Falling is a significant health problem. Fall detection, to alert for medical attention, has been gaining increasing attention. Still, most of the existing studies use falls simulated in a laboratory ...environment to test the obtained performance. We analyzed the acceleration signals recorded by an inertial sensor on the lower back during 143 real-world falls (the most extensive collection to date) from the FARSEEING repository. Such data were obtained from continuous real-world monitoring of subjects with a moderate-to-high risk of falling. We designed and tested fall detection algorithms using features inspired by a multiphase fall model and a machine learning approach. The obtained results suggest that algorithms can learn effectively from features extracted from a multiphase fall model, consistently overperforming more conventional features. The most promising method (support vector machines and features from the multiphase fall model) obtained a sensitivity higher than 80%, a false alarm rate per hour of 0.56, and an F-measure of 64.6%. The reported results and methodologies represent an advancement of knowledge on real-world fall detection and suggest useful metrics for characterizing fall detection systems for real-world use.
Falls are a common, serious threat to the health and self-confidence of the elderly. Assessment of fall risk is an important aspect of effective fall prevention programs.
In order to test whether it ...is possible to outperform current prognostic tools for falls, we analyzed 1010 variables pertaining to mobility collected from 976 elderly subjects (InCHIANTI study). We trained and validated a data-driven model that issues probabilistic predictions about future falls. We benchmarked the model against other fall risk indicators: history of falls, gait speed, Short Physical Performance Battery (Guralnik et al. 1994), and the literature-based fall risk assessment tool FRAT-up (Cattelani et al. 2015). Parsimony in the number of variables included in a tool is often considered a proxy for ease of administration. We studied how constraints on the number of variables affect predictive accuracy.
The proposed model and FRAT-up both attained the same discriminative ability; the area under the Receiver Operating Characteristic (ROC) curve (AUC) for multiple falls was 0.71. They outperformed the other risk scores, which reported AUCs for multiple falls between 0.64 and 0.65. Thus, it appears that both data-driven and literature-based approaches are better at estimating fall risk than commonly used fall risk indicators. The accuracy-parsimony analysis revealed that tools with a small number of predictors (~1-5) were suboptimal. Increasing the number of variables improved the predictive accuracy, reaching a plateau at ~20-30, which we can consider as the best trade-off between accuracy and parsimony. Obtaining the values of these ~20-30 variables does not compromise usability, since they are usually available in comprehensive geriatric assessments.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Highlights ► We compare a Smartphone-embedded accelerometer with a commercial ambulatory monitoring device. ► We evaluate the Smartphone validity on Timed Up and Go test parameters already used in ...the literature. ► We find good statistical agreement in most cases. ► It is crucial to define an appropriate pre-processing of Smartphone signals. ► Smartphones can provide suitable solutions for pervasive movement analysis.
Photoplethysmographic (PPG) signals are mainly employed for heart rate estimation but are also fascinating candidates in the search for cardiovascular biomarkers. However, their high susceptibility ...to motion artifacts can lower their morphological quality and, hence, affect the reliability of the extracted information. Low reliability is particularly relevant when signals are recorded in a real-world context, during daily life activities. We aim to develop two classifiers to identify PPG pulses suitable for heart rate estimation (Basic-quality classifier) and morphological analysis (High-quality classifier). We collected wrist PPG data from 31 participants over a 24 h period. We defined four activity ranges based on accelerometer data and randomly selected an equal number of PPG pulses from each range to train and test the classifiers. Independent raters labeled the pulses into three quality levels. Nineteen features, including nine novel features, were extracted from PPG pulses and accelerometer signals. We conducted ten-fold cross-validation on the training set (70%) to optimize hyperparameters of five machine learning algorithms and a neural network, and the remaining 30% was used to test the algorithms. Performances were evaluated using the full features and a reduced set, obtained downstream of feature selection methods. Best performances for both Basic- and High-quality classifiers were achieved using a Support Vector Machine (Acc: 0.96 and 0.97, respectively). Both classifiers outperformed comparable state-of-the-art classifiers. Implementing automatic signal quality assessment methods is essential to improve the reliability of PPG parameters and broaden their applicability in a real-world context.
Difficulty with turning is a major contributor to mobility disability and falls in people with movement disorders, such as Parkinson's disease (PD). Turning often results in freezing and/or falling ...in patients with PD. However, asking a patient to execute a turn in the clinic often does not reveal their impairments. Continuous monitoring of turning with wearable sensors during spontaneous daily activities may help clinicians and patients determine who is at risk of falls and could benefit from preventative interventions. In this study, we show that continuous monitoring of natural turning with wearable sensors during daily activities inside and outside the home is feasible for people with PD and elderly people. We developed an algorithm to detect and characterize turns during gait, using wearable inertial sensors. First, we validate the turning algorithm in the laboratory against a Motion Analysis system and against a video analysis of 21 PD patients and 19 control (CT) subjects wearing an inertial sensor on the pelvis. Compared to Motion Analysis and video, the algorithm maintained a sensitivity of 0.90 and 0.76 and a specificity of 0.75 and 0.65, respectively. Second, we apply the turning algorithm to data collected in the home from 12 PD and 18 CT subjects. The algorithm successfully detects turn characteristics, and the results show that, compared to controls, PD subjects tend to take shorter turns with smaller turn angles and more steps. Furthermore, PD subjects show more variability in all turn metrics throughout the day and the week.
