Research And Application Of Algorithms For CT Cerebral Perfusion

In recent years, with the development of biomedical technology, the research of CT cerebral perfusion algorithms base on the theory of nuclear contrast agent becoming popular, and studies for cerebral perfusion algorithms become one of the focus topics. A large number of experiments have shown that the CT perfusion imaging is an effective method to detect the physiological changes in organ and tissue with highly sensitive and resolution. With the development of CT imaging technology, conventional CT imaging and the functional perfusion image are playing an increasingly important role in the clinical diagnosis. In this thesis, the main study is the cerebral perfusion algorithm. This research combines modeling, numerical calculation, optimization and medical image informatics, the main studies as follow:Establish physiological models and simplify the parameters of models; Evaluate the results of different methods; Numerical calculation and optimization for multi-parameter model; Analyse the parameters of the perfusion imaging results.The focus of the study is the development of dynamic CT cerebral perfusion imaging algorithms and software, base on the basic principles and related algorithms, finally realize the function modules.The major work done including:(1) Develope a system which supports the DICOM standard images in cerebral perfusion CT-assisted diagnostic, and the system supports for the DICOM medical imaging standard format files read, storage and data conversion.(2) Simulated the compartment models which is the basis of the perfusion algorithms, and evaluated these algorithms, do researches on numerical calculation and optimization of the multi-parameter model.(3) The software involved the perfusion function is in achieving the clinical value, and as the tool in an "Eleventh Five-Year" project being used in many hospitals in our country.