Abstract

A few emerging medical imaging methods are being developed for breast imaging. Electrical impedance tomography (EIT) is an excellent candidate for safe, low cost, and noninvasive breast cancer monitoring. Despite early promises, the EIT faces a few challenges for the breast imaging application. It is mainly due to its limited resolution and especially for the tumors in depth. However, unlike the other medical applications of EIT, such as brain and thorax, breast tissues are deformable. This article exploits the deformation of breast shape to enhance the EIT resolution and its depth detection. Exterior boundary of the breast can be used to create deformable EIT with multiple shapes to enhance the imaging resolution, turning a challenge to an opportunity. With deformation of the boundary shape, more independent measurements can be obtained, and hence, more information can be gained. This can increase the resolution of the reconstructed image and possible detection for smaller tumors in depth. This article demonstrates it by experimental verification in phantom test representing tumor size inclusion deep inside the breast by a few deformed shape phantoms. To evaluate the experimental results, 3-D printed phantoms are built in several different shapes. Quantitative image analysis shows that some of the deformed shapes are superior to a traditional circular cross section. Additionally, we proposed a combination of data from all shapes so that all this information can be used in one step reconstruction to achieve higher imaging accuracy.

Original languageEnglish
Article number9025058
JournalIEEE Sensors Letters
Volume4
Issue number4
Early online date5 Mar 2020
DOIs
Publication statusPublished - 30 Apr 2020

Keywords

  • Sensor systems
  • breast cancer diagnosis
  • deformed boundaries in electrical impedance tomography (EIT)

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Instrumentation

Fingerprint Dive into the research topics of 'Combining Multiple Boundary Shapes in Deformable EIT a Potential Use in Breast Imaging'. Together they form a unique fingerprint.

Cite this