Consistency in Neuroimaging With AI

Reading Time: 2 minutes

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.

“In the dynamic realm of medical exploration and diagnostics, the realm of neuroimaging assumes a pivotal role in unraveling the complexities of neurological disorders, such as multiple sclerosis (MS). Nevertheless, this process is not devoid of its challenges. Neuroimaging data derived from diverse clinic centers, hospitals, and even distinct manufacturers frequently exhibit disparities in image contrast, engendering complexities when endeavoring to compare and culminate the data for conclusive analysis.

In a significant development, Lianrui Zuo, MSE, a dedicated PhD scholar within the Department of Electrical and Computer Engineering at Johns Hopkins University, unveiled a comprehensive exploration into the repercussions and potential remedies for incongruities in imaging data acquisition. This enlightening discourse took center stage during a focused platform session centered around imaging themes at the distinguished 2023 Consortium of Multiple Sclerosis Centers (CMSC) Annual Meeting, held from May 31 to June 3, in Aurora, Colorado.1 Notable participants in this symposium delved into progressive strides aimed at establishing a standardized MRI protocol, precise measurement of the spinal cord in clinical MRI settings, and an insightful investigation into the correlation between initial cognitive performance and subsequent brain volume outcomes among MS patients.

In an engaging dialogue with NeurologyLive, Zuo elaborated on the crux of his research findings, as highlighted in his presentation. He expounded on how the integration of artificial intelligence (AI) can serve as an instrumental tool in mitigating the challenges posed by the multifaceted effects of multisite data acquisition in the domain of neuroimaging. Furthermore, he accentuated the paramount significance of prioritizing uniform data acquisition techniques in studies encompassing aspects such as gray matter volume and the intricate interplay between patient age and MS. Zuo also delved into the pivotal role that AI-driven tools play in standardizing the gamut of image data gleaned from disparate centers and manufacturers, thereby heralding a new era of consistency in the field of neuroimaging.”

Source NeurologyLive

Neurologica
Author: Neurologica