Category: Tremor
Objective: To identify factors associated with outcomes after magnetic resonance-guided focused ultrasound (MRgFUS) ventral intermediate nucleus (Vim) thalamotomy using motion-capture analysis of tremor videos in essential tremor (ET).
Background: In MRgFUS Vim thalamotomy for ET, most studies have focused on surgical technique and targeting, whereas predictors of response from the perspective of patient selection remain unclear. Recent advances in artificial intelligence and machine learning have improved the accuracy of freely available motion-capture applications.
Method: One hundred and sixty patients with ET who underwent unilateral Vim thalamotomy between May 2016 and April 2024 were screened. Preoperative postural hand tremor was analyzed with both upper limbs extended forward using an AI-based markerless motion-capture application (MediaPipe). Patients without visible tremor in this posture or with inadequate tracking were excluded, leaving 47 patients for analysis. Spectrograms, amplitude spectral density (ASD), and power spectral density (PSD) were generated. A distinct single-peak band was defined as a frequency difference of at least 1.5 Hz between the first and second PSD peaks and a peak power difference of at least 10 dB. Sample entropy (SampEn) and tremor stability index (TSI; Biase et al. Brain 2017) were also calculated. Outcome was assessed by the improvement rate in the Clinical Rating Scale for Tremor (CRST) score at 6 or 12 months visit.
Results: The mean age was 66.4±14.3 years, and mean disease duration was 26.7±15.6 years. Of the 47 patients, 31 showed a distinct single-peak band and 16 did not. CRST total score improvement was significantly greater in patients with a single-peak band than in those without (54% vs 43%, p = 0.042; Mann–Whitney U test). Among patients with a single-peak band, higher tremor frequency was associated with greater improvement (r = 0.362, p = 0.045; Pearson’s). Neither SampEn nor TSI was significantly associated with outcome.
Conclusion: The presence of a distinct single-peak band on spectrogram analysis may be associated with better response after MRgFUS Vim thalamotomy in ET. Among patients with a single-peak band, higher tremor frequency was associated with greater improvement. Motion-capture-based tremor analysis may help predict responsiveness to MRgFUS treatment.
To cite this abstract in AMA style:
Y. Osakada, H. Hirabayashi, A. Kinoshita, H. Ohnishi, H. Ohnishi, R. Morihara, H. Ishiura. Motion-Capture features associated with Outcome after MRgFUS Vim thalamotomy for Essential Tremor [abstract]. Mov Disord. 2026; 41 (suppl 1). https://www.mdsabstracts.org/abstract/motion-capture-features-associated-with-outcome-after-mrgfus-vim-thalamotomy-for-essential-tremor/. Accessed October 1, 2026.« Back to 2026 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/motion-capture-features-associated-with-outcome-after-mrgfus-vim-thalamotomy-for-essential-tremor/
