Predicting and Monitoring Upper-Limb Rehabilitation Outcomes Using Clinical and Wearable Sensor Data in Brain Injury Survivors

Abstract

Objective: Rehabilitation specialists have shown considerable interest for the development of models, based on clinical data, to predict the response to rehabilitation interventions in stroke and traumatic brain injury survivors. However, accurate predictions are difficult to obtain due to the variability in patients' response to rehabilitation interventions. This study aimed to investigate the use of wearable technology in combination with clinical data to predict and monitor the recovery process and assess the responsiveness to treatment on an individual basis. Methods: Gaussian Process Regression-based algorithms were developed to estimate rehabilitation outcomes (i.e., Functional Ability Scale scores) using either clinical or wearable sensor data or a combination of the two. Results: The algorithm based on clinical data predicted rehabilitation outcomes with a Pearson's correlation of 0.79 compared to actual clinical scores provided by clinicians but failed to model the variability in responsiveness to the intervention observed across individuals. In contrast, the algorithm based on wearable sensor data generated rehabilitation outcome estimates with a Pearson's correlation of 0.91 and modeled the individual responses to rehabilitation more accurately. Furthermore, we developed a novel approach to combine estimates derived from the clinical data and the sensor data using a constrained linear model. This approach resulted in a Pearson's correlation of 0.94 between estimated and clinician-provided scores. Conclusion: This algorithm could enable the design of patient-specific interventions based on predictions of rehabilitation outcomes relying on clinical and wearable sensor data. Significance: This is important in the context of developing precision rehabilitation interventions.

IMU-Based Monitoring for Assistive Diagnosis and Management of IoHT: A Review

“…Research on the monitoring and rehabilitation of patients who suffered from neurological trauma, such as traumatic brain injury (BI), spinal cord injury (SCI), and brachial plexus injury (BPI), have also produced results. For monitoring the recovery process for stroke and traumatic brain injury survivors, Lee et al [ 56 ] proposed a GPR-based regression model to estimate rehabilitation outcomes using a combination of clinical and wearable inertial sensor data. This approach resulted in a Pearson’s correlation of 0.94 between the estimated and clinician-provided scores.…”

“…Linear Discriminant Analysis (LDA), Naïve Bayesian, k-nearest neighbor (k-NN), and shallow ANN are also used as typical ML techniques to build specific classifiers. For supervised regression tasks, such as joint movement tracking or clinical score estimating, Support Vector Regression (SVR) and Gaussian Progress Regression (GPR) models are the most popular choices [ 54 , 55 , 56 ]. Some articles also used linear regression (LR) to demonstrate the relation of the medical outcomes and handcrafted features [ 25 ].…”

IMU-Based Monitoring for Assistive Disease Diagnosis and Management of IoHT: A Review

“…Linear Discriminant Analysis (LDA), Naive Bayes (NB), k-nearest neighbor (k-NN), and shallow ANN are also used as typical ML techniques to build specific classifiers. For supervised regression tasks like joint movement tracking or clinical score estimating, Support Vector Regression (SVR) and Gaussian Progress Regression (GPR) models are the most popular choice [73], [79], [95]. Some articles also used linear regression (LR) to demonstrate the relation of the medical outcomes and handcrafted features [104].…”

“…Research on the monitoring and rehabilitation of patients who have suffered from neurological trauma such as traumatic brain injury (BI), spinal cord injury (SCI), and brachial plexus injury (BPI) have also produced results. For monitoring the recovery process for stroke and traumatic brain injury survivors, Lee et al [95] proposed a GPR-based regression model to estimate rehabilitation outcomes using a combination of clinical and wearable inertial sensor data. This approach resulted in a Pearson's correlation of 0.94 between estimated and clinician-provided scores.…”

ARMIA: A Sensorized Arm Wearable for Motor Rehabilitation

“…Moreover, kinematic metrics have a good correlation with conventional clinical scales, which provides an additional way of enhancing motor therapy [ 13 ]. Recent studies also show how ADLs can benefit from the application of this technology in rehabilitation [ 14 , 15 ]. Upper-limb motor rehabilitation can also benefit from the assessment of neuromuscular behavior.…”