Scholarly record
USING IMAGE PROCESSING SOFTWARE DEVELOPED FOR B-IMAGING ON MR IMAGES BASED ON THRESHOLDING ALGORITHM IN A SELECTABLE REGION OF INTEREST
Abstract
In 2011, we developed our application for B-imaging based on binary thresholding algorithm in a selectable Region of Interest. This application has been used in many studies focused on ultrasound B-imaging in neurology; primarily for substantia nigra echogenicity examination and also for some different brain structures, e.g. mediotemporal lobe atrophy investigation. One of principal output value is an echogenicity index which is used as a marker of the echogenicity grade. Many published studies showed that is well applicable in B-imaging analysis. The lower index corresponds to the lower echogenicity grade and vice versa. Due to the principle of the thresholding of the images in gray-scale, the application could be also used for MR images. Thus, we have analyzed totally of 52 MR images to observe changing of the echogenicity index. In MR images, we have measured a dark and a light ROI area similarly as for different echogenicity of US image. We use the same descriptor as in the case of US images; the echogenicity index. In this study, images in DCM and IMA formats were analyzed. In result, the echogenicity index is significantly different for light and dark area. Thus, the application could be well-applicable also for MR images analysis. This is not a clinical study; we use the same ROI independently on examined structure. The ROI is placed on any dark and on any light area only to observe changing the echogenicity index. We have analyzed two features: the difference between dark and light area and the difference between two non-experienced observers. In all cases, the echogenicity index was higher for light than for dark area and in many cases more than triple. The difference between observers for the same image and the same ROI is insignificant. Minimal value of the index was 9.44 and maximal achieved was 105.67. In majority, the index > 50 for light area and under 50 for dark area. This is key feature that echogenicity index could be applied as a marker also for MR images.
Publication Impact Profile
- Captures
- Mendeley - Readers: 2
Publication details
References1
1. Blahuta, J., Cermak, P. An Expert System Based on Using Artificial Neural Network and Region-Based Image Processing to Recognition Substantia Nigra and Atherosclerotic Plaques in B-Images: A Prospective Study. 14th International Work-Conference on Artificial Neural Networks, IWANN 2017, Cadiz, Spain, June 14-16, 2017, Proceedings, Part I. Lecture Notes in Computer Science 10305, Springer 2017, pp 236-245.
View or Download full articleAccess options
SWS access login
Login as SWS Scientific CommitteeLogin as SWS Scientific PartnerLogin as SWS AuthorAuthors and approved SWS contributors will read and export their own linked papers after identity matching by SWS profile, email and SGEM GlobalID.
For librarian assistance: [email protected]
Purchase Instant Access
- Article can be downloaded after successful payment.
- Article may be used according to SWS library access terms.
- Article cannot be redistributed.