Sliding multi-pixel method to improve sO2 estimation accuracy in in-vivo ovarian cancer diagnosis using photoacoustic tomography

In photoacoustic tomography (PAT), measurement errors arise from optical fluence spatial and temporal variations caused by tissue optical absorption and scattering heterogeneities, system noise, and motion. These errors influence the estimation accuracy of blood oxygenation saturation (sO2). In this study, we introduce a sliding multi-pixel approach to mitigate the effect of measurement errors before computing sO2 maps. As a result, the sO2 estimation is both more accurate, as evaluated by residual fitting errors, as well as smoother. We conclude by presenting diagnostic results from PAT of 33 patients with ovarian masses imaged by our coregistered PAT and ultrasound system.