Robust Phase Unwrapping via Deep Image Prior for Quantitative Phase Imaging

TL;DR

This paper introduces a deep image prior-based method for phase unwrapping in quantitative phase imaging, enabling accurate reconstruction of complex biological samples without requiring training data.

Contribution

It proposes a novel deep learning approach that unrolls phase unwrapping without supervised training, suitable for complex biological samples.

Findings

Successfully unwrapped phases of complex samples

Works on both real and simulated data

No training dataset needed

Abstract

Quantitative phase imaging (QPI) is an emerging label-free technique that produces images containing morphological and dynamical information without contrast agents. Unfortunately, the phase is wrapped in most imaging system. Phase unwrapping is the computational process that recovers a more informative image. It is particularly challenging with thick and complex samples such as organoids. Recent works that rely on supervised training show that deep learning is a powerful method to unwrap the phase; however, supervised approaches require large and representative datasets which are difficult to obtain for complex biological samples. Inspired by the concept of deep image priors, we propose a deep-learning-based method that does not need any training set. Our framework relies on an untrained convolutional neural network to accurately unwrap the phase while ensuring the consistency of the measurements. We experimentally demonstrate that the proposed method faithfully recovers the phase of complex samples on both real and simulated data. Our work paves the way to reliable phase imaging of thick and complex samples with QPI.