Prediction of Physical Frailty in Orthogeriatric Patients Using Sensor Insole–Based Gait Analysis and Machine Learning Algorithms: Cross-sectional Study

Background Assessment of the physical frailty of older patients is of great importance in many medical disciplines to be able to implement individualized therapies. For physical tests, time is usually used as the only objective measure. To record other objective factors, modern wearables offer great potential for generating valid data and integrating the data into medical decision-making. Objective The aim of this study was to compare the predictive value of insole data, which were collected during the Timed-Up-and-Go (TUG) test, to the benchmark standard questionnaire for sarcopenia (SARC-F: strength, assistance with walking, rising from a chair, climbing stairs, and falls) and physical assessment (TUG test) for evaluating physical frailty, defined by the Short Physical Performance Battery (SPPB), using machine learning algorithms. Methods This cross-sectional study included patients aged >60 years with independent ambulation and no mental or neurological impairment. A comprehensive set of parameters associated with physical frailty were assessed, including body composition, questionnaires (European Quality of Life 5-dimension [EQ 5D 5L], SARC-F), and physical performance tests (SPPB, TUG), along with digital sensor insole gait parameters collected during the TUG test. Physical frailty was defined as an SPPB score≤8. Advanced statistics, including random forest (RF) feature selection and machine learning algorithms (K-nearest neighbor [KNN] and RF) were used to compare the diagnostic value of these parameters to identify patients with physical frailty. Results Classified by the SPPB, 23 of the 57 eligible patients were defined as having physical frailty. Several gait parameters were significantly different between the two groups (with and without physical frailty). The area under the receiver operating characteristic curve (AUROC) of the TUG test was superior to that of the SARC-F (0.862 vs 0.639). The recursive feature elimination algorithm identified 9 parameters, 8 of which were digital insole gait parameters. Both the KNN and RF algorithms trained with these parameters resulted in excellent results (AUROC of 0.801 and 0.919, respectively). Conclusions A gait analysis based on machine learning algorithms using sensor soles is superior to the SARC-F and the TUG test to identify physical frailty in orthogeriatric patients.

Effect of Walking Adaptability on an Uneven Surface by a Stepping Pattern on Walking Activity After Stroke

Real-world walking activity is important for poststroke patients because it leads to their participation in the community and physical activity. Walking activity may be related to adaptability to different surface conditions of the ground. The purpose of this study was to clarify whether walking adaptability on an uneven surface by step is related to daily walking activity in patients after stroke. We involved 14 patients who had hemiparesis after stroke (age: 59.4 ± 8.9 years; post-onset duration: 70.7 ± 53.5 months) and 12 healthy controls (age: 59.5 ± 14.2 years). The poststroke patients were categorized as least limited community ambulators or unlimited ambulators. For the uneven surface, the study used an artificial grass surface (7 m long, 2-cm leaf length). The subjects repeated even surface walking and the uneven surface walking trials at least two times at a comfortable speed. We collected spatiotemporal and kinematic gait parameters on both the even and uneven surfaces using a three-dimensional motion analysis system. After we measured gait, the subjects wore an accelerometer around the waist for at least 4 days. We measured the number of steps per day using the accelerometer to evaluate walking activity. Differences in gait parameters between the even and uneven surfaces were calculated to determine how the subjects adapted to an uneven surface while walking. We examined the association between the difference in parameter measurements between the two surface properties and walking activity (number of steps per day). Walking activity significantly and positively correlated with the difference in paretic step length under the conditions of different surface properties in the poststroke patients (r = 0.65, p = 0.012) and step width in the healthy controls (r = 0.68, p = 0.015). The strategy of increasing the paretic step length, but not step width, on an uneven surface may lead to a larger base of support, which maintains stability during gait on an uneven surface in poststroke patients, resulting in an increased walking activity. Therefore, in poststroke patients, an increase in paretic step length during gait on an uneven surface might be more essential for improving walking activity.

