https://www.mdu.se/

Microwave imaging is an efficient technique to non-invasively visualizing dielectricproperties of non-metallic bodies. One potential of the technique is the high contrast in dielectricproperties between biological tissues. In the 80’s, Supélec developed a 2.45 GHz planarmicrowave camera, in the 90’s the group developed algorithms for quantitative microwaveimaging. The purpose of this study is to investigate the capability of these existing materials, oran extended version of them, in terms of quantitative imaging of high-contrast inhomogeneousobject for application of breast cancer detection. A two-dimensional formalization is consideredto be followed up with future three-dimensional investigations.

Microwave imaging is recognized as an efficient diagnostic modality for no invasively visualizingdielectric contrasts in non metallic bodies. The usefulness of this modality results from the existingcorrelation between dielectric properties and quantities of practical relevance for industrial orbiomedical applications. At the beginning of the 80’s, Supélec developed a 2.45 GHz planarmicrowave camera, in the 90’s the group developed algorithms for quantitative microwave imaging.The purpose of this study is to investigate the capability of these existing materials, or an extendedversion of it, in terms of quantitative imaging of high contrast inhomogeneous object for application ofbreast cancer detection.

Breast cancer is a global health problem, needing cheep and effective alternative diagnosis methods in order to minimize the mortality. This experimental study is performed in the context of an ongoing collaborative project towards a future planar three-dimensional microwave breast mammography system. Herein the first quantitative image of an inhomogeneous breast tumor phantom, composed by different Triton X-100/water/salt mixtures, is obtained by using the existing planar 2.45 GHz microwave camera.

In this paper an experimental prototype of a robot controlled data acquisition system for microwave imaging is presented, where the transmitting and receiving antennas are immersed in a water-tank. The scattered field from the object under test is acquired by using the robot and scanning a single receiving antenna in cylindrical or half spherical coordinates, while the transmitting antenna is fixed at one position with possibilities to be manually moved to different positions. Careful design and construction of the system has given accurate measurements of incident and total field with a SNR of 45dB. A validation of the robot system is performed by comparing measured and computed data for a sunflower oil object.In this paper an experimental prototype of a robot controlled data acquisition system for microwave imaging is presented, where the transmitting and receiving antennas are immersed in a water-tank. The scattered field from the object under test is acquired by using the robot and scanning a single receiving antenna in cylindrical or half spherical coordinates, while the transmitting antenna is fixed at one position with possibilities to be manually moved to different positions. Careful design and construction of the system has given accurate measurements of incident and total field with a SNR of 45dB. A validation of the robot system is performed by comparing measured and computed data for a sunflower oil object.

Microwave imaging is recognized as a potentialcandidate for biomedical applications, such as breast tumordetection. In this context a planar microwave camera isinvestigated for quantitative imaging of inhomogeneous objects.Promising simulation results indicates that the planar geometryis suitable for quantitative imaging, as long as the signal to noiseratio is higher than 40 dB. Different calibration techniques arediscussed with several proposed model improvements and thefirst reconstructed quantitative image of an inhomogeneous 2Dobject is obtained by using experimental data from the camera.

This paper deals with the breast tumor detectionability of a planar microwave imaging system. Indeed, microwaveimaging seems to have a significant potential in such a biomedicalapplication. By means of a numerical model based upon electricfield volume integral equations, we investigate the influence ofseveral parameters, such as the respective sizes of the breastand the tumor, the operating frequency and the electromagneticproperties of the different media, on the detection ability. Thelatter is estimated by comparing the scattered fields observed inthe presence and in the absence of tumor and by accounting forthe signal to noise ratio available with the experimental setupdeveloped at the laboratory.