Supervised learning technique for the automated identification of white matter hyperintensities in traumatic brain injury. (14th October 2016)

• Record Type: Journal Article
• Authors: Stone, James R.; Wilde, Elisabeth A.; Taylor, Brian A.; Tate, David F.; Levin, Harvey; Bigler, Erin D.; Scheibel, Randall S.; Newsome, Mary R.; Mayer, Andrew R.; Abildskov, Tracy; Black, Garrett M.; Lennon, Michael J.; York, Gerald E.; Agarwal, Rajan; DeVillasante, Jorge; Ritter, John L.; Walker, Peter B.; Ahlers, Stephen T.; Tustison, Nicholas J.
• Is Part Of: Brain injury. Volume 30:Number 12(2016)
• Page Start: 1458
• Page End: 1468
• Publication Date: 2016-10-14
• Subjects: Neuroimaging -- brain imaging -- magnetic resonance imaging -- machine learning -- random forest decision tree -- deep learning -- TBI; Brain damage -- Periodicals; Brain -- Wounds and injuries -- Periodicals; Brain Injuries -- Periodicals; 617.481
• DOI: 10.1080/02699052.2016.1222080
• Languages: English
• ISSNs: 0269-9052

Abstract

Abstract: Background : White matter hyperintensities (WMHs) are foci of abnormal signal intensity in white matter regions seen with magnetic resonance imaging (MRI). WMHs are associated with normal ageing and have shown prognostic value in neurological conditions such as traumatic brain injury (TBI). The impracticality of manually quantifying these lesions limits their clinical utility and motivates the utilization of machine learning techniques for automated segmentation workflows. Methods : This study develops a concatenated random forest framework with image features for segmenting WMHs in a TBI cohort. The framework is built upon the Advanced Normalization Tools (ANTs) and ANTsR toolkits. MR (3D FLAIR, T2- and T1-weighted) images from 24 service members and veterans scanned in the Chronic Effects of Neurotrauma Consortium's (CENC) observational study were acquired. Manual annotations were employed for both training and evaluation using a leave-one-out strategy. Performance measures include sensitivity, positive predictive value, score and relative volume difference. Results : Final average results were: sensitivity = 0.68 ± 0.38, positive predictive value = 0.51 ± 0.40, = 0.52 ± 0.36, relative volume difference = 43 ± 26%. In addition, three lesion size ranges are selected to illustrate the variation in performance with lesion size. Conclusion : Paired with correlative outcome data, supervised learning methods may allow for identification of imaging features predictive of diagnosis and prognosis in individual TBI patients.