine, multilayer perceptron neural network, and random forest. Machine learning is envisioned to serve an instrumental role for the objective of achieving closed-loop optimization of deep brain stimulation parameter configurations. In essence, machine learning is envisioned to function as a predominant role for the post-processing perspective of Network Centric Therapy.

The considerable advantage of integrating wearable and wireless systems with machine learning for the assessment of deep brain stimulation parameter configuration status is addressed. A wearable and wireless system, such as a smartphone with its accelerometer and gyroscope, provides the quantified basis for the efficacy determination of a treatment strategy. In particular, deep brain stimulation is well suited for being amalgamated with wearable and wireless systems. For a subject with Parkinson’s disease, deep brain stimulation set to “On” and “Off” status is distinguished through an assortment of machine learning algorithms, such as J48 decision tree, K-nearest neighbors, logistic regression, support vector machine, multilayer perceptron neural network, and random forest. The feature set is consolidated from the accelerometer signal and gyroscope signal from a smartphone using software automation. The appropriateness for these machine learning algorithms was assessed in terms of both classification accuracy and computational efficiency. These capabilities further refine the opportunities of machine learning classification being allocated local to the wearable and wireless system with an available Cloud computing resource. These findings establish a preliminary perspective regarding the utility of Network Centric Therapy, for which effectively real-time optimization of parameter configurations for deep brain stimulation can be developed in a patient-specific context. Furthermore, the real-time optimization process can be adaptive to the inherent temporal fluctuations of a progressive neurodegenerative movement disorder, such as Parkinson’s disease and Essential tremor.

The smartphone represents a wearable and wireless system with the potential to have transformative influence on the biomedical and healthcare industry. An intrinsic feature of the smartphone is a gyroscope sensor, for which with a software application the smartphone functions as a wearable and wireless gyroscope platform. The resultant gyroscope data recording presents a clinical recognizable signal, which has been successful demonstrated to quantify aspects of human movement characteristics, such as the patellar tendon reflex. Gait another associated feature of human movement can be readily quantified by a smartphone functioning as a wearable and wireless gyroscope platform. The research objective is to distinguish between an affected leg and unaffected leg during hemiplegic gait based on a smartphone functioning as a wearable and wireless gyroscope platform though machine learning classification. A single smartphone is applied to quantify hemiplegic gait. The smartphone is first mounted to the affected leg and then the unaffected leg with velocity constrained to a constant velocity by a treadmill. Through wireless connectivity to the Internet the gyroscope signal data is conveyed as an email attachment for post-processing at a remote location. Software automation consolidates the gyroscope signal data of hemiplegic gait to a feature set for machine learning classification. With the application of a multilayer perceptron neural network considerable classification accuracy is attained for distinguishing between the affected leg and unaffected leg of hemiplegic gait. Future implications of the successful implementation of a smartphone as a wearable and wireless gyroscope for machine learning classification of hemiplegic gait through a multilayer perceptron neural network elucidate pathways to highly optimized therapy through machine learning with the potential for patients to reside remote from their therapist.

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.

Current forecasts imply a significant increase in the quantity of lower limb amputations. Synergizing the capabilities of a conventional gait analysis system and machine learning facilitates the capacity to classify disparate types of transtibial prostheses. Automated classification of prosthesis type may eventually advance rehabilitative acuity for selecting an appropriate prosthesis for a given aspect of the rehabilitation process. The presented research utilized a force plate as a conventional gait analysis device to acquire a feature set for two types of prosthesis: passive Solid Ankle Cushioned Heel (SACH) and the iWalk BiOM powered prosthesis. The feature set consists of both temporal and kinetic data with respect to the force plate signal during stance. Intuitively a passive prosthesis and powered prosthesis generate distinctively different force plate recordings. A support vector machine, which is type of machine learning application, achieves 100% classification between a passive prosthesis and powered prosthesis regarding the feature set derived from force plate recordings.

Essential tremor (ET) is a highly prevalent movement disorder. Patients with ET exhibit a complex progressive and disabling tremor, and medical management often fails. Deep brain stimulation (DBS) has been successfully applied to this disorder, however there has been no quantifiable way to measure tremor severity or treatment efficacy in this patient population. The quantified amelioration of kinetic tremor via DBS is herein demonstrated through the application of a smartphone (iPhone) as a wireless accelerometer platform. The recorded acceleration signal can be obtained at a setting of the subject's convenience and conveyed by wireless transmission through the Internet for post-processing anywhere in the world. Further post-processing of the acceleration signal can be classified through a machine learning application, such as the support vector machine. Preliminary application of deep brain stimulation with a smartphone for acquisition of a feature set and machine learning for classification has been successfully applied. The support vector machine achieved 100% classification between deep brain stimulation in `on' and `off' mode based on the recording of an accelerometer signal through a smartphone as a wireless accelerometer platform.

