A. Ahmadian

This paper develops a CNN-based framework to track invisible needles in ultrasound sequences using spatiotemporal features of speckle dynamics, achieving accurate localization in challenging conditions.

This paper introduces D2BGAN, a dual discriminator Bayesian GAN for MR-ultrasound registration, achieving improved accuracy in brain shift compensation with a mean TRE of 0.75 mm.

This study proposes a co-sparse analysis model for registering intra-operative photoacoustic images with pre-operative MR images to improve brain shift compensation in neurosurgery.

This paper introduces SegCaps, a capsule network for glioma segmentation in MRI images, achieving higher accuracy with fewer data and parameters than U-Net.

This paper proposes a deep learning framework for automated segmentation of PDAC and surrounding vessels in CT images, achieving improved Dice scores using 3D-CNN and texture descriptors.