This study leverages DGE MRI to investigate glucose metabolism connectivity as a potential biomarker for Huntington's disease (HD) using two mouse models: zQ175 KI and R6/2 KI. Using Pearson correlation analysis, we calculated glucose connectivity between brain regions for both HD models and WT controls. Results revealed significant connectivity alterations in HD models, particularly in regions such as the thalamus, caudate putamen, and dentate gyrus, which are associated with HD-related pathophysiology. To further examine metabolic patterns, we employed self-organizing maps (SOM) to cluster DGE MRI signal curves, identifying brain regions with similar glucose metabolism dynamics. The clustering analysis revealed discrete glucose metabolism zones, providing insights into spatial and temporal connectivity variations that might not be apparent in anatomical imaging alone. While clusters showed distinct temporal patterns—some with rapid initial signal increases, others with gradual changes—these metabolic shifts highlight SOM’s utility in assessing brain region-specific metabolic behaviors. The analysis indicates that glucose metabolism connectivity is notably disrupted in HD models, aligning with known HD pathology and reflecting both rapid and gradual disease progression observed in R6/2 and zQ175 KI mice, respectively. These findings underscore DGE MRI’s potential as a novel imaging biomarker, offering insights into the metabolic disruptions in HD that may inform early diagnosis and therapeutic interventions. However, limitations in signal strength and the complexity of SOM clustering warrant further methodological refinements to enhance biomarker reliability. In conclusion, our study supports the use of DGE MRI in identifying glucose metabolism connectivity disruptions as a viable HD biomarker, providing a robust framework for future studies targeting metabolic dysregulation in neurodegenerative diseases.