A new paper published in the Annals of Neurology by trained pattern classifiers discriminated between patients with microbleeds and age-match controls with a high degree of accuracy, and outperformed other methods. “Individual prediction of white matter injury following traumatic brain injury,” Hellyer PJ, Leech R, Ham TE, Bonnelle V and Sharp DJ, Ann Neurol 2013.
In their article, the researchers note:
Traumatic brain injury often results in traumatic axonal injury (TAI). This can be difficult to identify using conventional imaging. Diffusion tensor imaging (DTI) offers a method of assessing axonal damage in vivo, but has previously mainly been used to investigate groups of patients. Machine learning techniques are increasingly used to improve diagnosis based on complex imaging measures. We investigated whether machine learning applied to DTI data can be used to diagnose white matter damage after TBI and to predict neuropsychological outcome in individual patients.
The researchers, “Trained pattern classifiers to predict the presence of white matter damage in twenty-five TBI patients with microbleed evidence of TAI compared to neurologically healthy age-match controls.” The researchers then applied these classifiers to, “Thirty-five additional patients with no conventional imaging evidence in TAI [mTBI patients]. Finally, using a regression analyses to predict indices of neuropsychological outcome for information processing speed, executive function and associative memory in a group of seventy heterogeneous patients.”
The study provides, “Proof of principal that multivariate techniques can be used with DTI to provide diagnostic information about clinically significant TAI.”
This is a very important study. While it was not exclusive to mTBI patients – about one half were, DTI clearly delineated a difference between controls and TBI subjects.
This study also debunks Larrabee (2013) in their response to Bigler (2013) in which Larrabee noted the inability of DTI to “diagnose” TBI.