Abstract
Balance disorders present a significant healthcare burden due to the potential for hospitalization or complications for the patient, especially among the elderly population when considering intangible losses such as quality of life, morbidities, and mortalities. This work is a continuation of our earlier works where we now examine feature extraction methodology on Dynamic Gait Index (DGI) tests and machine learning classifiers to differentiate patients with balance problems versus normal subjects on an expanded cohort of 60 patients. All data was obtained using our custom designed low-cost wireless gait analysis sensor (WGAS) containing a basic inertial measurement unit (IMU) worn by each subject during the DGI tests. The raw gait data is wirelessly transmitted from the WGAS for real-time gait data collection and analysis. Here we demonstrate predictive classifiers that achieve high accuracy, sensitivity, and specificity in distinguishing abnormal from normal gaits. These results show that gait data collected from our very low-cost wearable wireless gait sensor can effectively differentiate patients with balance disorders from normal subjects in real-time using various classifiers. Our ultimate goal is to be able to use a remote sensor such as the WGAS to accurately stratify an individual’s risk for falls.
Highlights
Patients presenting with balance insufficiency or vestibular hypofunction may be at risk for injuries from falling
An artificial neural network (ANN) classifier was trained to distinguish normal from abnormal gaits with a random 80–20% train–test split of the wireless gait analysis sensor (WGAS) dataset
All classifiers were evaluated on the same test set. It was vital true negative rate was shown, as clinicians and patients would certainly desire to avoid preemptive that a high true negative rate was shown, as clinicians and patients would certainly desire to avoid treatment for vestibular hypofunction when, no such pathology was truly present
Summary
Patients presenting with balance insufficiency or vestibular hypofunction may be at risk for injuries from falling. Biosensors 2019, 9, 29 with balance insufficiency or vestibular hypofunction versus individuals with normal gaits is important to identify patients at risk for falling. The defining events in walking—heel strike, toe strike, heel-off, and toe-off—were automatically extracted, again using accelerometers [8] Such measurements of walking motions have been used to categorize gaits as normal or abnormal [9]. In our efforts to evaluate predictive modeling of fall risk, machine learning is applied to classify patients as having either a normal or abnormal gait on the basis of measurements from a small low-cost wearable wireless gait analysis sensor (WGAS). The long-term aim is to determine the applicability of WGAS for fall prediction using classifiers that demonstrate high performance on the patient cohort, suggesting that accurate predictions would be made on patients presenting for evaluation of vestibular hypofunction
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
