Detecting a History of Repetitive Head Impacts from a Short Voice Recording

Abstract

Repetitive head impacts (RHI) are associated with an increased risk of developing various neurodegenerative disorders, such as Alzheimer's disease (AD), Parkinson's disease (PD), and most notably, chronic traumatic encephalopathy (CTE). While the clinical presentation of AD and PD is well established, CTE can only be diagnosed post-mortem. Therefore, a distinction can be made between the pathologically defined CTE and RHI-related functional or structural brain changes (RHI-BC) which may result in CTE. Unfortunately, there are currently no accepted biomarkers of CTE nor RHI-BC, a major hurdle to achieving clinical diagnoses. Interestingly, speech has shown promise as a potential biomarker of both AD and PD, being used to accurately classify individuals with AD and PD from those without. Given the overlapping symptoms between CTE, RHI-BC, PD and AD, we aimed to determine if speech could be used to identify individuals with a history of RHI from those without. We therefore created the Verus dataset, consisting of 13 second voice recordings from 605 professional fighters (RHI group) and 605 professional athletes in non-contact sports (control group) for a total of 1210 recordings. Using a deep learning approach, we achieved 85% accuracy in detecting individuals with a history of RHI from those without. We then used our model trained on the Verus dataset to fine-tune on publicly available AD and PD speech datasets and achieved new state-of-the-art accuracies of 84.99% on the AD dataset and 89% on the PD dataset. Finding a biomarker of CTE and RHI-BC that presents early in disease progression is critical to improve risk management and patient outcome. Our study is the first we are aware of to investigate speech as such a candidate biomarker of RHI-BC.

Articulation Rate, Pauses, and Disfluencies in Professional Fighters: Potential Speech Biomarkers for Repetitive Head Injury

“…Acoustic speech measures show great promise in early identification of RHI-related behavioral changes. Tauro and colleagues 13 recently differentiated speakers with and with no history of RHI with 85% accuracy using automated analysis of 13-second voice recordings, including the correct classification of boxers who were just 21 years of age. Articulation rate, pausing, and disfluencies are relatively easy to measure, even in everyday speaking conditions and by automatic recognition systems, so they have the potential for early clinical identification of RHIassociated neurological decline and to track changes in behavior over time.…”

“…10 Two recent reviews discuss the utility of speech biomarkers such as speech rate, pause time, and pause duration for the detection of neurological disturbance including Parkinson disease, amyotrophic lateral sclerosis, and RHI. 11,12 For example, Tauro et al 13 reported that acoustic machine learning techniques differentiated speakers with and with no history of RHI with 85% accuracy. In this study, we compared fighters exposed to RHI with controls who had no history of RHI exposure on several characteristics of connected speech that have been associated with various types of dysarthria and that can be reliably measured in speech samples collected in clinical conditions.…”