Despite extensive preventive efforts, falls continue to be a major source of morbidity and mortality among elderly. Real-time detection of falls and their urgent communication to a telecare center ...may enable rapid medical assistance, thus increasing the sense of security of the elderly and reducing some of the negative consequences of falls. Many different approaches have been explored to automatically detect a fall using inertial sensors. Although previously published algorithms report high sensitivity (SE) and high specificity (SP), they have usually been tested on simulated falls performed by healthy volunteers. We recently collected acceleration data during a number of real-world falls among a patient population with a high-fall-risk as part of the SensAction-AAL European project. The aim of the present study is to benchmark the performance of thirteen published fall-detection algorithms when they are applied to the database of 29 real-world falls. To the best of our knowledge, this is the first systematic comparison of fall detection algorithms tested on real-world falls. We found that the SP average of the thirteen algorithms, was (mean ± std) 83.0% ± 30.3% (maximum value = 98%). The SE was considerably lower (SE = 57.0% ± 27.3%, maximum value = 82.8%), much lower than the values obtained on simulated falls. The number of false alarms generated by the algorithms during 1-day monitoring of three representative fallers ranged from 3 to 85. The factors that affect the performance of the published algorithms, when they are applied to the real-world falls, are also discussed. These findings indicate the importance of testing fall-detection algorithms in real-life conditions in order to produce more effective automated alarm systems with higher acceptance. Further, the present results support the idea that a large, shared real-world fall database could, potentially, provide an enhanced understanding of the fall process and the information needed to design and evaluate a high-performance fall detector.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
This paper describes the second part of the PASSO (Parkinson smart sensory cues for older users) project, which designs and tests an innovative haptic biofeedback system based on a wireless body ...sensor network using a smartphone and different smartwatches specifically designed to rehabilitate postural disturbances in persons with Parkinson’s disease. According to the scientific literature on the use of smart devices to transmit sensory cues, vibrotactile feedback (particularly on the trunk) seems promising for improving people’s gait and posture performance; they have been used in different environments and are well accepted by users. In the PASSO project, we designed and developed a wearable device and a related system to transmit vibrations to a person’s body to improve posture and combat impairments like Pisa syndrome and camptocormia. Specifically, this paper describes the methodologies and strategies used to design, develop, and test wearable prototypes and the mHealth system. The results allowed a multidisciplinary comparison among the solutions, which led to prototypes with a high degree of usability, wearability, accessibility, and effectiveness. This mHealth system is now being used in pilot trials with subjects with Parkinson’s disease to verify its feasibility among patients.
Clinicians need a practical, objective test of postural control that is sensitive to mild neurological disease, shows experimental and clinical validity, and has good test-retest reliability. We ...developed an instrumented test of postural sway (ISway) using a body-worn accelerometer to offer an objective and practical measure of postural control.
We conducted two separate studies with two groups of subjects. Study I: sensitivity and experimental concurrent validity. Thirteen subjects with early, untreated Parkinson's disease (PD) and 12 age-matched control subjects (CTR) were tested in the laboratory, to compare sway from force-plate COP and inertial sensors. Study II: test-retest reliability and clinical concurrent validity. A different set of 17 early-to-moderate, treated PD (tested ON medication), and 17 age-matched CTR subjects were tested in the clinic to compare clinical balance tests with sway from inertial sensors. For reliability, the sensor was removed, subjects rested for 30 min, and the protocol was repeated. Thirteen sway measures (7 time-domain, 5 frequency-domain measures, and JERK) were computed from the 2D time series acceleration (ACC) data to determine the best metrics for a clinical balance test.
Both center of pressure (COP) and ACC measures differentiated sway between CTR and untreated PD. JERK and time-domain measures showed the best test-retest reliability (JERK ICC was 0.86 in PD and 0.87 in CTR; time-domain measures ICC ranged from 0.55 to 0.84 in PD and from 0.60 to 0.89 in CTR). JERK, all but one time-domain measure, and one frequency measure were significantly correlated with the clinical postural stability score (r ranged from 0.50 to 0.63, 0.01 < p < 0.05).
Based on these results, we recommend a subset of the most sensitive, reliable, and valid ISway measures to characterize posture control in PD: 1) JERK, 2) RMS amplitude and mean velocity from the time-domain measures, and 3) centroidal frequency as the best frequency measure, as valid and reliable measures of balance control from ISway.
Although turning has been reported as one of the leading activities performed during a fall, and falls during turning result in 8-times more hip fractures than falls during linear gait, the quantity ...and quality of turns resulting in falls remain unknown since turns are rarely assessed during activities of daily living. 160 community-dwelling older adults were monitored for one week using smartphone technology. Turn measures and activity rates were quantified. Fall incidence within 12 months from continuous monitoring defined fall status, with 7/153 prospective fallers/non-fallers. Based on the analysis of 718,582 turns, prospective fallers turned less frequently, took longer to turn, and were less consistent in turn angle (p = 0.007, 0.025, and 0.038, respectively). Prospective fallers also walked slower and spent less time walking and turning and more time engaged in sedentary behavior (p = 0.043, 0.012, and 0.015, respectively). Individuals experiencing decline in the control of gait and/or turning may attempt to reduce their risk of falling by limiting their exposure and implementing cautionary movement strategies while turning. Since there was no difference in the overall active rate between prospective fallers and non-fallers, impaired gait and turning ability, specifically, may attribute to elevated fall risk within this cohort.
Highlights • An IMU-based method to detect APAs was validated in healthy and subjects with PD. • The size of APAs was smaller in subjects with PD compared to healthy subjects. • Test–retest ...reliability for the IMU-based method was moderate-to-excellent.