An Advanced Integrated Sensor-Based Method for Fall Risk Assessment in Rehabilitation Setting

Mastery learning is a form of competency-based education in which learning time varies but outcomes are uniform. Trainees must meet a minimum passing standard (MPS) before completing a mastery learning curriculum. The objective of this study was to establish a curriculum for fall risk and gait assessment for medical students, determine an MPS for a fall risk and gait assessment clinical skills examination (CSE), and apply the MPS to a sample of medical students completing a fall risk and gait assessment CSE. Medical students completed an interactive session about fall risk and gait assessment including the Timed Up and Go (TUG) test and completed deliberate practice with 3 patients. Skills were evaluated using an 18-item skills checklist. A panel of clinical experts set the MPS at 82%. Eighty-seven medical students participated. The average score on the checklist was 14.7 of 18 (81.4%.) Although almost all performed the TUG correctly, only 61% met the MPS for the checklist. Our results suggest that a mastery learning approach may better prepare the 39% of students that did not meet MPS to complete a fall risk and gait assessment.

Strategies for fall risk assessment are currently not multifactorial neither implemented as a regular assessment of health status in clinics or hospital. The reason could be related with a lack of an easy to implement, complete and objective test to assess elderly's fall risk level. More recently, inertial wearable sensors have been used in combination with standard tests to evaluate the performance of the person during each phase of the test in an objective way. This paper proposes a methodology for collecting and analyzing the Timed-Up and Go (TUG) test instrumented with wearable inertial sensors. An automatic algorithm to segment the TUG test into three components was implemented prior to feature extraction. Overall, features from the walking and first turning phases of the tests could provide meaningful information to differentiate groups of high and low fall risk.

There are many attempts to provide the elderly with a life more independently. One of the main problems facing people in this age group is fall events. Falls are one of the most common accidents among the elderly and may result in extended hospitalization and increased medical costs. The requirement for care services, such as fall detection, is increasing because of the growing population of elderly people. In this study, machine learning techniques- Logistic Regression, Random Forest, and Decision Tree are used to predict fall risk of elderly people. Fall risk assessment methods are used to obtain inputs and outputs in addition to the physical and clinical features of people in the dataset. This study aimed to facilitate the fall risk assessment process of health professionals to determine the fall risk factors of elderly individuals and to make predictions. Based on the results of fall prediction, individualized fall prevention interventions can be developed to reduce the fall rates of elderly individuals.

Falls are caused by complex interaction between multiple risk factors which may be modified by age, disease and environment. A variety of methods and tools for fall risk assessment have been proposed, but none of which is universally accepted. Existing tools are generally not capable of providing a quantitative predictive assessment of fall risk. The need for objective, cost-effective and clinically applicable methods would enable quantitative assessment of fall risk on a subject-specific basis. Tracking objectively falls risk could provide timely feedback about the effectiveness of administered interventions enabling intervention strategies to be modified or changed if found to be ineffective. Moreover, some of the fundamental factors leading to falls and what actually happens during a fall remain unclear. Objectively documented and measured falls are needed to improve knowledge of fall in order to develop more effective prevention strategies and prolong independent living. In the last decade, several research groups have developed sensor-based automatic or semi-automatic fall risk assessment tools using wearable inertial sensors. This approach may also serve to detect falls. At the moment, i) several fall-risk assessment studies based on inertial sensors, even if promising, lack of a biomechanical model-based approach which could provide accurate and more detailed measurements of interests (e.g., joint moments, forces) and ii) the number of published real-world fall data of older people in a real-world environment is minimal since most authors have used simulations with healthy volunteers as a surrogate for real-world falls. With these limitations in mind, this thesis aims i) to suggest a novel method for the kinematics and dynamics evaluation of functional motor tasks, often used in clinics for the fall-risk evaluation, through a body sensor network and a biomechanical approach and ii) to define the guidelines for a fall detection algorithm based on a real-world fall database availability.

