Machine learning algorithms for activity recognition in ambulant children and adolescents with cerebral palsy

Abstract

BackgroundCerebral palsy (CP) is the most common physical disability among children (2.5 to 3.6 cases per 1000 live births). Inadequate physical activity (PA) is a major problem effecting the health and well-being of children with CP. Practical, yet accurate measures of PA are needed to evaluate the effectiveness of surgical and therapy-based interventions to increase PA. Accelerometer-based motion sensors have become the standard for objectively measuring PA in children and adolescents; however, current methods for estimating physical activity intensity in children with CP are associated with significant error and may dramatically underestimate HPA in children with more severe mobility limitations. Machine learning (ML) models that first classify the PA type and then predict PA intensity or energy expenditure using activity specific regression equations may be more accurate than standalone regression models. However, the feasibility and validity of ML methods has not been explored in youth with CP. Therefore, the purpose of this study was to develop and test ML models for the automatic identification of PA type in ambulant children with CP.MethodsTwenty two children and adolescents (mean age: 12.8 ± 2.9 y) with CP classified at GMFCS Levels I to III completed 7 activity trials while wearing an ActiGraph GT3X+ accelerometer on the hip and wrist. Trials were categorised as sedentary (SED), standing utilitarian movements (SUM), comfortable walking (CW), and brisk walking (BW). Random forest (RF), support vector machine (SVM), and binary decision tree (BDT) classifiers were trained with features extracted from the vector magnitude (VM) of the raw acceleration signal using 10 s non-overlapping windows. Performance was evaluated using leave-one-subject out cross validation.ResultsSVM (82.0–89.0%) and RF (82.6–88.8%) provided significantly better classification accuracy than BDT (76.1–86.2%). Hip (82.7–85.5%) and wrist (76.1–82.6%) classifiers exhibited comparable prediction accuracy, while the combined hip and wrist (86.2–89.0%) classifiers achieved the best overall performance. For all classifiers, recognition accuracy was excellent for SED (94.1–97.9%), good to excellent for SUM (74.0–96.6%) and brisk walking (71.5–86.0%), and modest for comfortable walking (47.6–70.4%). When comfortable and brisk walking were combined into a single walking class, recognition accuracy ranged from 90.3 to 96.5%.ConclusionsML methods provided acceptable classification accuracy for detection of a range of activities commonly performed by ambulatory children with CP. The resultant models can help clinicians more effectively monitor bouts of brisk walking in the community. The results indicate that 2-step models that first classify PA type and then predict energy expenditure using activity specific regression equations are worthy of exploration in this patient group.Electronic supplementary materialThe online version of this article (10.1186/s12984-018-0456-x) contains supplementary material, which is available to authorized users.

Instrumented assessment of motor function in dyskinetic cerebral palsy: a systematic review

“…In the recent years wearable sensor techniques have increasingly been used for detecting specific movements of interest, e.g. stereotypical movement patterns in epilepsies as well as for activity monitoring in neurological disorders [95] including the general population of CP [96,97]. However, no study was found specifically for dyskinetic CP.…”

From AAL to ambient assisted rehabilitation: a research pilot protocol based on smart objects and biofeedback

“…More generally speaking, the present study is also limited from the algorithmic point of view, not taking yet into consideration the growing and promising area of Machine Learning and AI techniques to classify and assess movements and performance from sensors' readings. These are perspectives that of course deserve to be further studied, having already produced interesting results in the field (Ahmadi et al 2018;Hagenbuchner et al 2015;Trost et al 2016). The difficulty however is also related to the freedom of the performed movements, since the tasks of each exercise could be accomplished by different subjects in completely different ways, according to the nature and severity of each child's pathology.…”

A Method to Detect Type 1 Diabetes Based on Physical Activity Measurements Using a Mobile Device

“…Assessment of the physical activity was based on ActiGraph activity monitor measurements. The ActiGraph measurements for health-related research were also carried out in published findings [17][18][19][20].…”

“…Features that can be derived from the accelerometer have been also used to recognize the presence and severity of motor fluctuations in patients with Parkinson's disease [18]. It has been also used with measurements of physical activity to evaluate the effectiveness of surgical and therapy-based interventions in children with cerebral palsy or to derive diurnal rest-activity patterns from actigraphy in adolescents and to analyze associations with adiposity measures and cardiometabolic risk factors [19,20].…”