Enhancing diagnosis of mild cognitive impairment through brain-heart-gut metabolic networks in whole-body PET imaging.

The brain-only model achieved diagnostic performance comparable to whole-body PET for MCI diagnosis.

• The framework was validated across four datasets comprising 1,543 whole-body PET and 1,721 brain PET images.
• The brain-only model showed 'promising generalizability' across all four datasets.
• The model demonstrated potential for clinical translation in MCI diagnosis.

Key brain regions identified for MCI diagnosis involved the limbic, parietal, frontal, and temporal cortices.

• These regions engage the default mode network, central autonomic network, and sensorimotor networks.
• These brain regions were identified as part of an integrated brain-heart-gut metabolic network.
• The identified regions were used in building the brain-only diagnostic model.

Specific myocardium and distal colon regions, together with identified brain regions, constitute an integrated brain-heart-gut metabolic network in MCI.

• The multi-organ crosstalk is mediated by neural, biochemical, and mechanical pathways.
• Whole-body PET imaging was used to identify peripheral organ involvement in MCI.
• Heart dysfunction and gut microbiota dysbiosis were incorporated as contributing factors to MCI pathogenesis.
• The distal colon was specifically identified as the gut component of the network, rather than the entire gut.

Whole-body PET imaging provided a framework for detecting brain-heart-gut interactions in MCI as a prodromal stage of dementia.

• MCI is described as 'the prodromal stage of dementia involving complex interactions between the brain and peripheral organs.'
• Emerging evidence indicates heart dysfunction and gut microbiota dysbiosis contribute to MCI pathogenesis.
• The framework integrates brain-heart-gut interactions to enhance brain-only diagnostic performance.
• The approach offers 'broad applicability to other systemic diseases beyond MCI.'