The synergy of gait analysis tools with machine learning enables the capacity to classify disparity existing in hemiplegic gait. Hemiplegic gait is characterized by an affected leg and unaffected leg, which can be quantified by the measurement of a force plate. The characteristic features of the force plate recording for gait consist of a two local maxima that represent the braking phase and push off phase of stance and their associated parameters. The quantified features of a hemiplegic pair of affected leg and unaffected leg force plate recordings are intuitively disparate. Logistic regression achieves 100% classification between an affected and unaffected hemiplegic leg pair based on the feature set of the force plate data.

The patellar tendon reflex response provides fundamental means of assessing a subject’s neurological health. Dysfunction regarding the characteristics of the reflex response may warrant the escalation to more advanced diagnostic techniques. Current strategies involve the manual elicitation of the patellar tendon reflex by a highly skilled clinician with subsequent interpretation according to an ordinal scale. The reliability of the ordinal scale approach is a topic of contention. Highly skilled clinicians have been in disagreement regarding even the observation of asymmetric reflex pairs. An alternative strategy incorporated the ubiquitous smartphone with a software application to function as a wireless gyroscope platform for quantifying the reflex response. Each gyroscope signal recording of the reflex response can be conveyed wirelessly through Internet connectivity as an email attachment. The reflex response is evoked through a potential energy impact pendulum that enables prescribed targeting and potential energy level. The smartphone functioning as a wireless gyroscope platform reveals an observationally representative gyroscope signal of the reflex response. Three notably distinguishable attributes of the reflex response are incorporated into a feature set for machine learning: maximum angular rate of rotation, minimum angular rate of rotation, and time disparity between maximum and minimum angular rate of rotation. Four machine learning platforms such as the J48 decision tree, K-nearest neighbors, logistic regression, and support vector machine, were applied to the patellar tendon reflex response feature set incorporating a hemiplegic patellar tendon reflex pair. The J48 decision tree attained 98% classification accuracy, and the K-nearest neighbors, logistic regression, and support vector machine achieved perfect classification accuracy for distinguishing between a hemiplegic affected leg and unaffected leg patellar tendon reflex pair. The research findings reveal the potential of machine learning for enabling advanced diagnostic acuity respective of the gyroscope signal of the patellar tendon reflex response.

The characteristics of the patellar tendon reflex provide fundamental insight regarding the diagnosis of neurological status. Based on the features of the tendon reflex response, a clinician may establish preliminary perspective regarding the global condition of the nervous system. Current techniques for quantifying the observations of the reflex response involve the application of ordinal scales, requiring the expertise of a highly skilled clinician. However, the reliability of the ordinal scale approach is debatable. Highly skilled clinicians have even disputed the presence of asymmetric reflex pairs. An alternative strategy was the implementation of an iPod wireless accelerometer application to quantify the reflex response acceleration waveform. An application enabled the recording of the acceleration waveform and later wireless transmission as an email attachment by connectivity to the Internet. A potential energy impact pendulum enabled the patellar tendon reflex to be evoked in a predetermined and targeted manner. Three feature categories of the reflex response acceleration waveform (global parameters, temporal organization, and spectral features) were incorporated into machine learning to distinguish a subject's hemiplegic and healthy reflex pair. Machine learning attained perfect classification of the hemiplegic and healthy reflex pair. The research findings implicate the promise of machine learning for providing increased diagnostic acuity regarding the acceleration waveform of the tendon reflex response.

The implementation of a wireless accelerometer application for the quantification of gait may enable an autonomous strategy for the quantification of gait. A wireless accelerometer application may potentially enable the remote, autonomous, and quantified evaluation of gait beyond the confines of a clinical facility. The accuracy and reliability for the application of accelerometers for gait evaluation has been established through the selection of specific mounting positions based on the anatomy of the subject. Wireless accelerometer systems have been successfully demonstrated for establishing a quantified gait disparity for hemiparetic subjects. For example, the quantification of hemiparetic gait has been conducted using the lateral epicondyle of the femur as a mounting position [1].Copyright © 2010 by ASME