Blows or Falls? Distinction by Random Forest Classification

SIMPLE SUMMARY: In forensic anthropology, skeletal trauma analysis can assist pathologists in determining the circumstance, cause, and manner of death. Determining whether the trauma is related to falls or induced by homicidal blows is often asked in relevance to legal issues. The hat brim line rule (HBL) is one of the most commonly used methods. The rule says that fractures resulting from blows may be found above and within the HBL, not on the skull’s base. Recent studies have found that the HBL rule must be used carefully, and postcranial skeletal trauma could be useful in this distinction. Evidence presented in court must follow Daubert’s guidelines for validity and reliability (evidence validated; error rates known; standards available; findings should be peer-reviewed and accepted by the scientific community). In this study, we assessed skeletal fracture patterns resulting from both etiologies. We tested various models for the method; the best one was based on the binary coding of 12 anatomical regions or 28 bones with or without baseline (age and sex). The results show the possible identification of the etiology in 83% of the cases. This method could be helpful for forensic experts in the interpretation of bone fractures. ABSTRACT: In this study, we propose a classification method between falls and blows using random forests. In total, 400 anonymized patients presenting with fractures from falls or blows aged between 20 and 49 years old were used. There were 549 types of fractures for 57 bones and 12 anatomical regions observed. We first tested various models according to the sensibility of random forest parameters and their effects on model accuracies. The best model was based on the binary coding of 12 anatomical regions or 28 bones with or without baseline (age and sex). Our method achieved the highest accuracy rate of 83% in the distinction between falls and blows. Our findings pave the way for applications to help forensic experts and archaeologists.