Deep Learning to Predict Falls in Older Adults Based on Daily-Life Trunk Accelerometry

Accurate fall risk classification in elderly using one gait cycle data and machine learning

BACKGROUND

Falls among the elderly are a major societal problem. While observations of medium-distance walking using inertial sensors identified potential fall predictors, classifying individuals at risk based on single gait cycles remains elusive. This challenge stems from individual variability and step-to-step fluctuations, making accurate classification difficult.

METHODS

We recruited 44 participants, equally divided into high and low fall-risk groups. A smartphone secured on their second sacral spinous process recorded data during indoor walking. Features were extracted at each gait cycle from a 6-dimensional time series (tri-axial angular velocity and tri-axial acceleration) and classified using the gradient boosting decision tree algorithm.

FINDINGS

Mean accuracy across five-fold cross-validation was 0.936. "Age" was the most influential individual feature, while features related to acceleration in the gait direction held the highest total relative importance when aggregated by axis (0.5365).

INTERPRETATION

Combining acceleration, angular velocity data, and the gradient boosting decision tree algorithm enabled accurate fall risk classification in the elderly, previously challenging due to lack of discernible features. We reveal the first-ever identification of three-dimensional pelvic motion characteristics during single gait cycles in the high-risk group. This novel method, requiring only one gait cycle, is valuable for individuals with physical limitations hindering repetitive or long-distance walking or for use in spaces with limited walking areas. Additionally, utilizing readily available smartphones instead of dedicated equipment has potential to improve gait analysis accessibility.

The Applications of Artificial Intelligence for Assessing Fall Risk: Systematic Review

BACKGROUND

Falls and their consequences are a serious public health problem worldwide. Each year, 37.3 million falls requiring medical attention occur. Therefore, the analysis of fall risk is of great importance for prevention. Artificial intelligence (AI) represents an innovative tool for creating predictive statistical models of fall risk through data analysis.

OBJECTIVE

The aim of this review was to analyze the available evidence on the applications of AI in the analysis of data related to postural control and fall risk.

METHODS

A literature search was conducted in 6 databases with the following inclusion criteria: the articles had to be published within the last 5 years (from 2018 to 2024), they had to apply some method of AI, AI analyses had to be applied to data from samples consisting of humans, and the analyzed sample had to consist of individuals with independent walking with or without the assistance of external orthopedic devices.

RESULTS

We obtained a total of 3858 articles, of which 22 were finally selected. Data extraction for subsequent analysis varied in the different studies: 82% (18/22) of them extracted data through tests or functional assessments, and the remaining 18% (4/22) of them extracted through existing medical records. Different AI techniques were used throughout the articles. All the research included in the review obtained accuracy values of >70% in the predictive models obtained through AI.

CONCLUSIONS

The use of AI proves to be a valuable tool for creating predictive models of fall risk. The use of this tool could have a significant socioeconomic impact as it enables the development of low-cost predictive models with a high level of accuracy.

TRIAL REGISTRATION

PROSPERO CRD42023443277; https://tinyurl.com/4sb72ssv.

Development and internal validation of a dynamic fall risk prediction and monitoring tool in aged care using routinely collected electronic health data: a landmarking approach

OBJECTIVES

Falls pose a significant challenge in residential aged care facilities (RACFs). Existing falls prediction tools perform poorly and fail to capture evolving risk factors. We aimed to develop and internally validate dynamic fall risk prediction models and create point-based scoring systems for residents with and without dementia.

MATERIALS AND METHODS

A longitudinal cohort study using electronic data from 27 RACFs in Sydney, Australia. The study included 5492 permanent residents, with a 70%-30% split for training and validation. The outcome measure was the incidence of falls. We tracked residents for 60 months, using monthly landmarks with 1-month prediction windows. We employed landmarking dynamic prediction for model development, a time-dependent area under receiver operating characteristics curve (AUROCC) for model evaluations, and a regression coefficient approach to create point-based scoring systems.

RESULTS

The model identified 15 independent predictors of falls in dementia and 12 in nondementia cohorts. Falls history was the key predictor of subsequent falls in both dementia (HR 4.75, 95% CI, 4.45-5.06) and nondementia cohorts (HR 4.20, 95% CI, 3.87-4.57). The AUROCC across landmarks ranged from 0.67 to 0.87 for dementia and from 0.66 to 0.86 for nondementia cohorts but generally remained between 0.75 and 0.85 in both cohorts. The total point risk score ranged from -2 to 57 for dementia and 0 to 52 for nondementia cohorts.

DISCUSSION

Our novel risk prediction models and scoring systems provide timely person-centered information for continuous monitoring of fall risk in RACFs.

CONCLUSION

Embedding these tools within electronic health records could facilitate the implementation of targeted proactive interventions to prevent falls.

Artificial intelligence based on falling in older people: A bibliometric analysis

Objectives

This study aimed to analyze publications on artificial intelligence (AI) for falls in older people from a bibliometric perspective.

Methods

The Web of Science database was searched for titles of English-language articles containing the words "artificial intelligence," "deep learning," "machine learning," "natural language processing,", "neural artificial network," "fall," "geriatric," "elderly," "aging," "older," and "old age." An R-based application (Biblioshiny for bibliometrics) and VOSviewer software were used for analysis.

Results

Thirty-seven English articles published between 2018 and 2024 were included. The year 2023 is the year with the most publications with 16 articles. The most productive research field was "Engineering Electrical Electronic" with seven articles. The most productive country was the United States, followed by China. The most common words were "injuries," "people," and "risk factors."

Conclusion

Publications on AI and falls in the elderly are both few in number and the number of publications has increased in recent years. Future research should include relevant analyses in scientific databases, such as Scopus and PubMed.

Empirical comparison of deep learning models for fNIRS pain decoding

Introduction

Pain assessment is extremely important in patients unable to communicate and it is often done by clinical judgement. However, assessing pain using observable indicators can be challenging for clinicians due to the subjective perceptions, individual differences in pain expression, and potential confounding factors. Therefore, the need for an objective pain assessment method that can assist medical practitioners. Functional near-infrared spectroscopy (fNIRS) has shown promising results to assess the neural function in response of nociception and pain. Previous studies have explored the use of machine learning with hand-crafted features in the assessment of pain.

Methods

In this study, we aim to expand previous studies by exploring the use of deep learning models Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and (CNN-LSTM) to automatically extract features from fNIRS data and by comparing these with classical machine learning models using hand-crafted features.

Results

The results showed that the deep learning models exhibited favourable results in the identification of different types of pain in our experiment using only fNIRS input data. The combination of CNN and LSTM in a hybrid model (CNN-LSTM) exhibited the highest performance (accuracy = 91.2%) in our problem setting. Statistical analysis using one-way ANOVA with Tukey's (post-hoc) test performed on accuracies showed that the deep learning models significantly improved accuracy performance as compared to the baseline models.

Discussion

Overall, deep learning models showed their potential to learn features automatically without relying on manually-extracted features and the CNN-LSTM model could be used as a possible method of assessment of pain in non-verbal patients. Future research is needed to evaluate the generalisation of this method of pain assessment on independent populations and in real-life scenarios.

