We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Quantitative Biology > Quantitative Methods

Title: Classification of head impacts based on the spectral density of measurable kinematics

Abstract: Traumatic brain injury can be caused by head impacts, but many brain injury risk estimation models are less accurate across the variety of impacts that patients may undergo. In this study, we investigated the spectral characteristics of different head impact types with kinematics classification. Data was analyzed from 3262 head impacts from head model simulations, on-field data from American football and mixed martial arts (MMA) using our instrumented mouthguard, and publicly available car crash data. A random forest classifier with spectral densities of linear acceleration and angular velocity was built to classify different types of head impacts (e.g., football, MMA), reaching a median accuracy of 96% over 1000 random partitions of training and test sets. Furthermore, to test the classifier on data from different measurement devices, another 271 lab-reconstructed impacts were obtained from 5 other instrumented mouthguards with the classifier reaching over 96% accuracy from these devices. The most important features in classification included both low-frequency and high-frequency features, both linear acceleration features and angular velocity features. It was found that different head impact types had different distributions of spectral densities in low-frequency and high-frequency ranges (e.g., the spectral densities of MMA impacts were higher in high-frequency range than in the low-frequency range). Finally, with head impact classification, type-specific, nearest-neighbor regression models were built for 95th percentile maximum principal strain, 95th percentile maximum principal strain in corpus callosum, and cumulative strain damage (15th percentile). This showed a generally higher R^2-value than baseline models without classification.
Comments: 16 pages, 5 figures
Subjects: Quantitative Methods (q-bio.QM); Machine Learning (cs.LG); Biological Physics (physics.bio-ph)
Cite as: arXiv:2104.09082 [q-bio.QM]
  (or arXiv:2104.09082v1 [q-bio.QM] for this version)

Submission history

From: Xianghao Zhan [view email]
[v1] Mon, 19 Apr 2021 06:58:43 GMT (1245kb)

Link back to: arXiv, form interface, contact.