Graph Attention Network For Microwave Imaging of Brain Anomaly
This work addresses the challenge of making microwave imaging models more practical for brain anomaly detection by reducing data needs, though it appears incremental in applying graph methods to a specific domain.
The paper tackles the problem of data-intensive microwave imaging models by proposing a graph formulation that incorporates the physical setup and symmetries of the imaging array, resulting in reduced data requirements, as evaluated on experimental brain anomaly localization.
So far, numerous learned models have been pressed to use in microwave imaging problems. These models however, are oblivious to the imaging geometry. It has always been hard to bake the physical setup of the imaging array into the structure of the network, resulting in a data-intensive models that are not practical. This work put forward a graph formulation of the microwave imaging array. The architectures proposed is made cognizant of the physical setup, allowing it to incorporate the symmetries, resulting in a less data requirements. Graph convolution and attention mechanism is deployed to handle the cases of fully-connected graphs corresponding to multi-static arrays. The graph-treatment of the problem is evaluated on experimental setup in context of brain anomaly localization with microwave imaging.