Electrical Impedance Tomography of Breast Cancer
Abstract
In screening of breast cancer, once abnormalities or lesions are discovered by the X-ray mammogram, generally, other imaging techniques are needed as an adjunct to diagnose the lesion as benign or malignant ft has been shown that cancer cells exhibit altered local electrical impedance. However, existing technology to measure the electrical impedance of the breast relies on a device that has poor spatial resolution. We proposed to map the impedance distribution in the tissue with high spatial resolution, by using it in conjunction with MRI to improve diagnostic accuracy of screening. During the first phase of this project, we developed necessary MRI pulse sequences, the software to analyze acquired images to extract impedance information and hardware components to inject electrical current synchronized with the MRI scanner. Several different pulse sequences were tested to determine their sensitivity and specificity. A spin-echo based sequence with alternating current injection at ^200Hz was found to be the most efficient among the tested. Various gel phantoms are prepared mimicking the electrical impedance variations in tissues. With these phantoms, spatial resolution and sensitivity of the method to various current amplitudes and impedance perturbations are measured. The method is also tested on a single animal with a malignant tumor. Results demonstrate that the proposed method clearly distinguishes objects separated by 1-2mm providing a good resolution and it can be used with safe current amplitudes (down to 0.5-ImA). Moreover, the method was able to distinguish 50% impedance variations. Therefore, these results support our hypothesis that MRI-impedance imaging can be used to distinguish benign and malignant tissues. In the second phase of the project, we will perform experiments on tumor induced rats.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 2003
- Accession Number
- ADA418049
Entities
People
- L. T. Muftuller
Organizations
- University of California, Irvine