Contextual associations represented both in neural networks and human behavior

Contextual associations facilitate object recognition in human vision. However, the role of context in artificial vision remains elusive as does the characteristics that humans use to define context. We investigated whether contextually related objects (bicycle-helmet) are represented more similarly in convolutional neural networks (CNNs) used for image understanding than unrelated objects (bicycle-fork). Stimuli were of objects against a white background and consisted of a diverse set of contexts (N=73). CNN representations of contextually related objects were more similar to one another than to unrelated objects across all CNN layers. Critically, the similarity found in CNNs correlated with human behavior across three experiments assessing contextual relatedness, emerging significant only in the later layers. The results demonstrate that context is inherently represented in CNNs as a result of object recognition training, and that the representation in the later layers of the network tap into the contextual regularities that predict human behavior.

A Biomechanical Evaluation of a Novel Airbag Bicycle Helmet Concept for Traumatic Brain Injury Mitigation

In this study, a novel expandable bicycle helmet, which integrates an airbag system into the conventional helmet design, was proposed to explore the potential synergetic effect of an expandable airbag and a standard commuter-type EPS helmet. The traumatic brain injury mitigation performance of the proposed expandable helmet was evaluated against that of a typical traditional bicycle helmet. A series of dynamic impact simulations on both a helmeted headform and a representative human head with different configurations were carried out in accordance with the widely recognised international bicycle helmet test standards. The impact simulations were initially performed on a ballast headform for validation and benchmarking purposes, while the subsequent ones on a biofidelic human head model were used for assessing any potential intracranial injury. It was found that the proposed expandable helmet performed admirably better when compared to a conventional helmet design—showing improvements in impact energy attenuation, as well as kinematic and biometric injury risk reduction. More importantly, this expandable helmet concept, integrating the airbag system in the conventional design, offers adequate protection to the cyclist in the unlikely case of airbag deployment failure.

Exploring the Relationship Between Mandatory Helmet Use Regulations and Adult Cyclists’ Behavior in California Using Hybrid Machine Learning Models

In California, bike fatalities increased by 8.1% from 2015 to 2016. Even though the benefits of wearing helmets in protecting cyclists against trauma in cycling crash has been determined, the use of helmets is still limited, and there is opposition against mandatory helmet use, particularly for adults. Therefore, exploring perceptions of adult cyclists regarding mandatory helmet use is a key element in understanding cyclists’ behavior, and determining the impact of mandatory helmet use on their cycling rate. The goal of this research is to identify sociodemographic characteristics and cycling behaviors that are associated with the use and non-use of bicycle helmets among adults, and to assess if the enforcement of a bicycle helmet law will result in a change in cycling rates. This research develops hybrid machine learning models to pinpoint the driving factors that explain adult cyclists’ behavior regarding helmet use laws.

An Overview of the Effectiveness of Bicycle Helmet Designs in Impact Testing

Cycling accidents are the leading cause of sports-related head injuries in the US. Conventional bicycle helmets typically consist of polycarbonate shell over Expanded Polystyrene (EPS) foam and are tested with drop tests to evaluate a helmet’s ability to reduce head kinematics. Within the last decade, novel helmet technologies have been proposed to mitigate brain injuries during bicycle accidents, which necessitates the evaluation of their effectiveness in impact testing as compared to conventional helmets. In this paper, we reviewed the literature to collect and analyze the kinematic data of drop test experiments carried out on helmets with different technologies. In order to provide a fair comparison across different types of tests, we clustered the datasets with respect to their normal impact velocities, impact angular momentum, and the type of neck apparatus. When we analyzed the data based on impact velocity and angular momentum clusters, we found that the bicycle helmets that used rotation damping based technology, namely MIPS, had significantly lower peak rotational acceleration (PRA) and Generalized Acceleration Model for Brain Injury Threshold (GAMBIT) as compared to the conventional EPS liner helmets (p < 0.01). SPIN helmets had a superior performance in PRA compared to conventional helmets (p < 0.05) in the impact angular momentum clustered group, but not in the impact-velocity clustered comparisons. We also analyzed other recently developed helmets that primarily use collapsible structures in their liners, such as WaveCel and Koroyd. In both of the impact velocity and angular momentum groups, helmets based on the WaveCel technology had significantly lower peak linear acceleration (PLA), PRA, and GAMBIT at low impact velocities as compared to the conventional helmets, respectively (p < 0.05). The protective gear with the airbag technology, namely Hövding, also performed significantly better compared to the conventional helmets in the analyzed kinematic-based injury metrics (p < 0.001), possibly due to its advantage in helmet size and stiffness. We also observed that the differences in the kinematic datasets strongly depend on the type of neck apparatus. Our findings highlight the importance and benefits of developing new technologies and impact testing standards for bicycle helmet designs for better prevention of traumatic brain injury (TBI).