Development and Validation of a New Pediatric Head Injury Assessment Tool for Possible Child Abuse Cases Considering Subject-Specific Child Head Anatomy

Establishing whether a head injury to a child was the result of a short fall or abuse is a fundamental problem in forensic investigation. Current injury assessments on possible child abuse cases are mainly based on clinical evidence and experts' professional experiences. However, expert opinions regarding how a head injury occurred to a child are often not based on scientific evidence and sometimes in direct conflict, which can result in wrongful legal action. The goal of this research is to establish a new paradigm for developing and using subject-specific pediatric head finite element (FE) models to provide objective and accurate assessments for judging the consistency of head injuries and the stated causes in infants or young children. The FE model-based injury assessment tool is designed to evaluate whether a non-abusive injury cause, such as a fall, described by the parents or caregivers can or cannot cause the medically observed head injury. However, it is not designed to directly predict shaken baby syndrome due to the controversial injury mechanism and difficulty in accurately quantifying the severity of a child abuse event. Our modeling paradigm will provide useful, accurate and objective information to both clinicians and forensic investigators which will help determine the true causes of a pediatric head injury. Five specific aims are proposed in this study: 1) to develop a statistical model that describes head geometry of 0-3 year-old (YO) children based on CT images, accounting for head size and shape, suture and fontanelles size, and skull thickness; 2) to conduct dynamic pediatric brain material-property tests using porcine brain tissues and quantify the age effects on brain material properties under loading conditions similar to real world injurious events; 3) to develop a fast and efficient method to automatically generate subject-specific pediatric head FE models capable of representing developmental changes during child growth; 4) to computationally reconstruct 50 infant cadaver drop tests reported by Weber (1984, 1985) for model validation/optimization and development of a skull-fracture injury criterion for 0-1 YO children; and 5) to prospectively collect clinical data on and computationally reconstruct 80 head-first shortfall accidents and child abuse cases for further model validation/optimization and development of a traumatic brain injury criterion for 0-3 YO children. The final product of this project will be a model-based head-injury assessment tool for 0-3 YO children. Users will only need to input patients basic information, such as age, weight, height, head circumference, and/or head CT/MRI images to rapidly generate the subject-specific head FE model. If incident conditions, such as fall height, fall angle, and impact surface material and thickness, are also defined, the model will be able to predict a range of skull/brain injury risks based on model-calculated results, and will provide a statistical assessment to whether the existing head injury is consistent with the stated injurious event. This pediatric head-injury assessment tool will provide a more objective, accurate, and cost effective way for forensic science researchers, practitioners, and policymakers in the criminal justice system to evaluate the consistency of a head injury with the stated cause in an infant or a young child. ca/ncf