Human Lower Limb Motion Capture and Recognition Based on Smartphones.

Human motion recognition based on wearable devices plays a vital role in pervasive computing. Smartphones have built-in motion sensors that measure the motion of the device with high precision. In this paper, we propose a human lower limb motion capture and recognition approach based on a Smartphone. We design a motion logger to record five categories of limb activities (standing up, sitting down, walking, going upstairs, and going downstairs) using two motion sensors (tri-axial accelerometer, tri-axial gyroscope). We extract the motion features and select a subset of features as a feature vector from the frequency domain of the sensing data using Fast Fourier Transform (FFT). We classify and predict human lower limb motion using three supervised learning algorithms: Naïve Bayes (NB), K-Nearest Neighbor (KNN), and Artificial Neural Networks (ANNs). We use 670 lower limb motion samples to train and verify these classifiers using the 10-folder cross-validation technique. Finally, we design and implement a live detection system to validate our motion detection approach. The experimental results show that our low-cost approach can recognize human lower limb activities with acceptable accuracy. On average, the recognition rate of NB, KNN, and ANNs are 97.01%, 96.12%, and 98.21%, respectively.

Unilateral arm rehabilitation for persons with multiple sclerosis using serious games in a virtual reality approach: Bilateral treatment effect?

IMPAIRED: arm function and loss of manual dexterity can lead to decreased independence in activities of daily living in persons with Multiple Sclerosis (MS). In this study we verified the feasibility and efficacy of a serious games approach to supervised upper limb rehabilitation of the more affected arm in persons with MS and the cross-over effect to the nontreated arm. METHODS: Eighteen persons with moderate to severe MS symptoms participated (mean age 56.1 (range 28-73) years; mean disease duration 17.6 (4-35) years). Each participant received 12 supervised sessions of serious games (45 min, 12 sessions) aimed at improving the most affected upper limb. Primary outcomes were the Nine Hole Peg Test (9HPT) and the Box and Blocks Test (BBT). Perceived health was evaluated pre and post intervention with SF-12 and the VAS of the EuroQual-5DL. Non parametric tests were used and P was set at 0.05. RESULTS: After the serious games training participants improved dexterity and arm function bilaterally (10-18%), however, there was a statistically significant improvement only in the treated arm (P<0.05). Perceived menthal health improved follwing training (P<0.05) but not perceived physical health. CONCLUSION: An in clinic intervention with a serious-games virtual reality approach positively influenced arm recovery in persons moderately to severely affected by MS, improving mainly the treated arm but with positive effects on the nontreated arm. The persons were motivated during the intervention and expressed being willing to continue this kind of training at home as part of continuity of care.

Serious games for arm rehabilitation of persons with multiple sclerosis. A randomized controlled pilot study.

OBJECTIVES: The feasibility and preliminary evidence for efficacy of a serious games platform compared to exergame using the Wii for arm rehabilitation in persons with multiple sclerosis (MS) was investigated. METHODS: A pilot single-blind randomized (2:1) controlled in clinic trial was carried out. Sixteen persons with MS participated (age years 56.8 (SD 12.3), MS-onset years 19.4 (SD 12.3), EDSS 6.5). Ten participants used a serious games platform (Rehab@Home) while 6 participants played with the commercial Wii platform, for four weeks (40min, 12 sessions/4 weeks). Feasibility and user experience measures were collected. Primary outcomes were the 9 Hole Peg Test (9HPT) and the Box and Block test (BBT). Secondary outcomes were the EQ-5D visual analogue scale (EQ-VAS) and the SF-12. Nonparametric analysis was used to verify changes from pre to post rehabilitation within group and treatment effect was verified with Mann-Whitney U test. P value was set at 0.10 and clinical improvement was set at 20% improvement from baseline. RESULTS: Serious games were perceived positively in terms of user experience and motivation. There were clinically significant improvements in arm function in the serious games group as measured by 9HPT (38-29.5s, P = 0.046, > 20%) and BBT 32-42 cubes, P = 0.19, > 20%) following the 12 gaming sessions while the exergame group did not improve on either test (9HPT 34.5-41.5s, P = 0.34; BBT 38,5 to 42 cubes, P = 0.34). Only the exergame group perceived themselves as having improved their health. There was a significant between groups treatment effect only in perception of health (EQ-VAS) (Z = 1.93, P = 0.06) favouring the exergame group. CONCLUSIONS: Virtual reality in a serious gaming approach was feasible and beneficial to arm function of persons with MS but motivational aspects of the approach may need further attention.

Detecting freezing of gait with a tri-axial accelerometer in Parkinson's disease patients.

Freezing of gait (FOG) is a common motor symptom of Parkinson's disease (PD), which presents itself as an inability to initiate or continue gait. This paper presents a method to monitor FOG episodes based only on acceleration measurements obtained from a waist-worn device. Three approximations of this method are tested. Initially, FOG is directly detected by a support vector machine (SVM). Then, classifier's outputs are aggregated over time to determine a confidence value, which is used for the final classification of freezing (i.e., second and third approach). All variations are trained with signals of 15 patients and evaluated with signals from another 5 patients. Using a linear SVM kernel, the third approach provides 98.7% accuracy and a geometric mean of 96.1%. Moreover, it is investigated whether frequency features are enough to reliably detect FOG. Results show that these features allow the method to detect FOG with accuracies above 90% and that frequency features enable a reliable monitoring of FOG by using simply a waist sensor.