RESUMO
Detecting tubular structures such as airways or vessels in medical images is important for diagnosis and surgical planning. Many state-of-the-art approaches address this problem by starting from the root and progressing towards thinnest tubular structures usually guided by image filtering techniques. These approaches need to be tailored for each application and can fail in noisy or low-contrast regions. In this work, we address these challenges by a two-layer model which consists of a low-level likelihood measure and a high-level measure verifying tubular branches. The algorithm starts by computing a robust measure of tubular presence using a discriminative classifier at multiple image scales. The measure is then used in an efficient multi-scale shortest path algorithm to generate candidate centerline branches and corresponding radii measurements. Finally, the branches are verified by a learning-based indicator function that discards false candidate branches. The experiments on detecting airways in rotational X-ray volumes show that the technique is robust to noise and correctly finds airways even in the presence of imaging artifacts.
Assuntos
Algoritmos , Inteligência Artificial , Imageamento Tridimensional/métodos , Pulmão/diagnóstico por imagem , Reconhecimento Automatizado de Padrão/métodos , Interpretação de Imagem Radiográfica Assistida por Computador/métodos , Tomografia Computadorizada por Raios X/métodos , Análise Discriminante , Humanos , Intensificação de Imagem Radiográfica/métodos , Reprodutibilidade dos Testes , Sensibilidade e EspecificidadeRESUMO
Long acquisition times lead to image artifacts in thoracic C-arm CT. Motion blur caused by respiratory motion leads to decreased image quality in many clinical applications. We introduce an image-based method to estimate and compensate respiratory motion in C-arm CT based on diaphragm motion. In order to estimate respiratory motion, we track the contour of the diaphragm in the projection image sequence. Using a motion corrected triangulation approach on the diaphragm vertex, we are able to estimate a motion signal. The estimated motion signal is used to compensate for respiratory motion in the target region, for example, heart or lungs. First, we evaluated our approach in a simulation study using XCAT. As ground truth data was available, a quantitative evaluation was performed. We observed an improvement of about 14% using the structural similarity index. In a real phantom study, using the artiCHEST phantom, we investigated the visibility of bronchial tubes in a porcine lung. Compared to an uncompensated scan, the visibility of bronchial structures is improved drastically. Preliminary results indicate that this kind of motion compensation can deliver a first step in reconstruction image quality improvement. Compared to ground truth data, image quality is still considerably reduced.