Joseph V. Lee

Transcranial magnetic stimulation (TMS) is a non-invasive method for treating neurological and psychiatric disorders. It is being tested as an experimental treatment for patients with mild to moderate traumatic brain injuries (mTBI). Due to the complex, heterogeneous composition of the brain, it is difficult to determine if targeted brain regions receive the correct amount of electric field (E-field) induced by the TMS coil. E-field distributions can be calculated by running time-consuming finite element analysis (FEA) simulations of TMS on patient head models. Using machine learning, the E-field can be predicted in real-time. Our prior work used a Deep Convolutional Neural Network (DCNN) to predict the E-field in healthy patients. This study applies the same DCNN to mTBI patients and investigates how model depth and color space of E-field images affect model performance. Nine DCNNs were created using combinations of 3, 4, or 5 encoder and decoder blocks with the color spaces RGB, LAB, and YCbCr. As depth increased, training and testing peak signal-to-noise ratios (PSNR) increased and mean squared errors (MSE) decreased. The depth 5 YCbCr model had the highest training and testing PSNRs of 34.77 and 29.08 dB and lowest training and testing MSEs of 3.335∗10−4 and 1.237∗10−3 respectively. Compared to the model in our prior work, models of depth 5 have higher testing PSNRs and lower MSEs and, except for RGB. Thus, DCNNs with depth 5 and alternative color spaces, despite losing information through color space conversions, resulted in higher PSNRs and lower MSEs.

Transcranial magnetic stimulation (TMS) is a non-invasive treatment protocol for treating several psychiatric conditions, including depression, migraine, smoking cessation, and obsessive-compulsive disorder. Past research suggests that TMS treatment outcomes vary based on neuroanatomy, functional connectivity, and tractography-based structural connectivity. In a previous study, 26 mild to moderate traumatic brain injury (mTBI) patients underwent repetitive transcranial magnetic stimulation (rTMS) and showed improvements in depression, post-concussive symptoms, and sleep dysfunction. The present study was a secondary analysis of that data. Anatomically accurate head models were derived from magnetic resonance imaging (MRI), and finite element analysis simulations were performed to mimic empirical data collection. This allowed for examination of the roles that age, brain scalp distance (BSD), gray matter volume (GMV), site-specific electrical field strength (EFS), and depolarized gray matter volume (DGMV) had on resting motor threshold (RMT) at the precentral gyrus (PreCG). We also investigated how EFS simulated at the dorsolateral prefrontal cortex (DLPFC) and RMT influenced rTMS treatment outcomes. Linear regression showed BSD was associated with EFS, RMT, and DGMV supporting efforts to derive accurate parameters from MRI-based modeling. Furthermore, linear mixed effects modeling showed RMT was associated with EFS and DGMV at the PreCG when age and individual neuroanatomy was accounted for suggesting MRI based anatomy and simulated EFS potentially determine TMS dosage. We did not observe any significant relationship between any of the measures from this study on empirically collected rTMS outcomes in mTBI suggesting further investigations into the mechanisms behind these outcomes are needed.