The underuse of AI in the health sector: Opportunity costs, success stories, risks and recommendations

Purpose

This contribution explores the underuse of artificial intelligence (AI) in the health sector, what this means for practice, and how much the underuse can cost. Attention is drawn to the relevance of an issue that the European Parliament has outlined as a "major threat" in 2020. At its heart is the risk that research and development on trusted AI systems for medicine and digital health will pile up in lab centers without generating further practical relevance. Our analysis highlights why researchers, practitioners and especially policymakers, should pay attention to this phenomenon.

Methods

The paper examines the ways in which governments and public agencies are addressing the underuse of AI. As governments and international organizations often acknowledge the limitations of their own initiatives, the contribution explores the causes of the current issues and suggests ways to improve initiatives for digital health.

Results

Recommendations address the development of standards, models of regulatory governance, assessment of the opportunity costs of underuse of technology, and the urgency of the problem.

Conclusions

The exponential pace of AI advances and innovations makes the risks of underuse of AI increasingly threatening.

Graphical Abstract

Association of Prospective Falls in Older People With Ubiquitous Step-Based Fall Risk Parameters Calculated From Ambulatory Inertial Signals: Secondary Data Analysis

Background

In recent years, researchers have been advocating for the integration of ambulatory gait monitoring as a complementary approach to traditional fall risk assessments. However, current research relies on dedicated inertial sensors that are fixed on a specific body part. This limitation impacts the acceptance and adoption of such devices.

Objective

Our study objective is twofold: (1) to propose a set of step-based fall risk parameters that can be obtained independently of the sensor placement by using a ubiquitous step detection method and (2) to evaluate their association with prospective falls.

Methods

A reanalysis was conducted on the 1-week ambulatory inertial data from the StandingTall study, which was originally described by Delbaere et al. The data were from 301 community-dwelling older people and contained fall occurrences over a 12-month follow-up period. Using the ubiquitous and robust step detection method Smartstep, which is agnostic to sensor placement, a range of step-based fall risk parameters can be calculated based on walking bouts of 200 steps. These parameters are known to describe different dimensions of gait (ie, variability, complexity, intensity, and quantity). First, the correlation between parameters was studied. Then, the number of parameters was reduced through stepwise backward elimination. Finally, the association of parameters with prospective falls was assessed through a negative binomial regression model using the area under the curve metric.

Results

The built model had an area under the curve of 0.69, which is comparable to models exclusively built on fixed sensor placement. A higher fall risk was noted with higher gait variability (coefficient of variance of stride time), intensity (cadence), and quantity (number of steps) and lower gait complexity (sample entropy and fractal exponent).

Conclusions

These findings highlight the potential of our method for comprehensive and accurate fall risk assessments, independent of sensor placement. This approach has promising implications for ambulatory gait monitoring and fall risk monitoring using consumer-grade devices.

CNN-Based Self-Attention Weight Extraction for Fall Event Prediction Using Balance Test Score

Injury, hospitalization, and even death are common consequences of falling for elderly people. Therefore, early and robust identification of people at risk of recurrent falling is crucial from a preventive point of view. This study aims to evaluate the effectiveness of an interpretable semi-supervised approach in identifying individuals at risk of falls by using the data provided by ankle-mounted IMU sensors. Our method benefits from the cause-effect link between a fall event and balance ability to pinpoint the moments with the highest fall probability. This framework also has the advantage of training on unlabeled data, and one can exploit its interpretation capacities to detect the target while only using patient metadata, especially those in relation to balance characteristics. This study shows that a visual-based self-attention model is able to infer the relationship between a fall event and loss of balance by attributing high values of weight to moments where the vertical acceleration component of the IMU sensors exceeds 5 m/s² during an especially short period. This semi-supervised approach uses interpretable features to highlight the moments of the recording that may explain the score of balance, thus revealing the moments with the highest risk of falling. Our model allows for the detection of 71% of the possible falling risk events in a window of 1 s (500 ms before and after the target) when compared with threshold-based approaches. This type of framework plays a paramount role in reducing the costs of annotation in the case of fall prevention when using wearable devices. Overall, this adaptive tool can provide valuable data to healthcare professionals, and it can assist them in enhancing fall prevention efforts on a larger scale with lower costs.

Automation in health care: the need for an ergonomics-based approach

Healthcare quality and efficiency challenges degrade outcomes and burden multiple stakeholders. Workforce shortage, burnout, and complexity of workflows necessitate effective support for patients and providers. There is interest in employing automation, or the use of 'computer[s] [to] carry out… functions that the human operator would normally perform', in health care to improve delivery of services. However, unique aspects of health care require analysis of workflows across several domains and an understanding of the ways work system factors interact to shape those workflows. Ergonomics has identified key work system issues relevant to effective automation in other industries. Understanding these issues in health care can direct opportunities for the effective use of automation in health care. This article illustrates work system considerations using two example workflows; discusses how those considerations may inform solution design, implementation, and use; and provides future directions to advance the essential role of ergonomics in healthcare automation.

Human Gait Activity Recognition Using Multimodal Sensors

Human activity recognition is an application of machine learning with the aim of identifying activities from the gathered activity raw data acquired by different sensors. In medicine, human gait is commonly analyzed by doctors to detect abnormalities and determine possible treatments for the patient. Monitoring the patient's activity is paramount in evaluating the treatment's evolution. This type of classification is still not enough precise, which may lead to unfavorable reactions and responses. A novel methodology that reduces the complexity of extracting features from multimodal sensors is proposed to improve human activity classification based on accelerometer data. A sliding window technique is used to demarcate the first dominant spectral amplitude, decreasing dimensionality and improving feature extraction. In this work, we compared several state-of-art machine learning classifiers evaluated on the HuGaDB dataset and validated on our dataset. Several configurations to reduce features and training time were analyzed using multimodal sensors: all-axis spectrum, single-axis spectrum, and sensor reduction.

Application of machine learning approaches in predicting clinical outcomes in older adults - a systematic review and meta-analysis

BACKGROUND

Machine learning-based prediction models have the potential to have a considerable positive impact on geriatric care.

DESIGN

Systematic review and meta-analyses.

PARTICIPANTS

Older adults (≥ 65 years) in any setting.

INTERVENTION

Machine learning models for predicting clinical outcomes in older adults were evaluated. A random-effects meta-analysis was conducted in two grouped cohorts, where the predictive models were compared based on their performance in predicting mortality i) under and including 6 months ii) over 6 months.

OUTCOME MEASURES

Studies were grouped into two groups by the clinical outcome, and the models were compared based on the area under the receiver operating characteristic curve metric.

RESULTS

Thirty-seven studies that satisfied the systematic review criteria were appraised, and eight studies predicting a mortality outcome were included in the meta-analyses. We could only pool studies by mortality as there were inconsistent definitions and sparse data to pool studies for other clinical outcomes. The area under the receiver operating characteristic curve from the meta-analysis yielded a summary estimate of 0.80 (95% CI: 0.76 - 0.84) for mortality within 6 months and 0.81 (95% CI: 0.76 - 0.86) for mortality over 6 months, signifying good discriminatory power.

CONCLUSION

The meta-analysis indicates that machine learning models display good discriminatory power in predicting mortality. However, more large-scale validation studies are necessary. As electronic healthcare databases grow larger and more comprehensive, the available computational power increases and machine learning models become more sophisticated; there should be an effort to integrate these models into a larger research setting to predict various clinical outcomes.