Method for Estimating Temporal Gait Parameters Concerning Bilateral Lower Limbs of Healthy Subjects Using a Single In-Shoe Motion Sensor through a Gait Event Detection Approach

To expand the potential use of in-shoe motion sensors (IMSs) in daily healthcare or activity monitoring applications for healthy subjects, we propose a real-time temporal estimation method for gait parameters concerning bilateral lower limbs (GPBLLs) that uses a single IMS and is based on a gait event detection approach. To validate the established methods, data from 26 participants recorded by an IMS and a reference 3D motion analysis system were compared. The agreement between the proposed method and the reference system was evaluated by the intraclass correlation coefficient (ICC). The results showed that, by averaging over five continuous effective strides, all time parameters achieved precisions of no more than 30 ms and agreement at the “excellent” level, and the symmetry indexes of the stride time and stance phase time achieved precisions of 1.0% and 3.0%, respectively, and agreement at the “good” level. These results suggest our method is effective and shows promise for wide use in many daily healthcare or activity monitoring applications for healthy subjects.

Application of vibration to the soles reduces minimum toe clearance variability during walking

Minimum toe clearance (MTC) is an important indicator of the risk of tripping. Aging and neuromuscular diseases often decrease MTC height and increase its variability, leading to a higher risk of tripping. Previous studies have developed visual feedback-based gait training systems to modify MTC. However, these systems are bulky and expensive, and the effects of the training continue only for a short time. We paid attention to the efficacy of vibration in decreasing the variability of gait parameters, and hypothesized that proper vibration applied to soles can reduce the MTC variability. Using shoes embedded with active vibrating insoles, we assessed the efficacy of both sub- and supra-threshold vibration in affecting MTC distribution. Experiment results with 17 young and healthy adults showed that vibration applied throughout the walking task with constant intensity of 130% of sensory threshold significantly decreased MTC variability, whereas sub-threshold vibration yielded no significant effect. These results demonstrate that a properly designed tactile sensory input which is controlled and delivered by a simple wearable device, the active insole, can reduce the MTC variability during walking.

Cardiovascular risk factors, living and ageing in Halle: the CARLA study

The CARLA study (Cardiovascular Disease, Living and Ageing in Halle) is a longitudinal population-based cohort study of the general population of the city of Halle (Saale), Germany. The primary aim of the cohort was to investigate risk factors for cardiovascular diseases based on comprehensive cardiological phenotyping of study participants and was extended to study factors associated with healthy ageing. In total, 1779 probands (812 women and 967 men, aged 45–83 years) were examined at baseline (2002–2005), with a first and second follow-up performed 4 and 8 years later. The response proportion at baseline was 64.1% and the reparticipation proportion for the first and second follow-up was 86% and 77% respectively. Sixty-four percent of the study participants were in retirement while 25% were full- or partially-employed and 11% were unemployed at the time of the baseline examination. The currently running third follow-up focuses on the assessment of physical and mental health, with an intensive 4 h examination program, including measurement of cardiovascular, neurocognitive, balance and gait parameters. The data collected in the CARLA Study resulted in answering various research questions in over 80 publications, of which two thirds were pooled analyses with other similar population-based studies. Due to the extensiveness of information on risk factors, subclinical conditions and evident diseases, the biobanking concept for the biosamples, the cohort representativeness of an elderly population, and the high level of quality assurance, the CARLA cohort offers a unique platform for further research on important indicators for healthy ageing.

Path Tracking of an Underwater Snake Robot and Locomotion Efficiency Optimization Based on Improved Pigeon-Inspired Algorithm

This paper considers the tracking control of curved paths for an underwater snake robot, and investigates the methods used to improve energy efficiency. Combined with the path-planning method based on PCSI (parametric cubic-spline interpolation), an improved LOS (light of sight) method is proposed to design the controller and guide the robot to move along the desired path. The evaluation of the energy efficiency of robot locomotion is discussed. In particular, a pigeon-inspired optimization algorithm improved by quantum rules (QPIO) is proposed for dynamically selecting the gait parameters that maximize energy efficiency. Simulation results show that the proposed controller enables the robot to accurately follow the curved path and that the QPIO algorithm is effective in improving robot energy efficiency.

Role of Artificial Neural Network for Prediction of Gait Parameters and Patterns

Walking is very important exercise. Walking is characterized by gait. Gait defines the bipedal and forward propulsion of center of gravity of the human body. This chapter describes the role of artificial neural network (ANN) for prediction of gait parameters and patterns for human locomotion. The artificial neural network is a mathematical model. It is computational system inspired by the structure, processing method, and learning ability of a biological brain. According to bio-mechanics perspective, the neural system is utilized to check the non-direct connections between datasets. Also, ANN model in gait application is more desired than bio-mechanics strategies or statistical methods. It produces models of gait patterns, predicts horizontal ground reactions forces (GRF), vertical GRF, recognizes examples of stand, and predicts incline speed and distance of walking.