Invertible Neural Networks for Uncertainty Quantification in Photoacoustic Imaging

TL;DR

This paper introduces a novel approach using conditional invertible neural networks to quantify uncertainty in multispectral photoacoustic imaging, enabling more reliable tissue parameter estimation and improved imaging strategies.

Contribution

The work presents a new cINN-based method for full posterior estimation in PAI, capturing ambiguity and multiple solutions, advancing uncertainty quantification in this imaging modality.

Findings

Demonstrated the ability to encode multiple tissue parameter modes

Enabled uncertainty-aware device design and image acquisition optimization

Showed potential for improved physiological parameter reconstruction

Abstract

Multispectral photoacoustic imaging (PAI) is an emerging imaging modality which enables the recovery of functional tissue parameters such as blood oxygenation. However, the underlying inverse problems are potentially ill-posed, meaning that radically different tissue properties may - in theory - yield comparable measurements. In this work, we present a new approach for handling this specific type of uncertainty by leveraging the concept of conditional invertible neural networks (cINNs). Specifically, we propose going beyond commonly used point estimates for tissue oxygenation and converting single-pixel initial pressure spectra to the full posterior probability density. This way, the inherent ambiguity of a problem can be encoded with multiple modes in the output. Based on the presented architecture, we demonstrate two use cases which leverage this information to not only detect and quantify but also to compensate for uncertainties: (1) photoacoustic device design and (2) optimization of photoacoustic image acquisition. Our in silico studies demonstrate the potential of the proposed methodology to become an important building block for uncertainty-aware reconstruction of physiological parameters with PAI.