: This paper describes a method for the enhancement of curvilinear structures such as vessels and bronchi in 3D medical images. A 3D line enhancement filter is developed with the aim of discriminating line structures from other structures and recovering line structures of various widths. The 3D line filter is based on a combination of the eigenvalues of the 3D Hessian matrix. Multi-scale integration is formulated by taking the maximum among single-scale filter responses, and its characteristics are examined to derive criteria for the selection of parameters in the formulation. The resultant multi-scale line-filtered images provide significantly improved segmentation and visualization of curvilinear structures. The usefulness of the method is demonstrated by the segmentation and visualization of brain vessels from MRI (magnetic resonance imaging) and MRA (magnetic resonance angiography), bronchi from a chest CT, and liver vessels (portal veins) from an abdominal CT. Keywords: 3D image ...

The World Wide Web as a repository of information has had a great influence on our lives. This influence is increasing as the web introduces applications in addition to information. These applications have several advantages, such as world wide accessibility, distance group learning and collaboration. Furthermore, the web encourages training applications since it offers multi-media that can support all stages of training. On the other hand, the virtual reality technology has been utilised to provide new systematic training methods for surgical procedures. These solutions are usually expensive in terms of cost and computation. In this thesis we propose a novel solution to fulfill the training needs of radiologists performing one type of minimally invasive surgery known as interventional radiology. Our training method combines the capabilities of virtual reality to provide realistic simulation environment together with the web environment to provide platform independent, scalable and acc...

This paper proposes a method for automated extraction of the aorta and pulmonary artery (PA) in the me- diastinum of the chest from uncontrasted chest X-ray CT images. Since the aorta and PA areas show low intensity contrast in the mediastinum, it is difficult to extract those areas by a procedure based on CT values. The proposed method employs a model fitting technique to use shape features of blood vessels for extraction. First, edge voxels are detected based on the standard deviation of CT values. A likelihood image, which shows the degree of "likelihood" on medial axes of vessels, are calculated by applying the Euclidean distance transformation to non-edge voxels. Second, the medial axis of each vessel is obtained by fitting the model. This is done by referring the likelihood image. Finally, the aorta and PA areas are recovered from the medial axes by executing the reverse Euclidean distance transformation. We applied the proposed method to seven cases of uncontrasted chest X-ray CT images and evaluated the results by calculating the coincidence index computed from the extracted regions and the regions manually traced. Experimental results showed that the extracted aorta and the PA areas coincides with manually input regions with the coincidence indexes values 90 % and 80-90 %, respectively.

Abstract. Image diagnosis is the task in which a physician searches for abnormal findings on the medical images that potentially involve a large amount of anatomical and pathological information, and identifies diseases from such findings. Since this task aims to reduce the risk (e.g., insult or death) that will occur in the near future, it can be seen as a case of chance discoveries. To investigate the cognitive process underlying the task, we conducted an experiment in which two independent physicians diagnosed the same CT images. Analysis of disagreements about final diagnosis and verbal protocols revealed that when observing the CT images on which one disagreed with the other, the physicians engaged in iterative cycles of the image-feature search and the schema construction. The result suggests the importance of investigating cyclic interactions between humans and environments when considering chance discovery. 1