Comparison of Two Methods of Deep Brain Stimulation Lead Reconstruction and Their Prediction Capacities

Renata Montes Garcia Barbosa; Joyce Yuri Silvestre Yamamoto; Miriam Carvalho Soares; Isabela Bruzzi Bezerra Paraguay; Sara Carvalho Barbosa Casagrande; Jorge Dornellys da Silva Lapa; Marcos Eugênio Ramalho Bezerra; Guilherme Sperling Torezani; Vanessa Milanese Holanda; Gabriel de Castro Micheli; Egberto Reis Barbosa; Rubens Gisbert Cury

doi:10.1016/j.neurom.2025.04.013

Comparison of Two Methods of Deep Brain Stimulation Lead Reconstruction and Their Prediction Capacities

Neuromodulation. 2025 Jun 20:S1094-7159(25)00191-6. doi: 10.1016/j.neurom.2025.04.013. Online ahead of print.

Authors

Affiliations

¹ Department of Neurology, School of Medicine, University of São Paulo, São Paulo, São Paulo, Brazil. Electronic address: renatamontesgarcia@hotmail.com.
² Department of Neurology, School of Medicine, University of São Paulo, São Paulo, São Paulo, Brazil.
³ Department of Neurology, School of Medicine, University of São Paulo, São Paulo, São Paulo, Brazil; Department of Neurology, Federal University of Pernambuco, Recife, Pernambuco, Brazil.
⁴ Department of Neurosurgery, School of Medicine, University of São Paulo, São Paulo, São Paulo, Brazil; Department of Medicine, Federal University of Sergipe, Aracaju, Sergipe, Brazil; Department Neurosurgery, Hospital de Cirurgia, Aracaju, Sergipe, Brazil.
⁵ Department of Neurology, Federal University of Pernambuco, Recife, Pernambuco, Brazil.
⁶ Department of Neurology, Federal Fluminense University, Niterói, Rio de Janeiro, Brazil.
⁷ Department of Neurosurgery, Mayo Clinic, Jacksonville, FL, USA; Department of Functional Neurosurgery, Beneficiência Portuguesa, São Paulo, São Paulo, Brazil.
⁸ Department of Neurology, Federal University of Rio de Janeiro, Rio de Janeiro, Rio de Janeiro, Brazil.
⁹ Department of Neurology, School of Medicine, University of São Paulo, São Paulo, São Paulo, Brazil; Hospital Israelita Albert Einstein, São Paulo, São Paulo, Brazil.

PMID: 40539948
DOI: 10.1016/j.neurom.2025.04.013

Abstract

Objectives: Deep brain stimulation (DBS) technology has become increasingly advanced, and tools capable of optimizing programming are necessary. Three-dimensional (3D) electrode imaging reconstruction has the potential to assist clinicians in selecting the most effective DBS contacts. In this study, we evaluated the predictive capacity of two different methods (Matlab software [Mathworks, Inc., Natick, MA] and Brainlab software [Brainlab, Munich, Germany]) in identifying effective DBS contacts, compared with clinical examination (the standard of care). The study also analyzes whether one method is superior.

Materials and methods: Overall, 29 patients were included: 27 with Parkinson's disease (PD) and two with dystonia (DT). Therefore, images of the 58 brain hemispheres were obtained. Of these, 56 had directional leads, and two had leads with eight-ring contacts. Electrode reconstruction was performed using Matlab software and Brainlab software. The anatomic relationship between the electrodes and the targets (subthalamic nucleus or globus pallidus internus) was analyzed. Contacts within the sweet spot of the target were chosen through 3D reconstruction visualization. In addition, if the best contact levels were directional, the best located directional contacts were chosen (this was possible in 40 leads). The best contact was defined as that with the highest motor symptoms improvement rate based on clinical examination (standard of care).

Results: The mean age of the patients was 56 years; 24 patients underwent surgery within ≤one year from the surgery, and the maximum time from surgery was three years. The difference in contact-level suggestion was only one hemisphere, with Matlab software correctly suggesting the best clinical contact in 83% (48/58) and Brainlab software in 84% (49/58). The suggestion capacity of the directional contacts was 80% for both methods. Furthermore, there was a 94.8% concordance rate between the two methods (55/58) on choosing the best contacts on the basis of image reconstruction.

Conclusion: Matlab software and Brainlab software are effective image reconstruction methods for accurately suggesting the best contact on the basis of clinical examination.

Keywords: Deep brain stimulation; Parkinson’s disease; dystonia; neuroimaging reconstruction.