Photoacoutic imaging of biological tissues is characterized by depth dependent optical fluence loss and acoustic variations. Here, we aim to correct for these inaccuracies aided by extrinsic imaging priors obtained through concurrent high-frequency ultrasound (US) imaging of tissue samples. We segmented the skin line and characterized tissue components using deformable model-based segmentation from the ultrasound images. The prior information from co-registered US images and tissue temperature was used to accurately model light fluence and speed of sound respectively. Methods applied here show significant improvement in beamforming performance, enhanced visual image quality and a higher PSNR.