RESUMEN
The use of standardized anthropomorphic test devices and test conditions prevent current vehicle development and safety assessments from capturing the breadth of variability inherent in real-world occupant responses. This study introduces a methodology that overcomes these limitations by enabling the assessment of occupant response while accounting for sources of human- and non-human-related variability. Although the methodology is generic in nature, this study explores the methodology in its application to human response in far-side motor vehicle crashes as an example. A total of 405 human body model simulations were conducted in a mid-sized sedan vehicle environment to iteratively train two neural networks to predict occupant head excursion and thoracic injury as a function of occupant anthropometry, impact direction and restraint configuration. The neural networks were utilized in Monte Carlo simulations to calculate the probability of head-to-intruding-door impacts and thoracic AIS 3+ as a function of the restraint configuration. This analysis indicated that the vehicle used in this study would lead to a range of 667 to 2,448 head-to-intruding-door impacts and a range of 3,041 to 3,857 cases of thoracic AIS 3+ in the real world, depending on the seatbelt load limiter. These real-world results were later successfully validated using United States field data. This far-side assessment illustrates how the methodology incorporates the human and non-human variability, generates response surfaces that characterize the effects of the variability, and ultimately permits vehicle design considerations and injury predictions appropriate for real-world field conditions.
Asunto(s)
Vehículos a Motor , Algoritmos , Antropometría , Fenómenos Biomecánicos , Simulación por Computador , Traumatismos Craneocerebrales/patología , Cabeza , Humanos , Persona de Mediana Edad , Redes Neurales de la Computación , ProbabilidadRESUMEN
Wearable sensors that accurately record head impacts experienced by athletes during play can enable a wide range of potential applications including equipment improvements, player education, and rule changes. One challenge for wearable systems is their ability to discriminate head impacts from recorded spurious signals. This study describes the development and evaluation of a head impact detection system consisting of a mouthguard sensor and machine learning model for distinguishing head impacts from spurious events in football games. Twenty-one collegiate football athletes participating in 11 games during the 2018 and 2019 seasons wore a custom-fit mouthguard instrumented with linear and angular accelerometers to collect kinematic data. Video was reviewed to classify sensor events, collected from instrumented players that sustained head impacts, as head impacts or spurious events. Data from 2018 games were used to train the ML model to classify head impacts using kinematic data features (127 head impacts; 305 non-head impacts). Performance of the mouthguard sensor and ML model were evaluated using an independent test dataset of 3 games from 2019 (58 head impacts; 74 non-head impacts). Based on the test dataset results, the mouthguard sensor alone detected 81.6% of video-confirmed head impacts while the ML classifier provided 98.3% precision and 100% recall, resulting in an overall head impact detection system that achieved 98.3% precision and 81.6% recall.