Evaluating the Potential of Machine Learning and Wearable Devices in End-of-Life Care in Predicting 7-Day Death Events Among Patients With Terminal Cancer: Cohort Study

BACKGROUND

An accurate prediction of mortality in end-of-life care is crucial but presents challenges. Existing prognostic tools demonstrate moderate performance in predicting survival across various time frames, primarily in in-hospital settings and single-time evaluations. However, these tools may fail to capture the individualized and diverse trajectories of patients. Limited evidence exists regarding the use of artificial intelligence (AI) and wearable devices, specifically among patients with cancer at the end of life.

OBJECTIVE

This study aimed to investigate the potential of using wearable devices and AI to predict death events among patients with cancer at the end of life. Our hypothesis was that continuous monitoring through smartwatches can offer valuable insights into the progression of patients at the end of life and enable the prediction of changes in their condition, which could ultimately enhance personalized care, particularly in outpatient or home care settings.

METHODS

This prospective study was conducted at the National Taiwan University Hospital. Patients diagnosed with cancer and receiving end-of-life care were invited to enroll in wards, outpatient clinics, and home-based care settings. Each participant was given a smartwatch to collect physiological data, including steps taken, heart rate, sleep time, and blood oxygen saturation. Clinical assessments were conducted weekly. The participants were followed until the end of life or up to 52 weeks. With these input features, we evaluated the prediction performance of several machine learning-based classifiers and a deep neural network in 7-day death events. We used area under the receiver operating characteristic curve (AUROC), F1-score, accuracy, and specificity as evaluation metrics. A Shapley additive explanations value analysis was performed to further explore the models with good performance.

RESULTS

From September 2021 to August 2022, overall, 1657 data points were collected from 40 patients with a median survival time of 34 days, with the detection of 28 death events. Among the proposed models, extreme gradient boost (XGBoost) yielded the best result, with an AUROC of 96%, F1-score of 78.5%, accuracy of 93%, and specificity of 97% on the testing set. The Shapley additive explanations value analysis identified the average heart rate as the most important feature. Other important features included steps taken, appetite, urination status, and clinical care phase.

CONCLUSIONS

We demonstrated the successful prediction of patient deaths within the next 7 days using a combination of wearable devices and AI. Our findings highlight the potential of integrating AI and wearable technology into clinical end-of-life care, offering valuable insights and supporting clinical decision-making for personalized patient care. It is important to acknowledge that our study was conducted in a relatively small cohort; thus, further research is needed to validate our approach and assess its impact on clinical care.

TRIAL REGISTRATION

ClinicalTrials.gov NCT05054907; https://classic.clinicaltrials.gov/ct2/show/NCT05054907.

The Reliability and Accuracy of a Fall Risk Assessment Procedure Using Mobile Smartphone Sensors Compared with a Physiological Profile Assessment

Falls in older people are a major health concern as the leading cause of disability and the second most common cause of accidental death. We developed a rapid fall risk assessment based on a combination of physical performance measurements made with an inertial sensor embedded in a smartphone. This study aimed to evaluate and validate the reliability and accuracy of an easy-to-use smartphone fall risk assessment by comparing it with the Physiological Profile Assessment (PPA) results. Sixty-five participants older than 55 performed a variation of the Timed Up and Go test using smartphone sensors. Balance and gait parameters were calculated, and their reliability was assessed by the (ICC) and compared with the PPAs. Since the PPA allows classification into six levels of fall risk, the data obtained from the smartphone assessment were categorised into six equivalent levels using different parametric and nonparametric classifier models with neural networks. The F1 score and geometric mean of each model were also calculated. All selected parameters showed ICCs around 0.9. The best classifier, in terms of accuracy, was the nonparametric mixed input data model with a 100% success rate in the classification category. In conclusion, fall risk can be reliably assessed using a simple, fast smartphone protocol that allows accurate fall risk classification among older people and can be a useful screening tool in clinical settings.

Preventing falls: the use of machine learning for the prediction of future falls in individuals without history of fall

Nowadays, it becomes of paramount societal importance to support many frail-prone groups in our society (elderly, patients with neurodegenerative diseases, etc.) to remain socially and physically active, maintain their quality of life, and avoid their loss of autonomy. Once older people enter the prefrail stage, they are already likely to experience falls whose consequences may accelerate the deterioration of their quality of life (injuries, fear of falling, reduction of physical activity). In that context, detecting frailty and high risk of fall at an early stage is the first line of defense against the detrimental consequences of fall. The second line of defense would be to develop original protocols to detect future fallers before any fall occur. This paper briefly summarizes the current advancements and perspectives that may arise from the combination of affordable and easy-to-use non-wearable systems (force platforms, 3D tracking motion systems), wearable systems (accelerometers, gyroscopes, inertial measurement units-IMUs) with appropriate machine learning analytics, as well as the efforts to address these challenges.

Machine learning versus binomial logistic regression analysis for fall risk based on SPPB scores in older adult outpatients

Objective

To compare the performance of the diagnostic model for fall risk based on the short physical performance battery (SPPB) developed using commercial machine learning software (MLS) and binomial logistic regression analysis (BLRA).

Methods

We enrolled 797 out of 850 outpatients who visited the clinic between March 2016 and November 2021. Patients were categorized into the development (n = 642) and validation (n = 155) datasets. Age, sex, number of comorbidities, number of medications, body mass index (BMI), calf circumference (left-right average), handgrip strength (left-right average), total SPPB score, and history of falls were determined. We defined fall risk by an SPPB score of ≤6 in men and ≤9 in women. The main metrics used for evaluating the machine learning model and BLRA were the area under the curve (AUC), accuracy, precision, recall (sensitivity), specificity, and F-measure. The commercial MLS automatically calculates the parameter range of the highest contribution.

Results

The participants included 797 outpatients (mean age, 76.3 years; interquartile range, 73.0-81.0; 288 men). The metrics of the current diagnostic model in the commercial MLS were as follows: AUC = 0.78, accuracy = 0.74, precision = 0.46, recall (sensitivity) = 0.81, specificity = 0.71, F-measure = 0.59. The metrics of the current diagnostic model in the BLRA were as follows: AUC = 0.77, accuracy = 0.75, precision = 0.47, recall (sensitivity) = 0.67, specificity = 0.77, F-measure = 0.55. The risk factors for falls in older adult outpatients were handgrip strength, female sex, experience of falls, BMI, and calf circumference in the commercial MLS.

Conclusions

The diagnostic model for fall risk based on SPPB scores constructed using commercial MLS is noninferior to BLRA.

Open-source dataset reveals relationship between walking bout duration and fall risk classification performance in persons with multiple sclerosis

Falls are frequent and associated with morbidity in persons with multiple sclerosis (PwMS). Symptoms of MS fluctuate, and standard biannual clinical visits cannot capture these fluctuations. Remote monitoring techniques that leverage wearable sensors have recently emerged as an approach sensitive to disease variability. Previous research has shown that fall risk can be identified from walking data collected by wearable sensors in controlled laboratory conditions however this data may not be generalizable to variable home environments. To investigate fall risk and daily activity performance from remote data, we introduce a new open-source dataset featuring data collected from 38 PwMS, 21 of whom are identified as fallers and 17 as non-fallers based on their six-month fall history. This dataset contains inertial-measurement-unit data from eleven body locations collected in the laboratory, patient-reported surveys and neurological assessments, and two days of free-living sensor data from the chest and right thigh. Six-month (n = 28) and one-year repeat assessment (n = 15) data are also available for some patients. To demonstrate the utility of these data, we explore the use of free-living walking bouts for characterizing fall risk in PwMS, compare these data to those collected in controlled environments, and examine the impact of bout duration on gait parameters and fall risk estimates. Both gait parameters and fall risk classification performance were found to change with bout duration. Deep learning models outperformed feature-based models using home data; the best performance was observed with all bouts for deep-learning and short bouts for feature-based models when evaluating performance on individual bouts. Overall, short duration free-living walking bouts were found to be the least similar to laboratory walking, longer duration free-living walking bouts provided more significant differences between fallers and non-fallers, and an aggregation of all free-living walking bouts yields the best performance in fall risk classification.

