Applying Artificial Intelligence and Neuroimaging Into Clinical Care: Michael Dwyer, PhD

SAP Partner | <b>Buffalo Neuroimaging Analysis Center</b>

The director of IT and Neuroinformatics Development at the Buffalo Neuroimaging Analysis Center provided a realistic perspective of when and how AI will be incorporated into the wider medical system.

"The floodgates are not fully open yet, but we’ve already seen the first kind of vanguard of AI tools that have been FDA-approved. I think we’ll see it in a lot of other fields, too. The question is just making sure that we have good validation so that we know why it’s making those decisions."

Artificial intelligence (AI) and its capability to improve the healthcare system has always been a phenomenon. Through deep learning algorithms, computers can be taught to answer questions that augment human capabilities in imaging. Deep learning is a form of machine learning that employing convolutional neural networks, which has shown potential in medical imaging applications. This niche approach can help elevate and increase human capacity as well as improve efficiency by assisting in reading and computing large sets of data.

For some, AI is not just the future of just medicine, but of all occupational sectors because of its power to enhance efficiency, accuracy, value, and quality. Since its founding in 2000, the Buffalo Neuroimaging Analysis Center (BNAC) has been developing and using state-of-the-art neuroimaging techniques to help better understand and combat neurological diseases and disorders. Michael Dwyer, PhD, director, IT and Neuroinformatics Development, BNAC, has also led research efforts that utilize AI; however, he considers it a "double-edged sword."

Dwyer, who also serves as an assistant professor of neurology and biomedical engineering at the University at Buffalo, claimed AI tends to have a black-box nature, which has limited it from becoming more relevant in the clinical space. In an interview with NeurologyLive, he provided more context on the knowledge of AI, current limitations of this technique, and when clinicians can realistically expect it to incorporate it into neurology.