Wearable airbag technology and machine learned models to mitigate falls after stroke

 Didn't your hospital bring in this 7 years ago?

Hip protector saves you when you slip February 2015

 

Do you prefer your  hospital incompetence NOT KNOWING? OR NOT DOING?

I personally prefer massive perturbations as you walk  so you know the movement necessary to prevent falls. That would build your confidence in safely walking more that this airbag technology. But since I'm not medically trained and stroke-addled besides, you can't listen to me.

The latest here:

Wearable airbag technology and machine learned models to mitigate falls after stroke

• Olivia K. Botonis,
• Yaar Harari,
• Kyle R. Embry,
• Chaithanya K. Mummidisetty,
• David Riopelle,
• Matt Giffhorn,
• Mark V. Albert,
• Vallery Heike &
• Arun Jayaraman 

Journal of NeuroEngineering and Rehabilitation volume 19, Article number: 60 (2022) Cite this article

• 2 Altmetric

• Metrics details

Abstract

Background

Falls are a common complication experienced after a stroke and can cause serious detriments to physical health and social mobility, necessitating a dire need for intervention. Among recent advancements, wearable airbag technology has been designed to detect and mitigate fall impact. However, these devices have not been designed nor validated for the stroke population and thus, may inadequately detect falls in individuals with stroke-related motor impairments. To address this gap, we investigated whether population-specific training data and modeling parameters are required to pre-detect falls in a chronic stroke population.

Methods

We collected data from a wearable airbag’s inertial measurement units (IMUs) from individuals with (n = 20 stroke) and without (n = 15 control) history of stroke while performing a series of falls (842 falls total) and non-falls (961 non-falls total) in a laboratory setting. A leave-one-subject-out crossvalidation was used to compare the performance of two identical machine learned models (adaptive boosting classifier) trained on cohort-dependent data (control or stroke) to pre-detect falls in the stroke cohort.

Results

The average performance of the model trained on stroke data (recall = 0.905, precision = 0.900) had statistically significantly better recall (P = 0.0035) than the model trained on control data (recall = 0.800, precision = 0.944), while precision was not statistically significantly different. Stratifying models trained on specific fall types revealed differences in pre-detecting anterior–posterior (AP) falls (stroke-trained model’s F₁-score was 35% higher, P = 0.019). Using activities of daily living as non-falls training data (compared to near-falls) significantly increased the AUC (Area under the receiver operating characteristic) for classifying AP falls for both models (P < 0.04). Preliminary analysis suggests that users with more severe stroke impairments benefit further from a stroke-trained model. The optimal lead time (time interval pre-impact to detect falls) differed between control- and stroke-trained models.

Conclusions

These results demonstrate the importance of population sensitivity, non-falls data, and optimal lead time for machine learned pre-impact fall detection specific to stroke. Existing fall mitigation technologies should be challenged to include data of neurologically impaired individuals in model development to adequately detect falls in other high fall risk populations.

Trial registration https://clinicaltrials.gov/ct2/show/NCT05076565; Unique Identifier: NCT05076565. Retrospectively registered on 13 October 2021

Background

Every year, approximately 13 million individuals around the world experience a stroke [1, 2]. Falls are one of the most common medical complications experienced by individuals after a stroke, reported in up to 65% of the stroke population during hospitalization and up to 75% in the community [3, 4]. Individuals who have experienced a stroke are at an increased vulnerability for falling, related to common correlates of high fall risk in this population such as impaired mobility, medication use, and cognitive impairment [5]. Falling after a stroke can have serious consequences. There are a high incidence of severe physical injuries, including fractures, soft tissue and head injuries, and at worst, death [6, 7]. Psychologically, individuals often develop a fear of falling, leading to reduced mobility, increased social isolation, and significant reduction in quality of life [8, 9]. Financially speaking, fall related injuries constitute a burden on healthcare systems through prolonged use of services and incurred high healthcare costs [10,11,12]. Despite evidence that multifactorial rehabilitation approaches such as improving strength, balance, and visual impairments can reduce fall incidence in the older adult population [13], a recent Cochrane review concluded that there is little to no evidence of interventions that can prevent falls from occurring in individuals experiencing falls after stroke [14]. Therefore, individuals who suffer from mobility deficits after a stroke continue to experience falls, frequently and repeatedly. Without a way to prevent these falls from occurring, there is a compelling need to develop methods and tools to detect these falls before impact with the ground and reduce the associated consequences.

One conventional solution to achieve some degree of fall impact mitigation is wearing padded hip protectors in or underneath clothing, yet their significance in reducing fractures and associated injuries is limited and their current usage in the community is insignificant (likely due to discomfort and poor compliance) [15, 16]. A more novel and recent fall impact mitigation approach to address these concerns is wearable airbag technology [17,18,19,20]. These devices generally include three design components: (1) at least one sensor, such as an inertial measurement unit (IMU), to record user motion; (2) a computational model that processes the sensor signals to pre-detect fall impact; and (3) an inflatable airbag mechanism that deploys upon detection of a fall, to mitigate contact forces with the ground.

Despite continued development, wearable airbag technologies are currently only developed with the non-neurologically impaired older adult population in mind. Furthermore, the internal fall impact detection algorithms are often developed on data from young participants [21,22,23,24,25]. The algorithms have neither been specifically designed nor validated for detecting falls of individuals presenting with stroke-related motor impairments. A presumption is that the design and computational models should seamlessly transfer to users in the stroke population. However, the underlying pathophysiology of a stroke, fundamentally related to the cerebrovascular territory that is compromised in the brain, may manifest with alterations in movement kinematics and a loss of ability to control movements. These characteristic changes in movement have been observed and quantified in existing literature [26,27,28] and may translate to observed and measurable differences leading up to or during falls [29,30,31,32,33]. For example, earlier studies have analyzed and compared falls between older able-bodied and stroke individuals, and found significantly different motor responses including postural stability, trunk control, fall velocity, and timely step compensation [29, 31,32,33]. Furthermore, Dusane et al. found that within a stroke population, kinematic responses differed depending on the side of the body which a fall was initiated on (paretic vs. non-paretic) [34]. Given falls in stroke may have distinct kinematic profiles, current fall mitigation technology developed on data of generally healthy individuals might not be sufficiently sensitive or specific to detect falls in individuals who have experienced a stroke. Such inaccuracy could result in failure to deploy the airbag during a fall or cause unnecessary airbag deployments (i.e. false positives) and consequently, may lead to poor user engagement.

To address these issues, we suggest that systematic fall detection models should consider incorporating training data of individuals specific to the intended user population. Using machine learning to tune movement recognition algorithms to unique movements of particular mobility-impaired populations has been demonstrated in various applications for Parkinson’s disease [35], incomplete spinal cord injury [36, 37], and stroke [38], yet has not been applied to pre-impact fall detection in stroke populations. Thus, this paper presents considerations for a sensor-based, machine learned wearable airbag system to demonstrate the importance of pre-impact fall detection models specific to stroke-related movement impairments. We hypothesize that for fall detection in the stroke population, a pre-impact fall detection model trained on data from a stroke population would perform better than models trained on data from a control population. In other words, failure to train a model on fall movements specific to individuals with a history of stroke will result in decreased pre-impact detection performance for users of the stroke population. Furthermore, we explore secondary considerations for model development, including dataset activity composition, severity of gait impairments across users, and lead time parameter tuning.

More at link.