Agreement and predictive power of six fall risk assessment methods in community-dwelling older adults

Category: Other; Midfoot/Forefoot Introduction/Purpose: Charcot-Marie-Tooth (CMT) disease is an inherited peripheral neuropathy known to cause progressive sensory and motor impairment. Patients often develop foot structure abnormalities, including pes cavus, due to intrinsic muscle weakness. These deformities, paired with the disease’s typical sensory and motor symptoms, can increase the risk of falls. Ankle foot orthoses (AFOs) are commonly used with the intent of reducing fall risk and restoring a more normal gait. The Timed Up and Go (TUG) test is widely used as a clinical screening tool to identify patients at risk of falling. No previous study has applied the TUG test in the setting of CMT fall risk assessment. The purpose of this study was to assess the overall predictive value of TUG on fall risk in CMT. Methods: Existing data from patients seen at the at the University of Iowa Hospitals and Clinics, as part of research with the Inherited Neuropathy Consortium (INC), were used for this investigation. Patients in the INC return to clinic on an annual basis and data from each patient’s most recent evaluation were included for analysis. All CMT sub-types were included. Clinical outcome measures were taken from the CMT Functional Outcome Measure (CMT-FOM) and the CMT Health Index (CMT-HI). Timed up and go (TUG) and patient reported ankle-foot orthoses (AFO) use was recorded as part of the CMT-FOM. The CMT-HI provides patient perspective on the impact of various health related topics, including difficulty getting up from the floor or ground, falling, and difficulty rising from a seated position. Kruskal-Wallis and Wilcoxon Rank Sum tests were utilized for statistical analysis and pair-wise group comparisons. Results: In total, 126 patients of 4,971 available records were included for analysis. The remainder were excluded due to missing data, unconfirmed CMT diagnosis, or statistical outliers. TUG scores ranged from 3.4s to 26.0s (mean 9.70s). Responses to the CMT-HI Falls question included no effect, little effect, moderate effect, and severely effecting, and poorer responses were correlated with a slower average TUG (p< 0.0001). Participants who reported AFO use had a average TUG score of 11.29 seconds while non-users averaged 10.0 seconds (p=0.0015). 117/126 (93%) patients reported foot deformity: 106 pes cavus, 6 pes planus, 5 “other”, 7 reported none, and 2 unreported. Of the 117 with reported deformities, those with pes planus averaged a TUG of 9.76 seconds compared to 9.45 seconds with pes cavus. Conclusion: This is the first study to identify a significant correlation between patient-reported fall risk and average TUG in CMT patients. This observation of slower TUG times in individuals with greater fall risk is consistent with findings available for other at-risk falls populations, such as stroke patients. The most predominant deformity was pes cavus. However, neither the presence nor absence of foot deformities showed a substantial difference in TUG results. Patients who wore Ankle-foot orthoses had, on average, slower times in the TUG test than non-users. This is likely due to greater a dependency on AFOs with increased disease progression.

Falls are a severe problem in older adults, often resulting in severe consequences such as injuries or loss of consciousness. It is crucial to screen fall risk in order to prescribe appropriate therapies that can potentially prevent falls. Identifying individuals who have experienced falls in the past, commonly known as fallers, is used to evaluate fall risk, as a prior fall indicates a higher likelihood of future falls. The methods that have the most support from evidence are Gait Speed (GS) and Time Up and Go (TUG), which use specific cut-off values to evaluate the fall risk. There have been proposals for alternative methods that use wearable sensor technology to improve fall risk assessment. Although these technological alternatives are promising, further research is necessary to validate their use in clinical settings. In this study, we propose a method for identifying fallers based on a Support Vector Machine (SVM) classifier. The inputs for the classifier are the gait parameters obtained from a 30-minute walk recorded using an Inertial Measurement Unit (IMU) placed at the foot of patients. We validated our proposed method using a sample of 157 patients aged over 70 years. Our findings indicate significant differences (p<0.05) in stride speed, clearance, angular velocity, acceleration, and coefficient of variability among steps between fallers and non-fallers. The proposed method demonstrates the its potential to classify fallers with an accuracy of 79.6%, slightly outperforming the GS method which provides an accuracy of 77.0%, and also overcomes its dependency on the cut-off speed to determine fallers. This method could be valuable in detecting fallers during long-term monitoring that does not require periodic evaluations in a clinical setting.

