Two-step training deep learning framework for computational imaging without physics priors.

Deep learning (DL) is a powerful tool in computational imaging for many applications. A common strategy is to use a preprocessor to reconstruct a preliminary image as the input to a neural network to achieve an optimized image. Usually, the preprocessor incorporates knowledge of the physics priors in the imaging model.

Preclinical imaging of epidermal growth factor receptor with ABY-029 in soft-tissue sarcoma for fluorescence-guided surgery and tumor detection.

Background And Objectives: Fluorescence-guided surgery using epidermal growth factor receptor (EGFR) targeting has been performed successfully in clinical trials using a variety of fluorescent agents. We investigate ABY-029 (anti-EGFR Affibody molecule labeled with IRDye 800CW) compared with a small-molecule perfusion agent, IRDye 700DX carboxylate, in a panel of soft-tissue sarcomas with varying levels of EGFR expression and vascularization.

Methods: Five xenograft soft-tissue sarcoma cell lines were implanted into immunosuppressed mice.