Andres M Lozano, MD, PhD, FRCSC, BMedSci

Dr. Lozano's research focuses on Functional Neurosurgery and the development of novel therapies for movement disorders and psychiatric disease. His lab studies brain mapping, deep brain stimulation (DBS) and focused ultrasound (FUS) in patients and in animal models of disease. His lab has approximately 850 publications and ranks first globally in total citations in the field of deep brain stimulation. Dr. Lozano is the most cited neurosurgeon in the world (Clarivate). He has been involved in phase I “first in man” applications of DBS in dystonia, Huntington’s, depression, anorexia and Alzheimer disease, and has been involved in phase II and III trials of DBS and brain intraparenchymal drug delivery in patients with Parkinson disease and other neurological disorders. His recent work in experimental animals has shown that DBS can drive neurogenesis and enhance memory function. Dr. Lozano serves as Editor-in-Chief of Stereotactic and Functional Neurosurgery.

Predicting the best deep brain stimulation parameters for Parkinson's disease using functional MRI and machine learning

Study Status: Active
Study Purpose: Deep brain stimulation helps people with Parkinson's disease feel better when the settings are just right. Finding the best settings for deep brain stimulation is known as programming. It is often done by trial and error and based on what the doctor observes and their experience. This means that finding the best settings can take a lot of visits to the doctor. This study looks at whether a brain scan and machine learning can help predict the best deep brain stimulation settings for each person.
Background: Deep brain stimulation is a common treatment for movement disorders like Parkinson’s disease. It works by placing an electrode in the brain to deliver electrical stimulation. This helps correct abnormal brain activity. New brain imaging techniques, like functional magnetic resonance imaging (fMRI), have helped us learn more about how deep brain stimulation affects brain activity. Previously, we analyzed the brain scans of 67 Parkinson’s disease patients with both good and not-so-good deep brain stimulation settings. We found that the best settings made a unique pattern in the brain, especially in the part that controls movement. We used this pattern and data from 39 patients who had already had their deep brain stimulation settings optimized, to create a computer program that could tell if the settings were good or not. The computer program was 88% accurate. We tested this program on new groups of patients, including some who had never had deep brain stimulation before, and it worked well. This means fMRI scans could help find the best deep brain stimulation settings without as many doctor visits.
Study Methods: We plan to enroll 200 Parkinson’s disease patients who have had deep brain stimulation to participate in this study. Once participants have provided their consent to participate in the study, they complete two fMRI sessions. In each session, magnetic waves are used to take pictures of the brain and brain activity. The two sessions are scheduled at least one week apart and are two hours each. Each session involves being scanned in the fMRI while their deep brain stimulation settings are kept at good settings and then again when they were set to not-so-good deep brain stimulation settings. So far 67 patients have completed the study already and these early results were published in the research journal Nature Communications (See the External Links section for more information).

 

For a list of Dr. Lozano's publications, please visit PubMed or Scopus.


Alan & Susan Hudson Cornerstone Chair in Neurosurgery, UHN
University Professor, Department of Surgery, University of Toronto