An official website of the United States government
Falls are the second leading cause of unintentional injury deaths worldwide. While numerous wearable fall detection devices incorporating AI models have been developed, none of them are used successfully in a fall detection application running on commodity-based smartwatches in real time. The system misses some falls, and generates an annoying amount of False Positives for practical use. We have investigated and experimented with an LSTM model for fall detection on a smartwatch. Even though the LSTM model has high accuracy during offline testing, the good performance of offline LSTM models cannot be translated to the equivalence of real-time performance. Transformers, on the other hand, can learn long-sequence data and patterns intrinsic to the data due to their self-attention mechanism. This paper compares three variants of LSTM and two variants of Transformer models for learning fall patterns. We trained all models using fall and activity data from three datasets, and the real-time testing of the model was performed using the SmartFall App. Our findings showed that in the offline training, the CNN-LSTM model was better than the Transformer model for all the datasets. However, the Transformer is a preferable choice for deployment in real-time fall detection applications.
more »
« less
Time-Critical Fall Prediction Based on Lipschitz Data Analysis and Design of a Reconfigurable Walker for Preventing Fall Injuries
Falls may cause serious injuries to older adults, leading to a deteriorated quality of life. Currently, there is a lack of fall injury prevention devices, especially for the balance impaired population who rely on mobility aids. Here, the functionality of a walker is augmented, so that it can predict a fall in real-time and prevent fall injuries via a rapidly reconfigurable mechanism. A key challenge is real-time fall prediction, which is a time-critical decision making process. A fall must be predicted preemptively so that the system has sufficient time to deploy the injury prevention mechanism. Data are collected from human subjects undergoing diverse loss-of-balance situations while using a walker. A predictor based on multiple Long-Short Term Memory (LSTM) networks is constructed based on three novel techniques. First, diverse fall types are identified by separately learning fast and slow falls. Second, a "Timer LSTM" is constructed that estimates the time remaining before an imbalance is unrecoverable and the fall prevention mechanism must be activated. Then if time allows, additional data are collected and the possibility of a fall is further examined. This approach lowered the fall prediction false positive rate. Third, confounding cases are further analyzed using a metric of data deficiency, called the Lipschitz quotient. Additional data features that lower the Lipschitz quotients and, thereby, increase data predictability, are sought and incorporated into the original input signals. Augmenting the data further improved performance, and the best model had a 97% success rate at identifying falls at a 0.17% false positive rate. The prediction method is implemented on a novel walkertype fall prediction and prevention prototype. The walker has a small footprint for improved maneuverability, and becomes untippable when its expandable legs are deployed in the event of a predicted fall. Thus, the older adult tethered to the untippable walker is protected from a fall. This work introduces techniques to improve the performance of real-time fall predictors when limited data is available, identifies how to select fruitful features from the available sensor signals, and incorporates a fall predictor into a physical device for a rapid fall injury prevention response. This promises immense benefits for future research on improving older adult wellbeing through real-time fall protection.
more »
« less
- Award ID(s):
- 2133072
- PAR ID:
- 10492964
- Editor(s):
- Derek Abbott
- Publisher / Repository:
- IEEE
- Date Published:
- Journal Name:
- IEEE Access
- Volume:
- 12
- ISSN:
- 2169-3536
- Page Range / eLocation ID:
- 1822 to 1838
- Subject(s) / Keyword(s):
- Fall prediction, Machine Learning, Mobility Aid, Fall Injury Prevention, Lipschitz Analysis, Eldercare, Assistive robots, Time-to-Event Prediction
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
It is often challenging to pick suitable data features for learning problems. Sometimes certain regions of the data are harder to learn because they are not well characterized by the selected data features. The challenge is amplified when resources for sensing and computation are limited and time-critical, yet reliable decisions must be made. For example, a robotic system for preventing falls of elderly people needs a real-time fall predictor, with low false positive and false negative rates, using a simple wearable sensor to activate a fall prevention mechanism. Here we present a methodology for assessing the learnability of data based on the Lipschitz quotient.We develop a procedure for determining which regions of the dataset contain adversarial data points, input data that look similar but belong to different target classes. Regardless of the learning model, it will be hard to learn such data. We then present a method for determining which additional feature(s) are most effective in improving the predictability of each of these regions. This is a model-independent data analysis that can be executed before constructing a prediction model through machine learning or other techniques. We demonstrate this method on two synthetic datasets and a dataset of human falls, which uses inertial measurement unit signals. For the fall dataset, we identified two groups of adversarial data points and improved the predictability of each group over the baseline dataset, as assessed by Lipschitz, by using 2 different sets of features. This work offers a valuable tool for assessing data learnability that can be applied to not only fall prediction problems, but also other robotics applications that learn from data.more » « less
