Extend Energy-Mamba to three-dimensional volumetric medical data

Extend Energy-Mamba from two-dimensional medical image classification to three-dimensional volumetric medical data while preserving its physics-informed state-space formulation.

Background

Energy-Mamba is evaluated only on two-dimensional MedMNIST datasets. The conclusion identifies extension to three-dimensional volumetric medical data as an unresolved direction, which would test whether its learnable potential-energy constraint and gradient-based forcing mechanism remain effective for volumetric representations and substantially longer spatial sequences.

References

Several directions remain open for future investigation, including the extension of Energy-Mamba to 3D volumetric medical data, the extension to other tasks such as segmentation, the exploration of alternative energy function architectures, and a systematic ablation of the interplay between the kinetic and potential branches across different imaging modalities and dataset scales.

— Energy-Mamba: A Physics-Constrained State-Space Model for Medical Image Classification  (2608.19813 - Mabrok et al., 20 Aug 2026) in Conclusion