ABSTRACT Transcranial magnetic stimulation (TMS) is an FDA-approved treatment for treatment-resistant major depression (TRD). However, at present the approach is only partially effective, due in part to relatively ineffective targeting approaches. Over recent years, there has been increasing realization that TMS for anhedonic-like symptoms of major depressive disorder (MDD) is most effective when targeted at the region of left dorsolateral prefrontal cortex (L-DLPFC) that is maximally anti-correlated with subgenual anterior cingulate cortex (sgACC) as determined using resting-state functional connectivity MRI (rsfcfMRI). Ideal approaches for identifying this location for individual subjects, however, still depend on processing pipelines that are not FDA cleared and thus and are not accessible to the TMS treatment community. In general, 3D brain reconstructions can be analyzed using either traditional, volumetric approaches, or more recently developed surface-based approaches developed as part of the human connectome project (HCP) and leading to publication of the HCP multimodal brain atlas (HCP MMP1.0). In volumetric approaches, structurally and functionally based approaches have both been implemented. In the surface-based HCP MMP1.0 approach, structural and functional data are merged to identify functionally discrete brain parcels across individuals. Here, during phase I we will first develop an investigational software package (“TMSMap”) that will permit TMS researchers to utilize either volumetric or surface-based analytic approaches for determining MR-based personalized L-DLPFC target locations on-line, and will create a parallel off-line virtual neuro-navigation tool to permit neuro-navigated TMS even for “community” practioners without access to on-line neuro-navigation capabilities. During phase II we will create a Picture Archiving and Communication System (PACS) compliant clinic-ready version, which we evaluate in a multicenter randomized controlled trial in TMS-resistant TRD individuals. We will ensure that all device (software developed to medical standards) and data (clinical validation) requirements are complete by Phase II end, to enable pursuing either a 510(k) or de-novo submission with positive clinical outcome. The project builds both from the long-standing interest of Soterix Medical Inc.(SMI) in 1) the use of combined TMS and neuro-navigation, and 2) the development of automated targeting software for non-invasive brain stimulation; and from recent research by the principal investigator at Columbia University Irving Medical Center (CUIMC) demonstrating 1) anti-correlation of HCP MMP1.0 parcel 46 with sgACC across individuals within the HCP sample and 2) 100% response rate among TMS-resistant TRD patients (n=10) treated with TMS targeted to HCP MMP1.0 parcel 46. Moreover, both effective electroconvulsive therapy and effective TMS were associated with rsfc changes involving DLPFC parcel 46 and additional brain regions. This will...