17 Jun Weill Cornell Study Develops AI Tool To Understand How Human Behavior Arises from Brain Networks

Prof. Kuceyeski

MedicalResearch.com Interview with:

Prof.  Amy Kuceyeski Ph.D.
Professor of Mathematics in Radiology and Neuroscience
Weill Cornell Medicine

MedicalResearch.com: What is the purpose of the Krankencoder tool?

Response: The Krakencoder is a tool that allows us to compactly represent brain networks, or the connections between different parts of the brain. This compact representation helps us to take a step toward achieving the goal of understanding how complex human behavior, like thinking, social interactions, and emotion, arise from the complex network that is the human brain.

MedicalResearch.com: What are the main findings? Can these finding help improve the capabilities of artificial intelligence?

Response:  One question that we as network neuroscientists want to answer is how brain function arises from structure – that is, how does brain activity travel through the brain’s physical wiring and allow communication and coordination of the different parts of the brain? We used an AI model to map structure and function to one another in a way that was highly accurate and that, importantly, preserved differences in brain networks across individuals.

My brain network is different from yours, and we wanted our models to preserve these differences.  We also found that the compact representation of the brain’s networks resulted in a stronger association with cognitive performance than the non-compact representations. 

MedicalResearch.com: What should readers take away from your report?

Response: We have developed a new AI-based tool to analyze brain networks that could allow us to better understand how complex human behavior arises from the brain’s networks.

MedicalResearch.com: What recommendations do you have for future research as a results of this study?

Response: We are working now on using the tool in populations of folks that have had a stroke or have a diagnosis of multiple sclerosis – the idea would be to create biomarkers of disease and recovery patterns that could be used to tailor treatments to a specific person, and ultimately to improve their outcomes.

Any disclosures?  Keith Jamison was the driver of this work. No financial disclosures, but this work was funded by the NIH and the Bowers Foundation.

Citation:

Jamison, K.W., Gu, Z., Wang, Q. et al. Krakencoder: a unified brain connectome translation and fusion tool. Nat Methods (2025). https://doi.org/10.1038/s41592-025-02706-2

—–

The information on MedicalResearch.com is provided for educational purposes only, and is in no way intended to diagnose, cure, or treat any medical or other condition.Some links may be sponsored. Products are not warranted or endorsed.
Always seek the advice of your physician or other qualified health and ask your doctor any questions you may have regarding a medical condition. In addition to all other limitations and disclaimers in this agreement, service provider and its third party providers disclaim any liability or loss in connection with the content provided on this website.

Last Updated on June 17, 2025 by Marie Benz MD FAAD