Local context encoding enables machine learning-based quantitative photoacoustics.

TL;DR: This work introduces the first machine learning-based approach to quantitative PA imaging (qPAI), which relies on learning the fluence in a voxel to deduce the corresponding optical absorption from PA images.

Abstract: Real-time monitoring of functional tissue parameters, such as local blood oxygenation, based on optical imaging could provide groundbreaking advances in the diagnosis and interventional therapy of various diseases. While photoacoustic (PA) imaging is a novel imaging concept with great potential to measure optical absorption deep inside tissue, quantification of the measurements remains a major challenge to be addressed. In this paper, we introduce the first machine learning-based approach to quantitative PA tomography (qPAT), which relies on learning the fluence in a voxel to deduce the corresponding optical absorption. The bottleneck of training data generation is overcome by encoding relevant information of the measured signal and the characteristics of the imaging system in voxel-based context images, which allow the generation of thousands of training samples from a single simulated PAT image. Comprehensive in silico experiments demonstrate that the concept of local context encoding (LCE) enables highly accurate and robust quantification (1) of the local fluence and the optical absorption from single wavelength PAT images as well as (2) of local oxygenation from multi wavelength PAT images.