Meeting Banner
Abstract #1373

HKMF: Hyperbolic Kernel-based Multimodal Fusion for HIV-Associated Neurocognitive Disorder Analysis

Meimei Yang1, Qianqian Wang1, Yongheng Sun1, Mengqi Wu1, Wei Wang2, Hong-Ju Li2, and Mingxia Liu1
1University of North Carolina at Chapel Hill, Chapel Hill, NC, United States, 2Department of Radiology, Beijing Youan Hospital, Capital Medical Unviersity, Beijing, China

Synopsis

Keywords: Diagnosis/Prediction, Diagnosis/Prediction, HIV-Associated Neurocognitive Disorder (HAND), Hyperbolic kernel, Multimodal fusion

Motivation: HIV-associated neurocognitive disorder (HAND) has generally been studied through a single imaging modality like functional Magnetic Resonance Imaging (fMRI), which restricts understanding of its complex neuropathological causes.

Goal(s): Our goal is to develop a multimodal approach that integrates fMRI and diffusion tensor imaging (DTI) for a more comprehensive HAND analysis.

Approach: We introduce hyperbolic kernel-based multimodal fusion (HKMF), a new method for HAND analysis, combining fMRI and DTI data using hyperbolic geometry for more comprehensive insights.

Results: Experimental results on 137 subjects show that our approach outperforms traditional methods in HAND diagnosis.

Impact: HKMF utilizes hyperbolic geometry, which is well-suited for capturing complex hierarchies in neuroimaging applications. This approach not only enhances HAND analysis but also can be extended to other medical imaging applications involving multimodal datasets.

How to access this content:

For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.

After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.

After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.

Click here for more information on becoming a member.

Keywords