Three-dimensional photoacoustic tomography and its application to detection of joint diseases in the hand

This thesis research presents the study of three-dimensional (3-D) photoacoustic tomography (PAT) and its application for the first time to detection of osteoarthritis (OA) in the hand. PAT is an emerging non-ionizing, non-invasive imaging modality that can visualize high optical contrast in biological tissues with high ultrasound resolution. Compared with other imaging modalities that have been conventionally used or recently investigated to visualize the structural abnormalities in the finger joints with OA, the 3-D PAT approach studied in this dissertation not only provides high-resolution anatomical structures, but also offers quantitative tissue optical property as well as physiological / functional information including concentrations of oxy-hemoglobin (HbO2), deoxy-hemoglobin (Hb) and water (H 2O) that could be used to detect OA in an early stage.;In this study, a 3-D high performance PAT reconstruction algorithm is developed based on parallel computing technique and finite element method. The optimal detector performance and scanning geometry for 3-D photoacoustic imaging of the finger joints are investigated with considerable phantom experiments. The results of phantom experiments show that a spherical scanning geometry appears to provide improved spatial solution over a cylindrical scanning geometry, and that 1mm thick "cartilage" can be accurately differentiated from the "bones" with a 1 MHz transducer in a spherical scanning geometry. In addition, the absorption coefficient of the "cartilage" can be effectively recovered when this optical property varied from 0.015mm -1 to 0.04mm-1. A 3-D PAT system in a spherical scanning geometry has been constructed and optimized for in-vivo examination of human finger joints. A distal interphalangeal (DIP) joint from a female healthy subject was photoacoustically examined by our 3-D PAT system, and major anatomical structures of the examined DIP joint along with the side arteries were clearly reconstructed in high quality, where joint space was well differentiated from surrounding finger phalanges. The performance of the 3-D PAT system was further improved with an ultrasound detection array composed of eight 1 MHz transducers and a 16-channel pulse/receive board. The 8-channel 3-D PAT system was carefully calibrated with controlled tissue phantom experiments, and is capable of completing a finger joint examination within 5 minutes (at single optical wavelength).;The 3-D PAT reconstruction algorithm and scanning system developed has also been applied to a pilot clinical study aiming to test the possibility of detecting OA in the hand joints using 3-D PAT. In this pilot clinical study, seven subjects (two OA patients and five healthy controls) were enrolled and photoacoustically examined. The image quality of the reconstructed finger joints was greatly improved with the 8-channel 3-D PAT system, and apparent differences, in both the reconstructed size of the joint space and the absorption coefficient of the joint cavity, has been observed between the OA and normal joints. The successful results obtained suggest the possibility of 3-D PAT as a potential clinical tool for early detection of OA in the finger joints. Major chromophore concentrations (HbO2 and Hb) of in-vivo finger joints have also been quantitatively imaged using multispectral 3-D PAT approach with six optical wavelengths from 730nm to 880nm. The multispectral results obtained further confirmed that the 3D PAT approach implemented in this thesis research is able to differentiate OA from normal joints.;While we target the detection of OA as a testing base for validating the single- and multi-spectral 3-D PAT approaches developed in this thesis research, many aspects of our work are fundamental to imaging in general. For example, the 3-D PAT approaches implemented are applicable to other biomedical problems such as rheumatoid arthritis (RA) detection and functional brain imaging.