Artificial Neural Network Analyzing Wearable Device Gait Data for Identifying Patients With Stroke Unable to Return to Work

Abstract

A potential dramatic effect of long-term disability due to stroke is the inability to return to work. An accurate prognosis and the identification of the parameters inflating the possibility of return to work after neurorehabilitation are crucial. Many factors may influence it, such as mobility and, in particular, walking ability. In this pilot study, two emerging technologies have been combined with the aim of developing a prognostic tool for identifying patients able to return to work: a wearable inertial measurement unit for gait analysis and an artificial neural network (ANN). Compared with more conventional statistics, the ANN showed a higher accuracy in identifying patients with respect to healthy subjects (90.9 vs. 75.8%) and also in identifying the subjects unable to return to work (93.9 vs. 81.8%). In this last analysis, the duration of double support phase resulted the most important input of the ANN. The potentiality of the ANN, developed also in other fields such as marketing on social networks, could allow a powerful support for clinicians that today should manage a large amount of instrumentally recorded parameters in patients with stroke.

Prognostic Factors in Neurorehabilitation of Stroke: A Comparison among Regression, Neural Network, and Cluster Analyses

“…The artificial neural network analysis was conducted by the ARIANNA model (ARtificial Intelligent Assistant for Neural Network Analysis), already used in other studies [ 14 , 15 ]. It is based on a multilayer perceptron procedure, and it is formed by the input layer (in which the above listed 30 variables are entered), two hidden layers (5 elements in each one), and a final output layer (the output of which was the predicted outcome).…”

“…The activation function for all the units in the hidden layers and for the output layer was a hyperbolic tangent. The chosen computational procedure was based on an online training [ 14 , 15 ].…”

“…The prognostic factors identified by AI, usually with an accuracy ≥ 70%, were similar to those classically accounted for: clinical test scores at admission, time from stroke onset to rehabilitation admission, age, sex, body mass index, and dysphasia [ 12 ]. Some other studies successfully used neural networks for assessing specific outcomes such as independency in toileting [ 13 ] or return to work [ 14 , 15 ].…”

“…Few sporadic studies have aimed at comparing two of these three methods (such as logistic regression vs. cluster analysis [ 17 ] or logistic regression vs. neural network [ 14 ]) or further testing their predictive accuracy on a sample independent of the originally analyzed dataset [ 16 ].…”

Assessment Methods of Post-stroke Gait: A Scoping Review of Technology-Driven Approaches to Gait Characterization and Analysis

“…Recently, artificial intelligence, including machine learning techniques has emerged as a promising tool for processing instrumented gait data efficiently (43)(44)(45)(46)(47). These techniques have found applications in numerous problems relating to gait assessment, including dimensionality reduction, feature extraction, and classification.…”