Learning High-Order Relationships of Brain Regions

Weikang Qiu, Huangrui Chu, Selena Wang, Haolan Zuo, Xiaoxiao Li, Yize Zhao, Rex Ying
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:41582-41603, 2024.

Abstract

Discovering reliable and informative relationships among brain regions from functional magnetic resonance imaging (fMRI) signals is essential in phenotypic predictions in neuroscience. Most of the current methods fail to accurately characterize those interactions because they only focus on pairwise connections and overlook the high-order relationships of brain regions. We propose that these high-order relationships should be maximally informative and minimally redundant (MIMR). However, identifying such high-order relationships is challenging and under-explored due to the exponential search space and the absence of a tractable objective. In response to this gap, we propose a novel method named HyBRiD, which aims to extract MIMR high-order relationships from fMRI data. HyBRiD employs a Constructor to identify hyperedge structures, and a Weighter to compute a weight for each hyperedge, which avoids searching in exponential space. HyBRiD achieves the MIMR objective through an innovative information bottleneck framework named multi-head drop-bottleneck with theoretical guarantees. Our comprehensive experiments demonstrate the effectiveness of our model. Our model outperforms the state-of-the-art predictive model by an average of 11.2%, regarding the quality of hyperedges measured by CPM, a standard protocol for studying brain connections.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-qiu24b, title = {Learning High-Order Relationships of Brain Regions}, author = {Qiu, Weikang and Chu, Huangrui and Wang, Selena and Zuo, Haolan and Li, Xiaoxiao and Zhao, Yize and Ying, Rex}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {41582--41603}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/qiu24b/qiu24b.pdf}, url = {https://proceedings.mlr.press/v235/qiu24b.html}, abstract = {Discovering reliable and informative relationships among brain regions from functional magnetic resonance imaging (fMRI) signals is essential in phenotypic predictions in neuroscience. Most of the current methods fail to accurately characterize those interactions because they only focus on pairwise connections and overlook the high-order relationships of brain regions. We propose that these high-order relationships should be maximally informative and minimally redundant (MIMR). However, identifying such high-order relationships is challenging and under-explored due to the exponential search space and the absence of a tractable objective. In response to this gap, we propose a novel method named HyBRiD, which aims to extract MIMR high-order relationships from fMRI data. HyBRiD employs a Constructor to identify hyperedge structures, and a Weighter to compute a weight for each hyperedge, which avoids searching in exponential space. HyBRiD achieves the MIMR objective through an innovative information bottleneck framework named multi-head drop-bottleneck with theoretical guarantees. Our comprehensive experiments demonstrate the effectiveness of our model. Our model outperforms the state-of-the-art predictive model by an average of 11.2%, regarding the quality of hyperedges measured by CPM, a standard protocol for studying brain connections.} }
Endnote
%0 Conference Paper %T Learning High-Order Relationships of Brain Regions %A Weikang Qiu %A Huangrui Chu %A Selena Wang %A Haolan Zuo %A Xiaoxiao Li %A Yize Zhao %A Rex Ying %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-qiu24b %I PMLR %P 41582--41603 %U https://proceedings.mlr.press/v235/qiu24b.html %V 235 %X Discovering reliable and informative relationships among brain regions from functional magnetic resonance imaging (fMRI) signals is essential in phenotypic predictions in neuroscience. Most of the current methods fail to accurately characterize those interactions because they only focus on pairwise connections and overlook the high-order relationships of brain regions. We propose that these high-order relationships should be maximally informative and minimally redundant (MIMR). However, identifying such high-order relationships is challenging and under-explored due to the exponential search space and the absence of a tractable objective. In response to this gap, we propose a novel method named HyBRiD, which aims to extract MIMR high-order relationships from fMRI data. HyBRiD employs a Constructor to identify hyperedge structures, and a Weighter to compute a weight for each hyperedge, which avoids searching in exponential space. HyBRiD achieves the MIMR objective through an innovative information bottleneck framework named multi-head drop-bottleneck with theoretical guarantees. Our comprehensive experiments demonstrate the effectiveness of our model. Our model outperforms the state-of-the-art predictive model by an average of 11.2%, regarding the quality of hyperedges measured by CPM, a standard protocol for studying brain connections.
APA
Qiu, W., Chu, H., Wang, S., Zuo, H., Li, X., Zhao, Y. & Ying, R.. (2024). Learning High-Order Relationships of Brain Regions. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:41582-41603 Available from https://proceedings.mlr.press/v235/qiu24b.html.

Related Material