Automated assessment of balance: A neural network approach based on large-scale balance function data

Balance impairment (BI) is an important cause of falls in the elderly. However, the existing balance estimation system needs to measure a large number of items to obtain the balance score and balance level, which is less efficient and redundant. In this context, we aim at building a model to automatically predict the balance ability, so that the early screening of large-scale physical examination data can be carried out quickly and accurately. We collected and sorted out 17,541 samples, each with 61-dimensional features and two labels. Moreover, using this data a lightweight artificial neural network model was trained to accurately predict the balance score and balance level. On the premise of ensuring high prediction accuracy, we reduced the input feature dimension of the model from 61 to 13 dimensions through the recursive feature elimination (RFE) algorithm, which makes the evaluation process more streamlined with fewer measurement items. The proposed balance prediction method was evaluated on the test set, in which the determination coefficient (R2) of balance score reaches 92.2%. In the classification task of balance level, the metrics of accuracy, area under the curve (AUC), and F1 score reached 90.5, 97.0, and 90.6%, respectively. Compared with other competitive machine learning models, our method performed best in predicting balance capabilities, which is especially suitable for large-scale physical examination.

Empirical mode decomposition using deep learning model for financial market forecasting

Financial market forecasting is an essential component of financial systems; however, predicting financial market trends is a challenging job due to noisy and non-stationary information. Deep learning is renowned for bringing out excellent abstract features from the huge volume of raw data without depending on prior knowledge, which is potentially fascinating in forecasting financial transactions. This article aims to propose a deep learning model that autonomously mines the statistical rules of data and guides the financial market transactions based on empirical mode decomposition (EMD) with back-propagation neural networks (BPNN). Through the characteristic time scale of data, the intrinsic wave pattern was obtained and then decomposed. Financial market transaction data were analyzed, optimized using PSO, and predicted. Combining the nonlinear and non-stationary financial time series can improve prediction accuracy. The predictive model of deep learning, based on the analysis of the massive financial trading data, can forecast the future trend of financial market price, forming a trading signal when particular confidence is satisfied. The empirical results show that the EMD-based deep learning model has an excellent predicting performance.

A Systematic Review of Wearable Sensor-Based Technologies for Fall Risk Assessment in Older Adults

Falls have been recognized as the major cause of accidental death and injury in people aged 65 and above. The timely prediction of fall risks can help identify older adults prone to falls and implement preventive interventions. Recent advancements in wearable sensor-based technologies and big data analysis have spurred the development of accurate, affordable, and easy-to-use approaches to fall risk assessment. The objective of this study was to systematically assess the current state of wearable sensor-based technologies for fall risk assessment among community-dwelling older adults. Twenty-five of 614 identified research articles were included in this review. A comprehensive comparison was conducted to evaluate these approaches from several perspectives. In general, these approaches provide an accurate and effective surrogate for fall risk assessment. The accuracy of fall risk prediction can be influenced by various factors such as sensor location, sensor type, features utilized, and data processing and modeling techniques. Features constructed from the raw signals are essential for predictive model development. However, more investigations are needed to identify distinct, clinically interpretable features and develop a general framework for fall risk assessment based on the integration of sensor technologies and data modeling.

Wearable Sensor Systems for Fall Risk Assessment: A Review

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.

Development of a machine learning model for the prediction of the short-term mortality in patients in the intensive care unit

PURPOSE

The aim of this study was to develop and evaluate a machine learning model that predicts short-term mortality in the intensive care unit using the trends of four easy-to-collect vital signs.

MATERIALS AND METHODS

The primary training cohort included 1968 patients at the Veterans Health Service Medical Center. The external validation cohort comprised 409 patients at Seoul National University Hospital. Datasets of heart rate, systolic blood pressure, diastolic blood pressure, and peripheral capillary oxygen saturation (SpO2) measured every hour for 10 h were used. The performances of mortality prediction models generated using five machine learning algorithms, Random Forest (RF), XGboost, perceptron, convolutional neural network, and Long Short-Term Memory, were calculated and compared using area under the receiver operating characteristic curve (AUROC) values and an external validation dataset.

RESULTS

The machine learning model generated using the RF algorithm showed the best performance. Its AUROC was 0.922, which is much better than the 0.8408 of the Acute Physiology and Chronic Health Evaluation II. The machine learning model developed using SpO2 showed the best performance (AUROC, 0.89).

CONCLUSIONS

This simple yet powerful new mortality prediction model could be useful for early detection of probable mortality and appropriate medical intervention, especially in rapidly deteriorating patients.

Improved screening of fall risk using free-living based accelerometer data

Falls are one of the most costly population health issues. Screening of older adults for fall risks can allow for earlier interventions and ultimately lead to better outcomes and reduced public health spending. This work proposes a solution to limitations in existing fall screening techniques by utilizing a hip-based accelerometer worn in free-living conditions. The work proposes techniques to extract fall risk features from periods of free-living ambulatory activity. Analysis of the proposed techniques is conducted and compared with existing screening methods using Functional Tests and Lab-based Gait Analysis. 1705 Older Adults from Umea (Sweden) were assessed. Data consisted of 1 Week of hip worn accelerometer data, gait measurements and performance metrics for 3 functional tests. Retrospective and Prospective fall data were also recorded based on the incidence of falls occurring 12 months before and after the study commencing respectively. Machine learning based experiments show accelerometer based measures perform best when predicting falls. Prospective falls had a sensitivity and specificity of 0.61 and 0.66 respectively while retrospective falls had a sensitivity and specificity of 0.61 and 0.68 respectively.

Towards Secure and Intelligent Internet of Health Things: A Survey of Enabling Technologies and Applications

With the growth of computing and communication technologies, the information processing paradigm of the healthcare environment is evolving. The patient information is stored electronically, making it convenient to store and retrieve patient information remotely when needed. However, evolving the healthcare systems into smart healthcare environments comes with challenges and additional pressures. Internet of Things (IoT) connects things, such as computing devices, through wired or wireless mediums to form a network. There are numerous security vulnerabilities and risks in the existing IoT-based systems due to the lack of intrinsic security technologies. For example, patient medical data, data privacy, data sharing, and convenience are considered imperative for collecting and storing electronic health records (EHR). However, the traditional IoT-based EHR systems cannot deal with these paradigms because of inconsistent security policies and data access structures. Blockchain (BC) technology is a decentralized and distributed ledger that comes in handy in storing patient data and encountering data integrity and confidentiality challenges. Therefore, it is a viable solution for addressing existing IoT data security and privacy challenges. BC paves a tremendous path to revolutionize traditional IoT systems by enhancing data security, privacy, and transparency. The scientific community has shown a variety of healthcare applications based on artificial intelligence (AI) that improve health diagnosis and monitoring practices. Moreover, technology companies and startups are revolutionizing healthcare with AI and related technologies. This study illustrates the implication of integrated technologies based on BC, IoT, and AI to meet growing healthcare challenges. This research study examines the integration of BC technology with IoT and analyzes the advancements of these innovative paradigms in the healthcare sector. In addition, our research study presents a detailed survey on enabling technologies for the futuristic, intelligent, and secure internet of health things (IoHT). Furthermore, this study comprehensively studies the peculiarities of the IoHT environment and the security, performance, and progression of the enabling technologies. First, the research gaps are identified by mapping security and performance benefits inferred by the BC technologies. Secondly, practical issues related to the integration process of BC and IoT devices are discussed. Third, the healthcare applications integrating IoT, BC, and ML in healthcare environments are discussed. Finally, the research gaps, future directions, and limitations of the enabling technologies are discussed.

