Microwave based breast cancer detection
Microwave based breast cancer detection
Breast cancer is the most common form of cancer amongst women in Sweden. The prognosis for the treatment is very good if the cancer is found at an early stage. X-ray mammography is successfully used for screening. Although about 20% of the tumors cannot be detected with this technique. In a MedTech West project a microwave based breast cancer imaging system is being developed, with the aim of improving the detection rate.
Our technique is based on the existence of a large contrast in dielectric properties between tumors and healthy tissue. Microwave measurements are performed and used in an algorithm to reconstruct internal images of the breast. The ability to create 3D images makes it easier to identify tumors. However, this technology is very computationally heavy and it is not until recent years that the required computational resources have become available. An advantage for the patient is that this technique needs neither breast compression nor the use of ionizing radiation. Thanks to the large contrast between tissues, our technique has the potential to become both sensitive and specific to the tumors.
Microwave technology for medical diagnostics is sensitive to variations in liquid and ion concentration between different tissues and disease-states. This is also the case for a cancer tumor inside normal tissue where these properties vary tocreate a detectable contrast. The propagation patterns of microwaves are however much more complex than that of for example x-rays, and this also makes the analysis of the measurements harder. In making the diagnosis the measured scattering pattern must be related to the corresponding tissue properties, i.e. the dielectric properties. We generate a dielectric image of the internal dielectric properties from the measured data. In mathematical terms this problem is both non-linear and ill-posed and this poses particular challenges on the algorithm in terms of the need for massive computational resources to reach the desired accuracy. The motivation for the research is on the otherhand driven by the potential to develop a more accurate modality than the x-ray mammography that is commonly used today for breast cancer screening. Our solution utilizes a microwave antenna array that is surrounding the breast, a microwave send-and-receive unit and an image reconstruction algorithm. With our system we will be able to detect tumors that are not seen with x-ray mammography.
We have developed a microwave antenna system connected to a microwave unit and a signal-processing algorithm. The algorithm and system has been tested and verified in the lab on phantom material. Work is now made to upgrade the system into a clinical prototype that will be used for tests on patients. In our work to reconstruct images it has become clear that it is essential to base the image reconstruction algorithm on an accurate three-dimensional electromagnetic model of the antenna array. This will unfortunately make the computational burdenvery large. Earlier work with two-dimensional models are significantly morecomputational efficient but instead unable to generate images of sufficiently accuracy and detail level. Experiences and results obtained in this projectwill be very valuable also in our other research projects based on microwave technology.
The final breast cancer detection system will provide a way to improve the detection of cancer tumors. Whether it will replace or complement the x-ray technology used today remains to be seen after the clinical evaluations. The examination will be painless and the patient will not be exposed to ionizing radiation. In the future we can also foresee a large number of other application areas where the microwave technology can improve the diagnostics.
MedTech West partner
Dr. Andreas Fhager, Department of Signal Processing and Biomedical Engineering, S2, Chalmers University of Technology.
Technical research partners
Prof. Mikael Persson, Department of Signal Processing and Biomedical Engineering, S2, Chalmers University of Technology.
Dr. Hoi-Shun Lui, Department of Signal Processing and Biomedical Engineering, S2, Chalmers University of Technology.
Dr. Johanna Gellermann, MD, Berlin
For more information on this project, contact Henrik Mindedal at email@example.com