Computerized image analysis is used to extract information from images. It can be used in medical applications to determine the size of organs or to build 3-D models of organs before surgery. For example, a PhD candidate at Uppsala University, Sweden, has developed new technology to make easier to diagnose and plan the treatment of cancer. He used haptics technology to develop new interactive methods ‘where the mouse and keyboard are replaced by a pen-like three-dimensional mouse that enables the user to feel the virtual organs.’ His thesis, ‘Visualization and Haptics for Interactive Medical Image Analysis‘ (PDF format, 84 pages, 3.73 MB), is a pleasure to read, but is quite long. If you don’t have enough time, read my summary.
You can see above the Reachin desktop display with a PHANToM Desktop haptic device. A researcher is working with the haptic pen at the specially constructed workstation that can show stereo graphics. (Credit: Erik Vidholm, Uppsala University) Erik Vidholm is the PhD candidate mentioned above and working at the Centre for Image Analysis at Uppsala University. And as he presented his thesis on February 8, 2008, he’s a real PhD by now.
On the images above, you can see on the left a screenshot from the application. The user is seeding the vascular tree in a head and upper thorax data set guided by volume haptics and a hardware accelerated maximum intensityprojection (MIP). On the right is an image of the ‘final gray-scale connectedness segmentation result.’ (Credit: Erik Vidholm, Uppsala University) The images avove have respectively extracted from pages 45 and 52 of Vidholm’s dissertation, which contains dozens of other pictures.
For more details, you can read an Uppsala University news release, ‘Doctors will soon be able to feel organs via a display screen.’ But for more information, here is the introduction of Vidholm’s thesis. “Modern medical imaging techniques provide an increasing amount of high-dimensional and high-resolution image data that need to be visualized, analyzed, and interpreted for diagnostic and treatment planning purposes. As a consequence, efficient ways of exploring these images are needed. In order to work with specific patient cases, it is necessary to be able to work directly with the medical image volumes and to generate the relevant 3D structures directly as they are needed for visualization and analysis. This requires efficient tools for segmentation, i.e., separation of objects from each other and from the background. Segmentation is hard to automate due to, e.g., high shape variability of organs and limited contrast between tissues. Manual segmentation, on the other hand, is tedious and error-prone. An approach combining the merits from automatic and manual methods is semi-automatic segmentation, where the user interactively provides input to the methods. For complex medical image volumes, the interactive part can be highly 3D oriented and is therefore dependent on the user interface.”
And here is the second paragraph of this introduction. “This thesis presents methods for interactive segmentation and visualization where true 3D interaction with haptic feedback and stereo graphics is used. Well-known segmentation methods such as fast marching, fuzzy connectedness, live-wire, and deformable models, have been tailored and extended for implementation in a 3D environment where volume visualization and haptics are used to guide the user. The visualization is accelerated with graphics hardware and therefore allows for volume rendering in stereo at interactive rates. The haptic feedback is rendered with constraint-based direct volume haptics in order to convey information about the data that is hard to visualize and thereby facilitate the interaction. The methods have been applied to real medical images, e.g., 3D liver CT data and 4D breast MR data with good results.”
Vidholm has assembled his methods in a software package that can be freely downloaded so that other researchers in medical image analysis can benefit from them. You can download the software from this link.
Sources: Uppsala University, Sweden, February 2008; and various websites
You’ll find related stories by following the links below.