Near-Fall Detection in Unexpected Slips during Over-Ground Locomotion with Body-Worn Sensors among Older Adults

Slip-induced falls are a growing health concern for older adults, and near-fall events are associated with an increased risk of falling. To detect older adults at a high risk of slip-related falls, this study aimed to develop models for near-fall event detection based on accelerometry data collected by body-fixed sensors. Thirty-four healthy older adults who experienced 24 laboratory-induced slips were included. The slip outcomes were first identified as loss of balance (LOB) and no LOB (NLOB), and then the kinematic measures were compared between these two outcomes. Next, all the slip trials were split into a training set (90%) and a test set (10%) at sample level. The training set was used to train both machine learning models (n = 2) and deep learning models (n = 2), and the test set was used to evaluate the performance of each model. Our results indicated that the deep learning models showed higher accuracy for both LOB (>64%) and NLOB (>90%) classifications than the machine learning models. Among all the models, the Inception model showed the highest classification accuracy (87.5%) and the largest area under the receiver operating characteristic curve (AUC), indicating that the model is an effective method for near-fall (LOB) detection. Our approach can be helpful in identifying individuals at the risk of slip-related falls before they experience an actual fall.

Initial Responses to False Positives in AI-Supported Continuous Interactions: A Colonoscopy Case Study

The use of artificial intelligence (AI) in clinical support systems is increasing. In this article, we focus on AI support for continuous interaction scenarios. A thorough understanding of end-user behaviour during these continuous human-AI interactions, in which user input is sustained over time and during which AI suggestions can appear at any time, is still missing. We present a controlled lab study involving 21 endoscopists and an AI colonoscopy support system. Using a custom-developed application and an off-the-shelf videogame controller, we record participants’ navigation behaviour and clinical assessment across 14 endoscopic videos. Each video is manually annotated to mimic an AI recommendation, being either true positive or false positive in nature. We find that time between AI recommendation and clinical assessment is significantly longer for incorrect assessments. Further, the type of medical content displayed significantly affects decision time. Finally, we discover that the participant’s clinical role plays a large part in the perception of clinical AI support systems. Our study presents a realistic assessment of the effects of imperfect and continuous AI support in a clinical scenario.

A Survey of IoT-Based Fall Detection for Aiding Elderly Care: Sensors, Methods, Challenges and Future Trends

Remote monitoring of a fall condition or activities and daily life (ADL) of elderly patients has become one of the essential purposes for modern telemedicine. Internet of Things (IoT) and artificial intelligence (AI) techniques, including machine and deep learning models, have been recently applied in the medical field to automate the diagnosis procedures of abnormal and diseased cases. They also have many other applications, including the real-time identification of fall accidents in elderly patients. The goal of this article is to review recent research whose focus is to develop AI algorithms and methods of fall detection systems (FDS) in the IoT environment. In addition, the usability of different sensor types, such as gyroscopes and accelerometers in smartwatches, is described and discussed with the current limitations and challenges for realizing successful FDSs. The availability problem of public fall datasets for evaluating the proposed detection algorithms are also addressed in this study. Finally, this article is concluded by proposing advanced techniques such as lightweight deep models as one of the solutions and prospects of futuristic smart IoT-enabled systems for accurate fall detection in the elderly.

Fall Risk Assessment Using Wearable Sensors: A Narrative Review

Recently, fall risk assessment has been a main focus in fall-related research. Wearable sensors have been used to increase the objectivity of this assessment, building on the traditional use of oversimplified questionnaires. However, it is necessary to define standard procedures that will us enable to acknowledge the multifactorial causes behind fall events while tackling the heterogeneity of the currently developed systems. Thus, it is necessary to identify the different specifications and demands of each fall risk assessment method. Hence, this manuscript provides a narrative review on the fall risk assessment methods performed in the scientific literature using wearable sensors. For each identified method, a comprehensive analysis has been carried out in order to find trends regarding the most used sensors and its characteristics, activities performed in the experimental protocol, and algorithms used to classify the fall risk. We also verified how studies performed the validation process of the developed fall risk assessment systems. The identification of trends for each fall risk assessment method would help researchers in the design of standard innovative solutions and enhance the reliability of this assessment towards a homogeneous benchmark solution.

Reliability of IMU-Based Gait Assessment in Clinical Stroke Rehabilitation

Background: Gait is often impaired in people after stroke, restricting personal independence and affecting quality of life. During stroke rehabilitation, walking capacity is conventionally assessed by measuring walking distance and speed. Gait features, such as asymmetry and variability, are not routinely determined, but may provide more specific insights into the patient’s walking capacity. Inertial measurement units offer a feasible and promising tool to determine these gait features. Objective: We examined the test–retest reliability of inertial measurement units-based gait features measured in a two-minute walking assessment in people after stroke and while in clinical rehabilitation. Method: Thirty-one people after stroke performed two assessments with a test–retest interval of 24 h. Each assessment consisted of a two-minute walking test on a 14-m walking path. Participants were equipped with three inertial measurement units, placed at both feet and at the low back. In total, 166 gait features were calculated for each assessment, consisting of spatio-temporal (56), frequency (26), complexity (63), and asymmetry (14) features. The reliability was determined using the intraclass correlation coefficient. Additionally, the minimal detectable change and the relative minimal detectable change were computed. Results: Overall, 107 gait features had good–excellent reliability, consisting of 50 spatio-temporal, 8 frequency, 36 complexity, and 13 symmetry features. The relative minimal detectable change of these features ranged between 0.5 and 1.5 standard deviations. Conclusion: Gait can reliably be assessed in people after stroke in clinical stroke rehabilitation using three inertial measurement units.

Using Sensor Graphs for Monitoring the Effect on the Performance of the OTAGO Exercise Program in Older Adults

The OTAGO exercise program is effective in decreasing the risk for falls of older adults. This research investigated if there is an indication that the OTAGO exercise program has a positive effect on the capacity and as well as on the performance in mobility. We used the data of the 10-months observational OTAGO pilot study with 15 (m = 1, f = 14) (pre-)frail participants aged 84.60 y (SD: 5.57 y). Motion sensors were installed in the flats of the participants and used to monitor their activity as a surrogate variable for performance. We derived a weighted directed multigraph from the physical sensor network, subtracted the weights of one day from a baseline, and used the difference in percent to quantify the change in performance. Least squares was used to compute the overall progress of the intervention (n = 9) and the control group (n = 6). In accordance with previous studies, we found indication for a positive effect of the OTAGO program on the capacity in both groups. Moreover, we found indication that the OTAGO program reduces the decline in performance of older adults in daily living. However, it is too early to conclude causalities from our findings because the data was collected during a pilot study.

