Loading...

Investigation, Design and Simulation of Elastography and IOTA Imaging Modes in Ultrasound Systems

Norouzi, Mohammad | 2019

242 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 54755 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Vosoughi Vahdat, Bijan; Kavehvash, Zahra
  7. Abstract:
  8. Ultrasound is one of the most widely used imaging techniques used by physicians and radiologists as a tool. It has many diagnostic and therapeutic uses. In particular, ultrasound imaging is used for prenatal imaging due to relative safety, low cost, nature Non-ionizing, instant display and ease of use by the operator, widely used in many countries around the world. According to the Ministry of Health, ovarian cancer is the most important cause of death in women The effect of cancer in Iran. Breast cancer is also one of the most common cancers in women, followed by cancer Cervix and ovarian and uterine cancers are also among the most common cancers in women. This issue is not only in Iran It is also pervasive in other developed countries and paying attention and the creation of structures for faster and cheaper detection in the early stages of this type of cancer is quite noticeable. Ultrasound is one of the diagnostic methods for this type of cancer. Different methods are used to judge whether tissues are benign or malignant for a long time.The most famous, most commonly found and used ultrasound method is IOTA. To achieve the desired result, we first used the simple network mode and just gave it the images to classify. At first, the size of the images was not uniform, and to solve this problem, in the first case, the tiles of the image were randomly separated And we gave it to the network to classify. But the images were too similar because the image tiles sometimes included unsuspected and irrelevant parts . In the next step, we merged the images and gave them to the network, but the problem with this method was that the size of the images was so large that it caused network training problems or it was so small that the details of the image were lost. So we thought of adding a margin of suspicious texture as network input. Ultrasound image zoning due Technical limitations have their own challenges, and conventional methods often do not. One of the solutions was the use of a trained network to classify another category of images. The latest method for semi-automatic classification is by using the active contour. In this method the physical elastic model is considered for the suspicious tissue boundaries and these boundaries got stretched towards the edges of the tissue to reach its minimum energy. the methods works in the way is that the sonologist must first give an initial bend as the initial state of the algorithm. This method is more accurate and safer than other methods because it is somewhat diagnosed by a sonologist helps
  9. Keywords:
  10. Deep Neural Networks ; Sonography ; Medical Images ; Image Processing ; Cancer Diagnosis ; Medical Imaging

 Digital Object List

 Bookmark

No TOC