Physical embedded cascade synergic multi-frequency electrical impedance tomography method and related device
By employing a physically embedded cascaded collaborative multi-frequency electrical impedance tomography method, the challenge of lesion reconstruction and classification in multi-layered heterogeneous structures of the brain has been solved, achieving accurate lesion reconstruction and type differentiation, and improving the reliability of imaging and clinical classification capabilities.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- FOURTH MILITARY MEDICAL UNIVERSITY
- Filing Date
- 2026-04-08
- Publication Date
- 2026-07-17
AI Technical Summary
Existing multi-frequency electrical impedance tomography (EIT) technology struggles to accurately reconstruct and classify lesions in multi-layered heterogeneous structures of the brain. Traditional methods cannot overcome nonlinear signal aliasing, and the effectiveness of deep learning methods in complex brain scenarios remains to be verified.
The physical embedded cascaded collaborative multi-frequency electrical impedance tomography method is adopted. Through the physical information and data feature extraction fusion module, the physical information of the brain impedance is accurately extracted and shared features are constructed. Combined with the lesion prediction module and the type differentiation module, the lesion reconstruction and classification are realized.
It enables precise reconstruction and type differentiation of lesions in complex cranial scenarios, improves the reliability of imaging, makes up for the shortcomings of traditional methods, and can simultaneously realize lesion localization and clinical classification.
Smart Images
Figure CN122398259A_ABST