Automatic Recognition and Analysis of Balance Activity in Community-Dwelling Older Adults: Algorithm Validation

BACKGROUND

Clinical mobility and balance assessments identify older adults who have a high risk of falls in clinics. In the past two decades, sensors have been a popular supplement to mobility and balance assessment to provide quantitative information and a cost-effective solution in the community environment. Nonetheless, the current sensor-based balance assessment relies on manual observation or motion-specific features to identify motions of research interest.

OBJECTIVE

The objective of this study was to develop an automatic motion data analytics framework using signal data collected from an inertial sensor for balance activity analysis in community-dwelling older adults.

METHODS

In total, 59 community-dwelling older adults (19 males and 40 females; mean age = 81.86 years, SD 6.95 years) were recruited in this study. Data were collected using a body-worn inertial measurement unit (including an accelerometer and a gyroscope) at the L4 vertebra of each individual. After data preprocessing and motion detection via a convolutional long short-term memory (LSTM) neural network, a one-class support vector machine (SVM), linear discriminant analysis (LDA), and k-nearest neighborhood (k-NN) were adopted to classify high-risk individuals.

RESULTS

The framework developed in this study yielded mean accuracies of 87%, 86%, and 89% in detecting sit-to-stand, turning 360°, and stand-to-sit motions, respectively. The balance assessment classification showed accuracies of 90%, 92%, and 86% in classifying abnormal sit-to-stand, turning 360°, and stand-to-sit motions, respectively, using Tinetti Performance Oriented Mobility Assessment-Balance (POMA-B) criteria by the one-class SVM and k-NN.

CONCLUSIONS

The sensor-based approach presented in this study provided a time-effective manner with less human efforts to identify and preprocess the inertial signal and thus enabled an efficient balance assessment tool for medical professionals. In the long run, the approach may offer a flexible solution to relieve the community's burden of continuous health monitoring.

Sensor-Based Fall Risk Assessment: A Survey

Fall is a major problem leading to serious injuries in geriatric populations. Sensor-based fall risk assessment is one of the emerging technologies to identify people with high fall risk by sensors, so as to implement fall prevention measures. Research on this domain has recently made great progress, attracting the growing attention of researchers from medicine and engineering. However, there is a lack of studies on this topic which elaborate the state of the art. This paper presents a comprehensive survey to discuss the development and current status of various aspects of sensor-based fall risk assessment. Firstly, we present the principles of fall risk assessment. Secondly, we show knowledge of fall risk monitoring techniques, including wearable sensor based and non-wearable sensor based. After that we discuss features which are extracted from sensors in fall risk assessment. Then we review the major methods of fall risk modeling and assessment. We also discuss some challenges and promising directions in this field at last.

Comparing Machine Learning Methods to Improve Fall Risk Detection in Elderly with Osteoporosis from Balance Data

Falls are a multifactorial cause of injuries for older people. Subjects with osteoporosis are particularly vulnerable to falls. We study the performance of different computational methods to identify people with osteoporosis who experience a fall by analysing balance parameters. Balance parameters, from eyes open and closed posturographic studies, and prospective registration of falls were obtained from a sample of 126 community-dwelling older women with osteoporosis (age 74.3 ± 6.3) using World Health Organization Questionnaire for the study of falls during a follow-up of 2.5 years. We analyzed model performance to determine falls of every developed model and to validate the relevance of the selected parameter sets. The principal findings of this research were (1) models built using oversampling methods with either IBk (KNN) or Random Forest classifier can be considered good options for a predictive clinical test and (2) feature selection for minority class (FSMC) method selected previously unnoticed balance parameters, which implies that intelligent computing methods can extract useful information with attributes which otherwise are disregarded by experts. Finally, the results obtained suggest that Random Forest classifier using the oversampling method to balance the data independent of the set of variables used got the best overall performance in measures of sensitivity (>0.71), specificity (>0.18), positive predictive value (PPV >0.74), and negative predictive value (NPV >0.66) independent of the set of variables used. Although the IBk classifier was built with oversampling data considering information from both eyes opened and closed, using all variables got the best performance (sensitivity >0.81, specificity >0.19, PPV = 0.97, and NPV = 0.66).

Random Forest for Automatic Feature Importance Estimation and Selection for Explainable Postural Stability of a Multi-Factor Clinical Test

Falling is a common incident that affects the health of elder adults worldwide. Postural instability is one of the major contributors to this problem. In this study, we propose a supplementary method for measuring postural stability that reduces doctor intervention. We used simple clinical tests, including the timed-up and go test (TUG), short form berg balance scale (SFBBS), and short portable mental status questionnaire (SPMSQ) to measure different factors related to postural stability that have been found to increase the risk of falling. We attached an inertial sensor to the lower back of a group of elderly subjects while they performed the TUG test, providing us with a tri-axial acceleration signal, which we used to extract a set of features, including multi-scale entropy (MSE), permutation entropy (PE), and statistical features. Using the score for each clinical test, we classified our participants into fallers or non-fallers in order to (1) compare the features calculated from the inertial sensor data, and (2) compare the screening capabilities of the multifactor clinical test against each individual test. We use random forest to select features and classify subjects across all scenarios. The results show that the combination of MSE and statistic features overall provide the best classification results. Meanwhile, PE is not an important feature in any scenario in our study. In addition, a t-test shows that the multifactor test of TUG and BBS is a better classifier of subjects in this study.

Exploration of machine algorithms based on deep learning model and feature extraction

The study expects to solve the problems of insufficient labeling, high input dimension, and inconsistent task input distribution in traditional lifelong machine learning. A new deep learning model is proposed by combining feature representation with a deep learning algorithm. First, based on the theoretical basis of the deep learning model and feature extraction. The study analyzes several representative machine learning algorithms, and compares the performance of the optimized deep learning model with other algorithms in a practical application. By explaining the machine learning system, the study introduces two typical algorithms in machine learning, namely ELLA (Efficient lifelong learning algorithm) and HLLA (Hierarchical lifelong learning algorithm). Second, the flow of the genetic algorithm is described, and combined with mutual information feature extraction in a machine algorithm, to form a composite algorithm HLLA (Hierarchical lifelong learning algorithm). Finally, the deep learning model is optimized and a deep learning model based on the HLLA algorithm is constructed. When K = 1200, the classification error rate reaches 0.63%, which reflects the excellent performance of the unsupervised database algorithm based on this model. Adding the feature model to the updating iteration process of lifelong learning deepens the knowledge base ability of lifelong machine learning, which is of great value to reduce the number of labels required for subsequent model learning and improve the efficiency of lifelong learning.

