Precision of handheld multispectral optoacoustic tomography for muscle imaging

Abstract

Photo-or optoacoustic imaging (OAI) allows quantitative imaging of target tissues. Using multi-wavelength illumination with subsequent ultrasound detection, it may visualize a variety of different chromophores at centimeter depth. Despite its non-invasive, label-free advantages, the precision of repeated measurements for clinical applications is still elusive. We present a multilayer analysis of n = 1920 imaging datasets obtained from a prospective clinical trial (NCT03979157) in n = 10 healthy adult volunteers. All datasets were analyzed for 13 single wavelengths (SWL) between 660 nm-1210 nm and five MSOT-parameters (deoxygenated/oxygenated/total hemoglobin, collagen and lipid) by a semi-automated batch mode software. Intraclass correlation coefficients (ICC) were good to excellent for intrarater (SWL: 0.82-0.92; MSOT-parameter: 0.72-0.92) and interrater reproducibility (SWL: 0.79-0.87; MSOT-parameter: 0.78-0.86), with the exception for MSOT-parameter lipid (interrater ICC: 0.56). Results were stable over time, but exercise-related effects as well as inter-and intramuscular variability were observed. The findings of this study provide a framework for further clinical OAI implementation.