Cerebellar lobules optimal stimulation (CLOS): A computational pipeline to optimize cerebellar lobule-specific electric field distribution

Abstract

Objective: Cerebellar transcranial direct current stimulation (ctDCS) is challenging due to the complexity of the cerebellar structure which is reflected by the well-known variability in ctDCS effects. Therefore, our objective is to present a freely available computational modeling pipeline for cerebellar lobules' optimal stimulation (CLOS). Methods: CLOS can optimize lobule-specific electric field distribution following finite element analysis (FEA) using freely available computational modeling pipelines. We modeled published ctDCS montages with 5 cm × 5 cm anode placed 3 cm lateral to inion, and the same sized cathode was placed on the: (1) contralateral supra-orbital area (called Manto montage), and (2) buccinators muscle (called Celnik montage). Also, a published (3) 4×1 HD-ctDCS electrode montage was modeled. We also investigated the effects of the subject-specific head model versus Colin 27 average head model on lobule-specific electric field distribution. Three-way analysis of variance (ANOVA) was used to determine the effects of lobules, montage, and head model on the electric field distribution. The differences in lobule-specific electric field distribution across different freely available computational pipelines were also evaluated using subject-specific head model. We also presented an application of our computational pipeline to optimize a ctDCS electrode montage to deliver peak electric field at the cerebellar lobules VII-IX related to ankle function. Results: Eta-squared effect size after three-way ANOVA for electric field strength was 0.05 for lobule, 0.00 for montage, 0.04 for the head model, 0.01 for lobule∗montage interaction, 0.01 for lobule∗ head model interaction, and 0.00 for montage∗head model interaction. The electric field strength of both the Celnik and the Manto montages affected the lobules Crus I/II, VIIb, VIII, and IX of the targeted cerebellar hemisphere where Manto montage had a spillover to the contralateral cerebellar hemisphere. The 4×1 HD-ctDCS montage primarily affected the lobules Crus I/II of the targeted cerebellar hemisphere. All three published ctDCS montages were found to be not optimal for ankle function (lobules VII-IX), so we presented a novel HD-ctDCS electrode montage. Discussion: Our freely available CLOS pipeline can be leveraged to optimize electromagnetic stimulation to target cerebellar lobules related to different cognitive and motor functions.