Research On Multi-modality Biomedical Imaging Algorithms

Medical imaging is an important diagnostic method in the field of modern medicine, and is of great significance. Many disease can not be easily detected at its early edge. However, with the assistance of medical imaging technique, the lesions can be found much more earlier, which earns more time for diagnose and improves the success rate of the diagnose effectively. Usually, single-modality imaging method is used, which means to judge the status of the biological tissue just according to one imaging method. Commonly known medical imaging methods include ultrasound imaging, X-ray imaging, sound-speed imaging and photoacoustic imaging. Among those methods, ultrasound imaging and X-ray imaging have been already applied to actual medical diagnose and have made great contributions to the prevention and treatment of some disease. Meanwhile, due to the advantage of photoacoustic imaging and sound-speed imaging that effecitive information can be provided from some different angles, these two methods have also been paid more and more attention.However, because of the complexity of the biological tissue, single-modality imaging method can hardly present all the information within the organization, which may lead to late discovery of the illness condition, and even wrong diagnosis results. Therefore, multi-modality imaging methods are studied in this thesis, mainly including three kinds of methods, named ultrasound and DBT(Digital Breast Tomosynthesis) imaging, photoacoustic and ultrasound imaging and BMode ultrasound and sound-speed imaging. The core concept of the multi-modality method is to combine some of the traditional single-modality imaging methods by optimizing the reconstruction algorithm or combining and displaying the imaging results from different single-modality imaging methods, which helps to make the result more accurate and informative.The theory of multi-modality imaging system is described in this thesis in detail, as well as the imaging setup and the imaging results. For the ultrasound and DBT imaging method, the imaging results show that the artifact along the z axis has been effectively reduced, which makes the quality in the imaging plan become much higher and the edges become much clearer. For the photoacoustic and ultrasound imaging method, since photocoustic imaging is much more sensitive of the functional information in the tissue, the combined imaging method offers more information that ultrasound imaging can not provide. For BMode ultrasound and sound-speed imaging method, the difference of sound speed inside the tissue can be effectively reflected by the sound-speed imaging method. As a result, this combined method can be used to make a great compensation for the BMode ultrasound imaging method, especially in the area where the acoustic impedence is nearly the same but sound speed differs. Through the comparison to single-modality method, multi-modality imaging methods introduced in this thesis exactly improves the imaging quality of the same tissue, which can improve the accuracy of diagnosis and provide protection for the subsequent treatment.