Patient specific intracranial neural signatures of obsessions and compulsions in the ventral striatum

Abstract

Objective: Deep brain stimulation is a treatment option for patients with refractory obsessive-compulsive disorder. A new generation of stimulators hold promise for closed loop stimulation, with adaptive stimulation in response to biologic signals. Here we aimed to discover a suitable biomarker in the ventral striatum in patients with obsessive compulsive disorder using local field potentials. Approach: We induced obsessions and compulsions in 11 patients undergoing deep brain stimulation treatment using a symptom provocation task. Then we trained machine learning models to predict symptoms using the recorded intracranial signal from the deep brain stimulation electrodes. Main results: Average areas under the receiver operating characteristics curve were 62.1% for obsessions and 78.2% for compulsions for patient specific models. For obsessions it reached over 85% in one patient, whereas performance was near chance level when the model was trained across patients. Optimal performances for obsessions and compulsions was obtained at different recording sites. Significance: The results from this study suggest that closed loop stimulation may be a viable option for obsessive-compulsive disorder, but that intracranial biomarkers are patient and not disorder specific. Clinical Trial: Netherlands trial registry NL7486

Wearable and Mobile Technologies for the Evaluation and Treatment of Obsessive-Compulsive Disorder: Scoping Review

“…They found significant correlations of facial affect with subjective improvement during DBS programming, negative correlations of delta band power with OCD symptom severity, and correlations of delta band power with ERP [ 66 ]. Fridgeirsson et al [ 69 ] used machine learning models to analyze LFP data collected during rest and symptom provocation in an outpatient setting while DBS was turned off. Machine learning modeling of resting LFP identified individual patients significantly above the chance level.…”

“…Machine learning modeling of resting LFP identified individual patients significantly above the chance level. Modeling of LFP collected during symptom provocation and relief of symptoms predicted symptom state with an average accuracy of 32.5% and 38.8% for the boosted trees and deep learning model, respectively, with the latter reaching an average area under the curve of 78.2% for compulsions, 62.1% for obsessions, 58.7% for baseline, and 59.7% for relief [ 69 ].…”

Local field potentials in major depressive and obsessive-compulsive disorder: a frequency-based review

“…Finally, because of the small sample sizes of patients ( Tables 1 , 2 ), more studies are needed to confirm the validity and reliability of LFPs. In addition, LFPs data are mostly recorded in the resting state, with few reports on the task ( 46 ) or sleep state ( 32 ), making it difficult to fully reflect the mechanisms of MDD and OCD.…”