Background: Falls are common and dangerous for stroke survivors. Current fall risk assessment methods rely on subjective scales. Objective sensor-based methods could improve prediction accuracy. Objective: Develop machine learning models using inertial sensors to objectively classify fall risk in stroke survivors. Determine optimal sensor configurations and clinical test protocols. Methods: 21 stroke survivors performed balance, Timed Up and Go, 10 Meter Walk, and Sit-to-Stand tests with and without dual-tasking. A total of 8 motion sensors captured lower limb and trunk kinematics, and 92 spatiotemporal gait and clinical features were extracted. Supervised models—Support Vector Machine, Logistic Regression, and Random Forest—were implemented to classify high vs. low fall risk. Sensor setups and test combinations were evaluated. Results: The Random Forest model achieved 91% accuracy using dual-task balance sway and Timed Up and Go walk time features. Single thorax sensor models performed similarly to multi-sensor models. Balance and Timed Up and Go best-predicted fall risk. Conclusion: Machine learning models using minimal inertial sensors during clinical assessments can accurately quantify fall risk in stroke survivors. Single thorax sensor setups are effective. Findings demonstrate a feasible objective fall screening approach to assist rehabilitation.

Fall-risk assessment of aged workers using wearable inertial measurement units based on machine learning

Fall risk assessment and fall detection are crucial for the prevention of adverse and long-term health outcomes. Wearable sensor systems have been used to assess fall risk and detect falls while providing additional meaningful information regarding gait characteristics. Commonly used wearable systems for this purpose are inertial measurement units (IMUs), which acquire data from accelerometers and gyroscopes. IMUs can be placed at various locations on the body to acquire motion data that can be further analyzed and interpreted. Insole-based devices are wearable systems that were also developed for fall risk assessment and fall detection. Insole-based systems are placed beneath the sole of the foot and typically obtain plantar pressure distribution data. Fall-related parameters have been investigated using inertial sensor-based and insole-based devices include, but are not limited to, center of pressure trajectory, postural stability, plantar pressure distribution and gait characteristics such as cadence, step length, single/double support ratio and stance/swing phase duration. The acquired data from inertial and insole-based systems can undergo various analysis techniques to provide meaningful information regarding an individual's fall risk or fall status. By assessing the merits and limitations of existing systems, future wearable sensors can be improved to allow for more accurate and convenient fall risk assessment. This article reviews inertial sensor-based and insole-based wearable devices that were developed for applications related to falls. This review identifies key points including spatiotemporal parameters, biomechanical gait parameters, physical activities and data analysis methods pertaining to recently developed systems, current challenges, and future perspectives.

BackgroundEarly detection of falls risk among older adults using simple tools may assist in fall prevention strategies. The aim of this study was to identify the best parameters associated with previous falls, either the timed up and go (TUG) test combined with sociodemographic factors and a self-rated multifactorial questionnaire (SRMQ) on falls risk or the TUG on its own. Falls risk was determined based on parameters associated with previous falls.DesignThis was a retrospective cohort study.SettingThe study was conducted in a community setting.ParticipantsThe participants were 1,086 community-dwelling older adults, with mean age of 69.6±5.6 years. Participants were categorized into fallers and nonfallers based on their history of falls in the past 12 months.MethodParticipants’ sociodemographic data was taken, and SRMQ consisting of five falls-related questions was administered. Participants performed the TUG test twice, and the mean was taken as the result.ResultsA total of 161 participants were categorized as fallers (14.8%). Multivariate logistic regression analysis showed that the model (χ2(6)=61.0, p<0.001, Nagelkerke R2=0.10) consisting of the TUG test, sociodemographic factors (gender, cataract/glaucoma and joint pain), as well as the SRMQ items “previous falls history” (Q1) and “worried of falls” (Q5), was more robust in terms of falls risk association compared to that with TUG on its own (χ2(1)=10.3, p<0.001, Nagelkerke R2=0.02).ConclusionCombination of sociodemographic factors and SRMQ with TUG is more favorable as an initial falls risk screening tool among community-dwelling older adults. Subsequently, further comprehensive falls risk assessment may be performed in clinical settings to identify the specific impairments for effective management.

The effectiveness of multifactorial fall risk assessment and intervention strategies is well documented. Although identifying feet/footwear-related influences on fall risk is a vital fall risk assessment component, few evidence-based resources or screening tools are available. To address this need, we developed the Screening Tool for Feet/Footwear-Related Influences on Fall Risk. Our tool is designed for older adults who are identified as at risk for falling, based on the CDC’s Stopping Elderly Accidents, Deaths, and Injuries (STEADI) Algorithm for Fall Risk Screening, Assessment, and Intervention. Tool development was informed by results of our systematic review of lower-limb factors associated with balance and falls. Our initial tool was evaluated by an external group of 9 interprofessional content experts. Those experts recommended modification of 8 items and rated the tool’s clarity as 81.2/100, appeal as 79.1/100, and clinical feasibility as 76.1/100. After incorporating recommended changes, we completed a modified Delphi study using 8 new interprofessional experts (average years of experience: 19.3). During Phase 1, Delphi participants recommended we combine items with similar treatment recommendations, add a question about orthoses, and increase the specificity of 9 items. This refinement resulted in a 20-item screening tool, which met approval after two rounds of consensus voting. Approval was defined based on the Item Content Validation Index, percentage of agreement > 80% on each item. The high level of agreement illustrates the tool’s content validity. Using our tool, an older adult’s feet/footwear-related risk factors can be identified and incorporated into an effective multifactorial fall prevention intervention.

