MRI Soft Visualization
Neuro-Fuzzy Visualization Scheme in MRI
MRI Knowledge Hub
What is Soft Segmentation or Soft Visualization
The radiological interpretation of image has undergone a paradigm shift towards computer-aided visualization. Today visualization tools (2-D and 3-D) are extensively used at the workstation by a radiologist while reporting. These tools are equipped with Dicom compatibility besides a number of post-processing features to enhance image quality. The technical aspects of the image acquisition that affect image quality have been the major focus of technology development. A large amount of effort has gone into innovating and improving modalities to produce images with high spatial resolution, high contrast, less noise. Computational methods for manipulating and visualizing medical images have been improved to meet the demand of an ever-increasing influx of image data. But the interpretation part by lot of radiologist is still dependent on the film box and films due to ease in operations or lack of good visualization tools. Window width and Window level is one of the main tool exploited by the radiologist in MR while diagnosing and reporting the patient. This tool operates between lower and upper threshold values to map the range of data values. Whereas the anatomical and pathological boundaries in MR image varies due to many reasons like MR system, acquisition parameters, time after the pathology changes etc. So the tool should be capable of visualizing soft segments in the image, which will improves the interpretation in demarcating tissue boundaries. Soft Segmentation and Soft visualization generate different interpretation of image (normal and diseased cases).
Why MRI Soft Visualization?
Interpretation of 2D MR images is important for surgical planning, treatment strategies and therapy for cases of cancer, multiple sclerosis lesions and other diseases like Alzheimer’s disease, Dementia in brain. An image visualization tool for diagnostic use can shorten the time required for diagnosis, improve diagnostic accuracy and increase the diagnosis throughput of the doctor. Identifying different types of tissues in the MR image of brain is the most challenging problem for interpretation and diagnosis. Difficulties in the 2D MR image based identification of brain tissues arise due to:
- Partial volume averaging: more than one type of tissue can lie within finite spatial extent of an imaging system’s point spread function;
- Tissue inhomogeneity and non-uniformity which leads to fuzziness at the boundary regardless of the imaging system’s quality;
- Variability in tissue types; and
- Imaging variabilities.
Limitation of Window Width and Window Level Scheme
One of the drawbacks in the present windowing system is that, window width and window level defines the narrow view to the data’s scalar values, for enhancing the contrast of certain structures within the image data. The linear function by changing WW and WL sets the display window to enhance portion of the image with optimum image contrast and brightness. The gray variations of remaining portion of the image will be suppressed. On the other hand MR image tissue boundaries causing intensity variations are soft and change due to many reasons like patient position, different pulse sequence, type of system etc.
Neuro Fuzzy Visualization Scheme
An interactive segmentation based visualization tool can enable a user to circumvent these difficulties with the help of his anatomical knowledge and experience. A Soft visualization scheme explained, which exploits fuzzy logic for generating alternate and selective segmentation of the 2D MR image for detailed analysis by the user. Nature of the possible views and segments in our system are not dependent upon the choice of morphological operators. A neural network based clustering scheme is used to determine the possible views. This neuro-fuzzy visualization system will help radiologists to obtain a better view of the tissue anatomy while analyzing MR images.