-
Gaugler, Joseph E (Ed.)Abstract Background and ObjectivesFalls are an increasing problem among older adults. Older adults’ self-report of falls is the primary method of fall risk identification. However, up to 72% of Medicare beneficiaries who have fallen do not report falls and fall-related injuries to their healthcare providers. Research suggests that older adults prefer the term “balance problems” instead of “fall risk.” The purpose of this study was to examine if perceived balance problem is a predictor of self-reported falls after controlling for known predictors of falls among older adults. Research Design and MethodsThe Health Belief Model served as the theoretical framework. A longitudinal secondary analysis was conducted using data from a subsample of independently living participants (N = 5,446) from the National Health and Aging Trends Study. Baseline data was from the year 2015, and the outcome was self-reported falls in 2016. ResultsComplex samples multiple logistic regression analyses revealed that the single item perceived balance problem question (odds ratios, OR = 1.69, p < .001) predicted falls in 2016, whereas, the balance performance measure, Short Physical Performance Battery, did not (OR = 0.98, p = .06). Non-Hispanic White participants were more likely to report falling compared to non-Hispanic Black and Hispanic participants, as were females compared to males. A hospital stay in 2015, comorbidities, fear of falling, and a fall in 2015 were also predictive of falls. Discussion and ImplicationsAssessing older adults’ perceived balance is important in primary care to identify fall risk and recommend appropriate home modifications, assistive devices, and/or interventions.more » « less
-
Fall prevention has always been a crucial topic for injury prevention. Research shows that real-time posture monitoring and subsequent fall prevention are important for the prevention of fall-related injuries. In this research, we determine a real-time posture classifier by comparing classical and deep machine learning classifiers in terms of their accuracy and robustness for posture classification. For this, multiple classical classifiers, including classical machine learning, support vector machine, random forest, neural network, and Adaboost methods, were used. Deep learning methods, including LSTM and transformer, were used for posture classification. In the experiment, joint data were obtained using an RGBD camera. The results show that classical machine learning posture classifier accuracy was between 75% and 99%, demonstrating that the use of classical machine learning classification alone is sufficient for real-time posture classification even with missing joints or added noise. The deep learning method LSTM was also effective in classifying the postures with high accuracy, despite incurring a significant computational overhead cost, thus compromising the real-time posture classification performance. The research thus shows that classical machine learning methods are worthy of our attention, at least, to consider for reuse or reinvention, especially for real-time posture classification tasks. The insight of using a classical posture classifier for large-scale human posture classification is also given through this research.more » « less
-
Abstract AimsTo examine how perceived balance problems are associated with self‐reported falls in the past month after controlling for known correlates of falls among older adults. BackgroundApproximately 30% of adults age 65 and older fall each year. Most accidental falls are preventable, and older adults' engagement in fall prevention is imperative. Limited research suggest that older adults do not use the term ‘fall risk’ to describe their risk for falls. Instead, they commonly use the term ‘balance problems’. Yet, commonly used fall risk assessment tools in both primary and acute care do not assess older adults' perceived balance. Design and MethodThe Health Belief Model and the concept of perceived susceptibility served as the theoretical framework. A retrospective, cross‐sectional secondary analysis using data from the National Health and Aging Trends Study from year 2015 was conducted. The outcome variable was self‐reported falls in the last month. ResultsA subsample of independently living participants (N = 7499) was selected, and 10.3% of the sample reported a fall. Multiple logistic regression analysis revealed that the odds of reporting a fall in the past month was 3.4 times (p < .001) greater for participants who self‐reported having a balance problem compared to those who did not. In contrast, fear of falling and perceived memory problems were not uniquely associated with falls. Using a mobility device, reporting pain, poor self‐rated health status, depression and anxiety scores were also associated with falling. Conclusion and ImplicationsOlder adults' perceived balance problem is strongly associated with their fall risk. Perceived balance may be important to discuss with older adults to increase identification of fall risk. Older adults' perceived balance should be included in nursing fall risk assessments and fall prevention interventions. A focus on balance may increase older adults' engagement in fall prevention.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Journal Article:
- https://doi.org/10.1109/ACCESS.2023.3347263
