Precision functional network imaging-guided transcranial magnetic stimulation: A review of clinical applications...

Transcranial magnetic stimulation (TMS) is a non-invasive neuromodulation technique approved for treatment-resistant depression and increasingly applied to neuropsychiatric and neurologic conditions. However, conventional anatomical targeting overlooks individual variability in brain network architecture, limiting treatment efficacy. We aim to evaluate the potential of network-level imaging-guided TMS to enhance precision neuromodulation by leveraging multimodal structural and functional connectivity, including diffusion MRI with constrained spherical deconvolution (CSD) and resting-state BOLD fMRI. We additionally highlight emerging artificial intelligence (AI)-based analytic platforms capable of integrating these modalities to derive individualized connectivity patterns and optimized stimulation targets. We conducted a comprehensive literature review of TMS applications across neuropsychiatric and neurologic conditions, focusing on studies integrating diffusion MRI, functional connectivity, and network-guided targeting strategies. Studies were included from peer-reviewed clinical trials, meta-analyses, and neuroimaging investigations published between 1990 and 2024. Evidence for the clinical superiority of connectivity-guided TMS over standard anatomical targeting is mixed. Across randomized and large real-world cohorts, anatomically guided rTMS remains consistently effective, and connectivity-guided approaches are biologically grounded and feasible but have not demonstrated a clear overall advantage. These patterns suggest that the method of personalization, specifically circuit selection and dose/schedule, may be more influential than connectivity or anatomical guidance alone. Network-level imaging for connectivity-guided TMS remains a promising direction for precision neurostimulation. Future work should prioritize standardizing multimodal imaging pipelines, improving integration of CSD-based tractography and functional connectivity, aligning stimulation resolution with connectomic detail, and conducting pragmatic comparative trials that vary both target and protocol. Developing scalable targeting solutions, such as AI-assisted, atlas-based, or normative connectomic frameworks will be essential when individual imaging is not feasible.