An Optimization-Based Approach to Radar Image Reconstruction in Breast Microwave Sensing

Abstract

Breast microwave sensing (BMS) has been studied as a potential technique for cancer detection due to the observed microwave properties of malignant and healthy breast tissues. This work presents a novel radar-based image reconstruction algorithm for use in BMS that reframes the radar image reconstruction process as an optimization problem. A gradient descent optimizer was used to create an optimization-based radar reconstruction (ORR) algorithm. Two hundred scans of MRI-derived breast phantoms were performed with a preclinical BMS system. These scans were reconstructed using the ORR, delay-and-sum (DAS), and delay-multiply-and-sum (DMAS) beamformers. The ORR was observed to improve both sensitivity and specificity compared to DAS and DMAS. The estimated sensitivity and specificity of the DAS beamformer were 19% and 44%, respectively, while for ORR, they were 27% and 56%, representing a relative increase of 42% and 27%. The DAS reconstructions also exhibited a hot-spot image artifact, where a localized region of high intensity that did not correspond to any physical phantom feature would be present in an image. This artifact appeared like a tumour response within the image and contributed to the lower specificity of the DAS beamformer. This artifact was not observed in the ORR reconstructions. This work demonstrates the potential of an optimization-based conceptualization of the radar image reconstruction problem in BMS. The ORR algorithm implemented in this work showed improved diagnostic performance and fewer image artifacts compared to the widely employed DAS algorithm.

Monostatic microwave imaging of cancerous breast phantom with a frequency to time domain conversion technique

“…The two tumors embedded inside the gland layer have been detected upto a reasonable extent in this work. The proposed process will be used as the pre‐processing stage for several algorithms such as Delay Multiply and Sum, iterative Delay and Sum, and iterative Delay Multiply and Sum 9,10 algorithms in the future, for imaging various cancerous phantoms.…”

Review and Analysis of Tumour Detection and Image Quality Analysis in Experimental Breast Microwave Sensing

“…The majority of estimates have come from patient datasets (see Table A1(v)) with sample sizes varying from 5 patients in [141] up to 225 patients in [143]. Two articles [100,120] estimated the diagnostic performance using breast phantoms, under well-controlled experimental conditions. Notably, the poorest specificity estimates in the literature are found in these works (20% in [100] and as low as 40% in [120]); the majority of trials performed using patient data were not designed to evaluate specificity.…”

“…Two articles [100,120] estimated the diagnostic performance using breast phantoms, under well-controlled experimental conditions. Notably, the poorest specificity estimates in the literature are found in these works (20% in [100] and as low as 40% in [120]); the majority of trials performed using patient data were not designed to evaluate specificity. The works in [47,94,111,141,143,170] all exclusively examined patients with known breast lesions.…”

Designs of three different octagonal-shaped antennas for microwave-based breast cancer detection

“…This filtering used only S-parameters from phantoms to classify phantom textures and reduce the feature data size. Each sample was identified with 35 × 72 features corresponding to 35 responses in the time domain for each of the 72 antenna positions used in the scan [12].…”