Prediction of gait trajectories based on the Long Short Term Memory neural networks

The forecasting of lower limb trajectories can improve the operation of assistive devices and minimise the risk of tripping and balance loss. The aim of this work was to examine four Long Short Term Memory (LSTM) neural network architectures (Vanilla, Stacked, Bidirectional and Autoencoders) in predicting the future trajectories of lower limb kinematics, i.e. Angular Velocity (AV) and Linear Acceleration (LA). Kinematics data of foot, shank and thigh (LA and AV) were collected from 13 male and 3 female participants (28 ± 4 years old, 1.72 ± 0.07 m in height, 66 ± 10 kg in mass) who walked for 10 minutes at preferred walking speed (4.34 ± 0.43 km.h-1) and at an imposed speed (5km.h-1, 15.4% ± 7.6% faster) on a 0% gradient treadmill. The sliding window technique was adopted for training and testing the LSTM models with total kinematics time-series data of 10,500 strides. Results based on leave-one-out cross validation, suggested that the LSTM autoencoders is the top predictor of the lower limb kinematics trajectories (i.e. up to 0.1s). The normalised mean squared error was evaluated on trajectory predictions at each time-step and it obtained 2.82-5.31% for the LSTM autoencoders. The ability to predict future lower limb motions may have a wide range of applications including the design and control of bionics allowing improved human-machine interface and mitigating the risk of falls and balance loss.

Machine Learning in Aging: An Example of Developing Prediction Models for Serious Fall Injury in Older Adults

BACKGROUND

Advances in computational algorithms and the availability of large datasets with clinically relevant characteristics provide an opportunity to develop machine learning prediction models to aid in diagnosis, prognosis, and treatment of older adults. Some studies have employed machine learning methods for prediction modeling, but skepticism of these methods remains due to lack of reproducibility and difficulty in understanding the complex algorithms that underlie models. We aim to provide an overview of two common machine learning methods: decision tree and random forest. We focus on these methods because they provide a high degree of interpretability.

METHOD

We discuss the underlying algorithms of decision tree and random forest methods and present a tutorial for developing prediction models for serious fall injury using data from the Lifestyle Interventions and Independence for Elders (LIFE) study.

RESULTS

Decision tree is a machine learning method that produces a model resembling a flow chart. Random forest consists of a collection of many decision trees whose results are aggregated. In the tutorial example, we discuss evaluation metrics and interpretation for these models. Illustrated using data from the LIFE study, prediction models for serious fall injury were moderate at best (area under the receiver operating curve of 0.54 for decision tree and 0.66 for random forest).

CONCLUSIONS

Machine learning methods offer an alternative to traditional approaches for modeling outcomes in aging, but their use should be justified and output should be carefully described. Models should be assessed by clinical experts to ensure compatibility with clinical practice.

Latest Research Trends in Fall Detection and Prevention Using Machine Learning: A Systematic Review

Falls are unusual actions that cause a significant health risk among older people. The growing percentage of people of old age requires urgent development of fall detection and prevention systems. The emerging technology focuses on developing such systems to improve quality of life, especially for the elderly. A fall prevention system tries to predict and reduce the risk of falls. In contrast, a fall detection system observes the fall and generates a help notification to minimize the consequences of falls. A plethora of technical and review papers exist in the literature with a primary focus on fall detection. Similarly, several studies are relatively old, with a focus on wearables only, and use statistical and threshold-based approaches with a high false alarm rate. Therefore, this paper presents the latest research trends in fall detection and prevention systems using Machine Learning (ML) algorithms. It uses recent studies and analyzes datasets, age groups, ML algorithms, sensors, and location. Additionally, it provides a detailed discussion of the current trends of fall detection and prevention systems with possible future directions. This overview can help researchers understand the current systems and propose new methodologies by improving the highlighted issues.

Classical Machine Learning Versus Deep Learning for the Older Adults Free-Living Activity Classification

Physical activity has a strong influence on mental and physical health and is essential in healthy ageing and wellbeing for the ever-growing elderly population. Wearable sensors can provide a reliable and economical measure of activities of daily living (ADLs) by capturing movements through, e.g., accelerometers and gyroscopes. This study explores the potential of using classical machine learning and deep learning approaches to classify the most common ADLs: walking, sitting, standing, and lying. We validate the results on the ADAPT dataset, the most detailed dataset to date of inertial sensor data, synchronised with high frame-rate video labelled data recorded in a free-living environment from older adults living independently. The findings suggest that both approaches can accurately classify ADLs, showing high potential in profiling ADL patterns of the elderly population in free-living conditions. In particular, both long short-term memory (LSTM) networks and Support Vector Machines combined with ReliefF feature selection performed equally well, achieving around 97% F-score in profiling ADLs.

Reducing Fall-related Revisits for Elderly Diabetes Patients in Emergency Departments: A Transition Flow Model

This paper introduces a transition flow model to study fall-related emergency department (ED) revisits for elderly patients with diabetes. Five diabetes classes are used to classify patients at discharge, within 7-day revisits, and between 8 and 30-day revisits. Analytical formulas to evaluate patient revisiting risks are derived. To reduce revisits, sensitivity analysis is introduced to identify the most critical, i.e., dominant, factors whose changes can lead to the largest reduction in revisits. In addition, a case study at University of Wisconsin (UW) Hospital ED is described to illustrate the applicability of the model.

Wearables and Deep Learning Classify Fall Risk From Gait in Multiple Sclerosis

Falls are a significant problem for persons with multiple sclerosis (PwMS). Yet fall prevention interventions are not often prescribed until after a fall has been reported to a healthcare provider. While still nascent, objective fall risk assessments could help in prescribing preventative interventions. To this end, retrospective fall status classification commonly serves as an intermediate step in developing prospective fall risk assessments. Previous research has identified measures of gait biomechanics that differ between PwMS who have fallen and those who have not, but these biomechanical indices have not yet been leveraged to detect PwMS who have fallen. Moreover, they require the use of laboratory-based measurement technologies, which prevent clinical deployment. Here we demonstrate that a bidirectional long short-term (BiLSTM) memory deep neural network was able to identify PwMS who have recently fallen with good performance (AUC of 0.88) based on accelerometer data recorded from two wearable sensors during a one-minute walking task. These results provide substantial improvements over machine learning models trained on spatiotemporal gait parameters (21% improvement in AUC), statistical features from the wearable sensor data (16%), and patient-reported (19%) and neurologist-administered (24%) measures in this sample. The success and simplicity (two wearable sensors, only one-minute of walking) of this approach indicates the promise of inexpensive wearable sensors for capturing fall risk in PwMS.

Fall risk classification for people with lower extremity amputations using random forests and smartphone sensor features from a 6-minute walk test

Fall-risk classification is a challenging but necessary task to enable the recommendation of preventative programs for individuals identified at risk for falling. Existing research has primarily focused on older adults, with no predictive fall-risk models for lower limb amputees, despite their greater likelihood of fall-risk than older adults. In this study, 89 amputees with varying degrees of lower limb amputation were asked if they had fallen in the past 6 months. Those who reported at least one fall were considered a fall risk. Each participant performed a 6 minute walk test (6MWT) with an Android smartphone placed in a holder located on the back of the pelvis. A fall-risk classification method was developed using data from sensors within the smartphone. The Ottawa Hospital Rehabilitation Center Walk Test app captured accelerometer and gyroscope data during the 6MWT. From this data, foot strikes were identified, and 248 features were extracted from the collection of steps. Steps were segmented into turn and straight walking, and four different data sets were created: turn steps, straightaway steps, straightaway and turn steps, and all steps. From these, three feature selection techniques (correlation-based feature selection, relief F, and extra trees classifier ensemble) were used to eliminate redundant or ineffective features. Each feature subset was tested with a random forest classifier and optimized for the best number of trees. The best model used turn data, with three features selected by Correlation-based feature selection (CFS), and used 500 trees in a random forest classifier. The resulting metrics were 81.3% accuracy, 57.2% sensitivity, 94.9% specificity, a Matthews correlation coefficient of 0.587, and an F1 score of 0.83. Since the outcomes are comparable to metrics achieved by existing clinical tests, the classifier may be viable for use in clinical practice.

