We develop an image analysis system to automatically detect pleural thickenings and assess their characteristic values from patients' thoracic spiral CT images. Algorithms are described to carry out the segmentati...
详细信息
ISBN:
(纸本)9780819466327
We develop an image analysis system to automatically detect pleural thickenings and assess their characteristic values from patients' thoracic spiral CT images. Algorithms are described to carry out the segmentation of pleural contours and to find the pleural thickenings. The method of thresholding was selected as the technique to separate lung's tissue from other. Instead thresholding based only on empirical considerations, the so-called "supervised range-constrained thresholding" is applied. The automatic detection of pleural thickenings is carried out based on the examination of its concavity and on the characteristic Hounsfield unit of tumorous tissue. After detection of pleural thickenings, in order to assess their growth rate, a spline-based interpolation technique is used to create a model of healthy pleura. Based on this healthy model, the size of the pleural thickenings is calculated. In conjunction with the spatio-temporal matching of CT images acquired at different times, the oncopathological. assessment of morbidity can be documented. A graphical user interface is provided which is also equipped with 3D visualization of the pleura. Our overall aim is to develop an image analysis system for an efficient and reliable diagnosis of early stage pleural mesothelioma in order to ease the consequences of the expected peak of malignant pleural mesothelioma caused by asbestos exposure.
A new approach to segment pleurae from CT data with high precision is introduced. This approach is developed in the segmentation's framework of an image analysis system to automatically detect pleural thickenings....
详细信息
ISBN:
(纸本)9780819470980
A new approach to segment pleurae from CT data with high precision is introduced. This approach is developed in the segmentation's framework of an image analysis system to automatically detect pleural thickenings. The new technique to carry out the 3D segmentation of lung pleura is based on supervised range-constrained thresholding and a Gibbs-Markov random field model. An initial segmentation is done using the 3D histogram by supervised range-constrained thresholding. 3D connected component labelling is then applied to find the thorax. In order to detect and remove trachea and bronchi therein, the 3D histogram of connected pulmonary organs is modelled as a finite mixture of Gaussian distributions. Parameters are estimated using the Expectation-Maximization algorithm, which leads to the classification of that pulmonary region. As consequence left and right lungs are separated. Finally we apply a Gibbs-Markov random field model to our initial segmentation in order to achieve a high accuracy segmentation of lung pleura. The Gibbs-Markov random field is combined with maximum a posteriori estimation to estimate optimal pleural contours. With these procedures, a new segmentation strategy is developed in order to improve the reliability and accuracy of the detection of pleural contours and to achieve a better assessment performance of pleural thickenings.
暂无评论