Enhancing Consistency in Neuroimaging With AI and Image Harmonization: Lianrui Zuo, MSE


The PhD student in the department of electrical and computer engineering at Johns Hopkins University discussed the use of artificial intelligence and image harmonization techniques to address the challenges caused by multisite effects in neuroimaging. [WATCH TIME: 4 minutes]

WATCH TIME: 4 minutes

“After harmonization, we observed that the images are more consistent as if they were acquired using a standardized, consistent protocol. We know that it's easier for radiologists to compare data across different centers, different time points, and also for machine learning other computer algorithms to more accurately more consistently processed data."

In the advancing field of medical research and diagnostics, neuroimaging plays an important role in understanding complex neurological disorders such as multiple sclerosis (MS); however, it is not a process without challenges. Neuroimaging data collected from different clinic centers, hospitals, and even manufacturers, often exhibit variations in image contrast, which may lead to some difficulties comparing the data and final analysis.

Recently, Lianrui Zuo, MSE, a PhD student in the Department of Electrical and Computer Engineering at Johns Hopkins University, presented on impacts and solutions of inconsistent imaging data acquisition in a platform session focused on imaging topics at the 2023 Consortium of Multiple Sclerosis Centers (CMSC) Annual Meeting, May 31 to June 3, in Aurora, Colorado.1 The rest of the speakers in the session discussed topics such as progressive steps towards a standardized MRI protocol, spinal cord measurement in clinical MRI, and the relationship between baseline cognitive performance and brain volume outcomes in patients with MS.

Zuo sat down in an interview with NeurologyLive® at the meeting to provide an overview of his research based on the presentation. He discussed how image harmonization using artificial intelligence (AI) can help to alleviate the challenges caused by multisite effects in neuroimaging. He also talked about the importance of prioritizing consistent data acquisition in studies involving gray matter volume and patient age relationships among patients with MS. Furthermore, Zuo spoke about the role of AI tools in standardizing image data acquired from different centers and manufacturers in the field of neuroimaging.

Click here for more coverage of CMSC 2023.

1. Zuo L. Inconsistent MR Acquisition in Longitudinal Volumetric Analysis: Impacts and Solutions. Presented at: 2023 CMSC Annual Meeting; May 31-June 3; Aurora, CO. IMG05.
Related Videos
Shahid Nimjee, MD, PhD
Peter J. McAllister, MD, FAAN
Video 6 - "Utilization of Neuroimaging in Alzheimer’s Disease"
Video 5 - "Contribution of Multiple Pathways to the Development of Alzheimer’s Disease"
Michael Levy, MD, PhD
Michael Levy, MD, PhD
Michael Kaplitt, MD, PhD
© 2024 MJH Life Sciences

All rights reserved.