Investigating Group-Specific Models of Hospital Workers’ Well-Being: Implications for Algorithmic Bias

Hospital workers often experience burnout due to the demanding job responsibilities and long work hours. Data yielding from ambulatory monitoring combined with machine learning algorithms can afford us a better understanding of the naturalistic processes that contribute to this burnout. Motivated by the challenges related to the accurate tracking of well-being in real-life, prior work has investigated group-specific machine learning (GS-ML) models that are tailored to groups of participants. We examine a novel GS-ML for estimating well-being from real-life multimodal measures collected in situ from hospital workers. In contrast to the majority of prior work that uses pre-determined clustering criteria, we propose an iterative procedure that refines participant clusters based on the representations learned by the GS-ML models. Motivated by prior work that highlights the differential impact of job demands on well-being, we further explore the participant clusters in terms of demography and job-related attributes. Results indicate that the GS-ML models mostly outperform general models in estimating well-being constructs. The GS-ML models further depict different degrees of predictive power for each participant cluster, as distinguished upon age, education, occupational role, and number of supervisees. The observed discrepancies with respect to the GS-ML model decisions are discussed in association with algorithmic bias.

Fall risk assessment in the wild: A critical examination of wearable sensor use in free-living conditions

BACKGROUND

Despite advances in laboratory-based supervised fall risk assessment methods (FRAs), falls still remain a major public health problem. This can be due to the alteration of behavior in laboratory due to the awareness of being observed (i.e., Hawthorne effect), the multifactorial complex etiology of falls, and our limited understanding of human behaviour in natural environments, or in the' wild'. To address these imitations, a growing body of literature has focused on free-living wearable-sensor-based FRAs. The objective of this narrative literature review is to discuss papers investigating natural data collected by wearable sensors for a duration of at least 24 h to identify fall-prone older adults.

METHODS

Databases (Scopus, PubMed and Google Scholar) were searched for studies based on a rigorous search strategy.

RESULTS

Twenty-four journal papers were selected, in which inertial sensors were the only wearable system employed for FRA in the wild. Gait was the most-investigated activity; but sitting, standing, lying, transitions and gait events, such as turns and missteps, were also explored. A multitude of free-living fall predictors (FLFPs), e.g., the quantity of daily steps, were extracted from activity bouts and events. FLFPs were further categorized into discrete domains (e.g., pace, complexity) defined by conceptual or data-driven models. Heterogeneity was found within the reviewed studies, which includes variance in: terminology (e.g., quantity vs macro), hyperparameters to define/estimate FLFPs, models and domains, and data processing approaches (e.g., the cut-off thresholds to define an ambulatory bout). These inconsistencies led to different results for similar FLFPs, limiting the ability to interpret and compare the evidence.

CONCLUSION

Free-living FRA is a promising avenue for fall prevention. Achieving a harmonized model is necessary to systematically address the inconsistencies in the field and identify FLFPs with the highest predictive values for falls to eventually address intervention programs and fall prevention.

A Deep Learning Approach for TUG and SPPB Score Prediction of (Pre-) Frail Older Adults on Real-Life IMU Data

Since older adults are prone to functional decline, using Inertial-Measurement-Units (IMU) for mobility assessment score prediction gives valuable information to physicians to diagnose changes in mobility and physical performance at an early stage and increases the chances of rehabilitation. This research introduces an approach for predicting the score of the Timed Up & Go test and Short-Physical-Performance-Battery assessment using IMU data and deep neural networks. The approach is validated on real-world data of a cohort of 20 frail or (pre-) frail older adults of an average of 84.7 years. The deep neural networks achieve an accuracy of about 95% for both tests for participants known by the network.

Time-Frequency Analysis of Daily Activities for Fall Detection

Fall events in elderly populations often result in serious injury and may lead to long-term disability and/or death. While many fall detection systems have been developed using wearable sensors to distinguish falls from other daily activities, detection sensitivity and specificity decreases when exposed to more rigorous activities such as running and jumping. This research uses time-frequency analysis of accelerometer-only activity data to develop a strategy for improving fall detection accuracy. In this study, a wireless sensor system (WSS) consisting of a three-axis accelerometer, microprocessor and wireless communications module is used to collect daily activities performed following a script in the laboratory setting. Experiments were conducted on 36 healthy human subjects performing four types of falls (i.e., forward, backward, and left/right sideway falls) as well as normal movements such as standing, walking, stand-to-sit, sit-to-stand, stepping, running and jumping. In total, 1227 different activities were collected and analyzed. The developed algorithm computes the magnitude of three-axis accelerometer data to detect if a critical fall threshold is passed, then analyzes the power spectral density within a critical fall duration window (500 ms) to differentiate fall events from other rigorous activities. Fall events were observed with high energy in the 2–3.5 Hz range and distinct from other rigorous activities such as running (3.5–5.5 Hz) and jumping (1–2 Hz). Preliminary results indicate the power spectral density (PSD)-based algorithm can detect falls with high sensitivity (98.4%) and specificity (98.6%) using lab-based daily activity data. The proposed algorithm has the benefit of improved accuracy over existing time-domain only strategies and multisensor strategies.

Human Activity Recognition using Inertial, Physiological and Environmental Sensors: A Comprehensive Survey

In the last decade, Human Activity Recognition (HAR) has become a vibrant research area, especially due to the spread of electronic devices such as smartphones, smartwatches and video cameras present in our daily lives. In addition, the advance of deep learning and other machine learning algorithms has allowed researchers to use HAR in various domains including sports, health and well-being applications. For example, HAR is considered as one of the most promising assistive technology tools to support elderly's daily life by monitoring their cognitive and physical function through daily activities. This survey focuses on critical role of machine learning in developing HAR applications based on inertial sensors in conjunction with physiological and environmental sensors.

Foreseeing future falls with accelerometer features in active community-dwelling older persons with no recent history of falls

BACKGROUND

Acceleration sensors are a viable option for monitoring gait patterns and its application on monitoring falls and risk of falling. However the literature still lacks prospective studies to investigate such risk before the occurrence of falls.

OBJECTIVE

To investigate features extracted from accelerometer signals with the purpose of predicting future falls in individuals with no recent history of falls.

METHODS

In this study we investigate the risk of fall in active and healthy community-dwelling living older persons with no recent history of falls, using a single accelerometer and variants of the Timed Up and Go (TUG) test. A prospective study was conducted with 74 healthy non-fallers older persons. After collecting acceleration data from the participants at the baseline, the occurrence of falls (outcome) was monitored quarterly during one year. A set of frequency features were extracted from the signal and their ability to predict falls was evaluated.

RESULTS

The best individual feature result shows an accuracy of 0.75, sensitivity of 0.71 and specificity of 0.76. A fusion of the three best features increases the sensitivity to 0.86. On the other hand, the cut-off points of the TUG seconds, often used to assess fall risk, did not demonstrate adequate sensitivity.

CONCLUSION

The results confirms previous evidence that accelerometer features can better estimate fall risk, and support potential applications that try to infer falls risk in less restricted scenarios, even in a sample stratified by age and gender composed of active and healthy community-dwelling living older persons.