Elliptical ultra-wideband antenna sensor with modified time reversal algorithm for breast tumour localization
Breast cancer is a serious health problem and the second leading cause of death around the globe. The skin around fatty tissues like the breast are the most common and fastest growing of all cancer types. Thus, ultra-wideband (UWB) antenna sensors are considered a promising instrument in microwave i...
Saved in:
| 總結: | Breast cancer is a serious health problem and the second leading cause of death around the globe. The skin around fatty tissues like the breast are the most common and fastest growing of all cancer types. Thus, ultra-wideband (UWB) antenna sensors are considered a promising instrument in microwave imaging tomography with the least destruction in the body’s tissue. Several studies have recommended UWB antennas to acquire the target’s characteristics and localize them. However, they showed some limitations in terms of bandwidth and resolution to even spending larger printing areas despite showing acceptable outcomes. In addition, the imaging algorithms are usually utilized to combine and sum all the received signals, create an image of the dielectric scatter (tumour) within the human organ, and remove the clutters and artifacts. However, the problem is how to apply and utilize an algorithm for microwave imaging to reconstruct the image of the target in a cluttered environment (like skin and breast) with the most possible artifact and clutter removal (not to mask the tumour response). Therefore, this thesis proposes a UWB antenna sensor with good directional beamwidth, high fidelity, high efficiency, and low group delay at the frequency range of 0-30 GHz. A novel weighted Time Reversal algorithm is introduced to improve the image quality along with removing the clutter in the imaging environment. The proposed elliptical patch antenna sensors loaded by stubs, slots, and truncated ground show complementary results for the imaging of breast skin as they achieve broad BW (>18 GHz), simple elliptical shape of the patch, miniaturized dimensions (15 × 15 mm2), and high fidelity (> 90 % hence low distortion). Notably, there is a good agreement shown between the simulation and measurement results of the proposed antenna. The imaging results show that the proposed algorithm obtained better results in terms of accurate localization, and better removal of image artifacts and clutter. It has shown the accuracy with more than 95 % detection of tumours in breast skin and can perform a hollow with a diameter of 3 mm in any location within the trunk. Besides, no significant discrepancy exists between the images using simulated and measured scattering indicates the system’s ability to the detection of tumours in the breast skin. |
|---|