Cognitive biases and contextual factors explaining variability in nurses’ fall risk judgements: a multi-centre cross-sectional study

Objective: To determine if a modified Timed Up and Go (TUG) test can be used to monitor motor improvement in patients with Parkinson's disease (PD) dependent on ambulatory assistance during a 2-week inpatient multimodal intensive neurorehabilitation and care (i-MINC) stay.Background: The TUG test is a widely used measure of functional mobility in the elderly and PD population.Its use is generally limited to patients able to walk independently.Methods: 76 PD patients (H&Y stage 3) were included.All required an assistive device and/or hands-on assistance on admission to i-MINC.They were assessed in their best ON state with a UPDRS examination and a modified TUG test allowing for ambulatory assistance at admission and prior to discharge.A paired T-test was used to analyze changes in TUG completion time and UPDRS scores, and a Wilcoxon signed-rank test for changes in TUG level of assistance (Table 1).The type of assistive device used during the test was recorded.Results: 75 out of 76 patients showed improvement in TUG completion time.76 patients improved in UPDRS total and motor scores.The average improvement in TUG completion time was 12.88 secs (p<0.0001),paralleled by a 16.84-point improvement in UPDRS total score and 6.41-point improvement in UPDRS motor score (p<0.0001).79% of patients required a reduced level of assistance at discharge (p<0.0001).Only 8 patients required a more restrictive assistive device.Conclusions: This study highlights the effectiveness of a modified TUG test as a marker for motor improvement in PD patients undergoing i-MINC.The joint reduction in TUG completion times, UPDRS total and motor scores strongly suggests that the modified test is a reliable and sensitive measure of motor function changes.Moreover, improvements in the level of assistance required during the test support the conclusion that i-MINC positively impacted functional mobility in the majority of patients.Adapting the TUG to accommodate patients dependent on ambulatory assistance creates a more inclusive assessment standard that can be used across the PD severity spectrum.Future studies may focus on formal reliability evaluation and application of the modified TUG test in other rehabilitation settings.

Fall risk assessment can be informed by understanding mobility/gait. Contemporary mobility analysis is being progressed by wearable inertial measurement units (IMU). Typically, IMUs gather temporal mobility-based outcomes (e.g., step time) from labs/clinics or beyond, capturing data for habitually informed fall risk. However, a thorough understanding of free-living IMU-based mobility is currently limited due to a lack of context. For example, although IMU-based length variability can be measured, no absolute clarity exists for factors relating to those variations, which could be due to an intrinsic or an extrinsic environmental factor. For a thorough understanding of habitual-based fall risk assessment through IMU-based mobility outcomes, use of wearable video cameras is suggested. However, investigating video data is laborious i.e., watching and manually labelling environments. Additionally, it raises ethical issues such as privacy. Accordingly, automated artificial intelligence (AI) approaches, that draw upon heterogenous datasets to accurately classify environments, are needed. Here, a novel dataset was created through mining online video and a deep learning-based tool was created via chained convolutional neural networks enabling automated environment (indoor or outdoor) and terrain (e.g., carpet, grass) classification. The dataset contained 146,624 video-based images (environment: 79,251, floor visible: 28,347, terrain: 39,026). Upon training each classifier, the system achieved F1-scores of ≥0.84 when tested on a manually labelled unseen validation dataset (environment: 0.98, floor visible indoor: 0.86, floor visible outdoor: 0.96, terrain indoor: 0.84, terrain outdoor: 0.95). Testing on new data resulted in accuracies from 51 to 100% for isolated networks and 45–90% for complete model. This work is ongoing with the underlying AI being refined for improved classification accuracies to aid automated contextual analysis of mobility/gait and subsequent fall risk. Ongoing work involves primary data capture from within participants free-living environments to bolster dataset heterogeneity.