Shape-based disease grading via functional maps and graph convolutional networks with application to Alzheimer’s disease
2024

Shape-based disease grading for Alzheimer's disease

Sample size: 238 publication 10 minutes Evidence: high

Author Information

Author(s): Mayer Julius, Baum Daniel, Ambellan Felix, von Tycowicz Christoph

Primary Institution: Zuse Institute Berlin

Hypothesis

This work aims to alleviate limitations in shape analysis by adapting the concept of functional maps and presenting a graph-based learning approach for morphometric classification of disease states.

Conclusion

The proposed classifier for Alzheimer's disease from hippocampi shapes surpasses previous designs by a significant margin.

Supporting Evidence

  • The classifier achieved an average classification accuracy of 86.0% for the extended dataset.
  • The approach outperformed state-of-the-art methods for shape-based learning.
  • Functional correspondences allowed for the inclusion of topologically varying shapes in the analysis.

Takeaway

The study uses shapes of the brain to help tell if someone has Alzheimer's disease, and it found a better way to do this using new technology.

Methodology

The study employed a graph convolutional network for semi-supervised node classification based on shape data derived from MRI scans.

Potential Biases

Potential biases may arise from the reliance on specific imaging techniques and the characteristics of the dataset used.

Limitations

The study's approach may not generalize to other types of diseases or different anatomical structures.

Participant Demographics

The study included 238 subjects, with 101 diagnosed with Alzheimer's disease and 137 normal controls, considering sex and ApoE genotype.

Statistical Information

P-Value

0.008

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/s12880-024-01513-z

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication