Impact of Censoring on the Quality of Cortical Parcellations and Personalized TMS Targets

Head motion systematically biases functional connectivity (FC) estimates in resting-state functional MRI (rs-fMRI). A common mitigation strategy is to censor high-motion volumes and discard high-motion runs. However, overly stringent censoring risks discarding signal alongside noise, potentially degrading FC estimates. Here, we test the efficacy of various censoring strategies on individual-specific cortical parcellations and personalized transcranial magnetic stimulation (TMS) target selection. We leverage precision-fMRI datasets to define individualized "ground-truth" references based on [≥]1 hour of low-motion data per participant. We then simulate 10-min or 20-min rs-fMRI sessions with varying motion levels from the remaining data. Unsurprisingly, higher motion leads to parcellations and TMS targets that deviate further from the "ground truth" references. However, at any given motion level, lenient censoring consistently produces higher quality parcellations and personalized TMS targets than strict censoring. With the advent of personalized connectome-guided TMS in the clinic, a common dilemma is whether to re-scan a patient with high motion. We find that lenient censoring and retaining high-motion runs in high-motion sessions yields parcellations and TMS targets that are comparable to - or even better than - those derived from strict censoring and discarding of high-motion runs in mixed-motion sessions. These results suggest that within the upper bound of motion explored in this study, patients might not need to be re-scanned for personalized TMS. Overall, our study suggests that lenient censoring better preserves individual-specific information, with direct implications for personalized TMS and other precision neuroimaging applications.