ANALYSIS is a CAD system designed and developed in the Biomedical Simulations and Imaging Laboratory, School of Electrical and Computer Engineering, National Technical University of Athens, Greece. The system incorporates the main parts of a typical CAD system and has been used to support the diagnosis of vascular disease (carotid atherosclerosis). ANALYSIS can be installed on PC-based platforms operating under the Microsoft Windows operating system. Minimum requirements in processor speed and RAM are 600 MHz and 256 MB, respectively. Using these minimum requirements, motion analysis of two ROIs using block-matching takes 20 min.
The main user interface of ANALYSIS is shown in Fig. 1.7.1. The two main windows of the interface are designed to support motion and texture analysis. Tool buttons and menu bars allow the user to select the parameters of the analysis, e.g. the number of investigated ROIs, the size and shape of ROIs, the inter-pixel distance for the calculation of second-order statistics and neighbourhood grey tone difference matrix, etc. The system includes a friendly wizard for importing results files, which can be subsequently analysed using ANOVA statistics and clustered into predefined groups using fuzzy c-means.
An important feature of ANALYSIS is the possibility of on-line tele-collaboration between health care professionals under a secure framework. The continuous advances in telecommunications allow exchange of information not only throughout but also beyond the hospital as well as real-time collaboration between remote users . Thus, with the introduction of computer-aided diagnosis, a wide area of possible telematics applications for vascular disease can be identified. These include support of the diagnosis procedure [11, 28] and continuous medical education in arterial wall disease . The collaborative environment of ANALYSIS extends the capabilities of the stand-alone CAD system, giving local and remote health care professionals the opportunity to co-operate for the delineation and image analysis of ROIs in ultrasound images. The implementation is based on a point-to-point communication scheme following the "What You See Is What I See" (WYSIWIS) paradigm. To support the collaborative environment, both asynchronous (off-line transfer of data sets) and synchronous (on-line collaboration) data exchange schemes are employed. The term "off-line" indicates the absence of user's interaction with ANALYSIS and involves the transfer of large datasets. A sequence of B-mode ultrasound images after lossless compression requires 5-20 MB of storage depending on the number of images.
Was this article helpful?