Individualized fMRI-Neuromodulation to Strengthen Swallowing Networks Targeted Towards Head and Neck Cancer Survivors, ALS, and Anterior Cervical Discectomy and Fusion Surgery Patients
Sandesh Reddy1, Rohan Vemu1, Cameron Noorbaksh1, Anthony Allam1, Peyton Presto7, Erich Sturgis2, Katherine Hutcheson8, James Orengo3, Sarah Heilbronner4, Kelly Bijanki4, Emmanouel Froudarakis9, Susan Hilsenbeck5, Dorina Papageorgiou6
1Baylor College of Medicine, 2Otolaryngology - Head & Neck Surgery, 3Neurology, 4Neurosurgery, 5Dan L Duncan Comprehensive Cancer Center, 6Neuroscience, Baylor College of Medicine, 7Texas Tech University Health Sciences Center, 8Head and Neck Surgery, M.D. Anderson Cancer Center, 9Institute of Molecular Biology and Biotechnology, FORTH
Objective:
By decoding motor and sensory control network selectivity involved in swallow function, we aim to deliver precise rehabilitation via individualized neuromodulation tailored for patients with head and neck cancers (HNC), amyotrophic lateral sclerosis (ALS), and anterior cervical discectomy and fusion surgery (ACDFS).
Background:
The Papageorgiou lab developed an individualized real-time fMRI closed-loop neuromodulation (iNM) to strengthen swallowing following surgery- and/or radiation-induced cranial neuropathy in HNC survivors, early phase ALS, or post-ACDFS by targeting swallow motor and sensory control (SwMSC) cortical networks. As a prelude to treating patients with iNM, we conducted a feasibility study in 30 healthy participants.
Design/Methods:
On study-day one, we decoded cortical magnitude and spatial patterns (n=30) during swallowing and tongue movement. On study-day two, participants underwent both iNM and control-NO-iNM scans. Linear support vector machines (SVMs) were trained to distinguish between swallowing and tongue movement under iNM and control conditions, by iteratively training on one participant and testing on the other 29.
Results:
iNM improved the spatial precision of SwMSC networks (80.6%) compared to the control condition (73%). SwMSC via iNM showed statistically significant increases in blood-oxygen-level-dependent signal magnitude: a 34% rise in the sensory network’s area under the curve and a 22% increase in the motor network. Spatiotemporal causal modeling identified two brain states during SwMSC iNM trials: (1) a dominant state present in 78% of iNM trials (45% of control trials); and (2) a non-dominant, noise state in 22% of iNM trials (55% of control trials). Enhanced activity was noted for motor and sensory cerebellum and basal ganglia under iNM.
Conclusions:
Our study demonstrates that iNM increases the signal-to-noise ratio in regions regulating SwMSC networks. These findings suggest that iNM could be a valuable intervention by identifying targets for optimal neurorehabilitation in patients with HNC, ALS or ACDFS treatment sequelae, as well as other neurological disorders.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.