Scientists interested in MR guided focused ultrasound treatment of the thalamus.Focused ultrasound (FUS) enables surgical ablation of deep brain structures under MR guidance. Essential tremor (ET) is a disorder that can be treated with selective FUS ablation of the ventral intermediate nucleus (Vim).¹ The white-matter-nulled MP-RAGE (WMnMPRAGE) contrast at 7T permits detailed visualization of the thalamus.² In the present work, we used pre- and post-FUS WMnMPRAGE images in ET patients to segment ablation zones and correlate the type and extent of ablation with treatment outcome. We further defined a novel target, derived from the ablation zone of our best-responding patient, and tested the hypothesis that targeting this sub-region of Vim reduces tremor in the dominant hand by evaluating its coverage across subjects.Following informed consent, 10 ET patients were first scanned at 7T (Discovery MR950, GE Healthcare) using a 32 channel head coil (Nova Medical) with WMnMPRAGE scan parameters: TS 6s, TI 680ms, TR 10ms, BW 9kHz, flip angle 4°, FOV 18cm, 180x180x200 matrix, slice thickness 1mm, ARC parallel imaging 1.1x1.1 (2D radial fanbeam), scan time 10 min.³ WMnMPRAGE images were also collected at 3T (Discovery MR750, GE Healthcare) using an 8-channel head coil: TS 4.5s, TI 500ms, TR 10ms, BW 9kHz, flip angle 7°, FOV 18cm, 180x180x200 matrix, slice thickness 1mm, no parallel imaging, scan time ~10 min with prospective motion correction. Patients received MRgFUS treatment at 3T (ExAblate 4000, InSightec, Israel) within 2-4 weeks of these screening scans. Images were acquired immediately after the treatment. Manual delineation of the thalamic nuclei based on the Morel atlas⁴ and ablation lesions was performed for each patient, dividing each lesion into 3 regions based on their appearance on WMnMPRAGE: focal hypointensity, focal hyperintensity, and diffuse hyperintensity. Patients were registered to a study-specific average brain built from 9 3T pre-treatment images using ANTS⁵; one was rejected due to motion. This permitted direct comparison between subjects at every location in the thalamus in a normalized coordinate space. We then gathered simple metrics about the ablation zones such as their volumes, as well as more sophisticated ones like the coverage of a designated target region. These were correlated with clinical outcome as measured by Clinical Rating Scale for Tremor (CRST) A+B.Fig. 1 shows the manually segmented outlines of ablation lesions overlaid on post-treatment 3T WMnMPRAGE images. Fig. 2 displays the volume of the ablation regions in each patient on a log scale. We correlated the volumes of the baseline ablation zones against the clinical outcome at 1-month in Fig. 3. There appears to be some correlation with the volume of the focal hyperintensity but after Bonferroni correction for multiple comparisons, it cannot be considered significant (p = 0.042*5 = 0.21). However, qualitatively this suggests that the focal hyperintensity may be the important zone of the ablation to study. More importantly, we noted that patient E was the strongest responder to the FUS treatment (best clinical outcome with CRST A+B = 2 at 1-month) and additionally had a reduced volume of tissue affected by the FUS compared to a majority of the group of 10. Therefore we assessed his/her focal ablation region (combined focal hypointensity and hyperintensity), as seen in Fig. 4, in greater detail to evaluate whether coverage of this particular region was critical to the outcome of other subjects. We measured the fraction of this region that was covered by the focal ablation regions in other subjects in the average brain space. Fig. 5 shows this metric correlated against clinical outcome at 1-month and 6-months. Note that 2 subjects are omitted due to their ablations being on the right side of the brain, making it difficult to evaluate coverage against patient E’s left side ablation. This result shows a strong correlation at 1-month (r²=0.8) with a p-value that is significant after correction for multiple comparisons (p=0.003*5 < 0.05).We demonstrate that WMnMPRAGE reveals both thalamic anatomy and the internal structure of MRgFUS ablation, allowing detailed tracking of this procedure. We were able to characterize ablation sub-regions at millimeter-scale resolution and to compare these to an exemplary case. The resulting metric, fraction of volumetric coverage of each individual’s focal ablation over the strongest responder’s focal ablation, is highly correlated with the outcome of the patients, better than any other image-derived metric discovered to date. Ongoing work will continue to test the hypothesis that this region, located within Vim, represents a more effective treatment target. Longer term, we hope to deliver on the promise of patient-specific MRgFUS with WMnMPRAGE.