Radar-Based Microwave Breast Imaging Using Neurocomputational Models

Abstract

In this study, neurocomputational models are proposed for the acquisition of radar-based microwave images of breast tumors using deep neural networks (DNNs) and convolutional neural networks (CNNs). The circular synthetic aperture radar (CSAR) technique for radar-based microwave imaging (MWI) was utilized to generate 1000 numerical simulations for randomly generated scenarios. The scenarios contain information such as the number, size, and location of tumors for each simulation. Then, a dataset of 1000 distinct simulations with complex values based on the scenarios was built. Consequently, a real-valued DNN (RV-DNN) with five hidden layers, a real-valued CNN (RV-CNN) with seven convolutional layers, and a real-valued combined model (RV-MWINet) consisting of CNN and U-Net sub-models were built and trained to generate the radar-based microwave images. While the proposed RV-DNN, RV-CNN, and RV-MWINet models are real-valued, the MWINet model is restructured with complex-valued layers (CV-MWINet), resulting in a total of four models. For the RV-DNN model, the training and test errors in terms of mean squared error (MSE) are found to be 103.400 and 96.395, respectively, whereas for the RV-CNN model, the training and test errors are obtained to be 45.283 and 153.818. Due to the fact that the RV-MWINet model is a combined U-Net model, the accuracy metric is analyzed. The proposed RV-MWINet model has training and testing accuracy of 0.9135 and 0.8635, whereas the CV-MWINet model has training and testing accuracy of 0.991 and 1.000, respectively. The peak signal-to-noise ratio (PSNR), universal quality index (UQI), and structural similarity index (SSIM) metrics were also evaluated for the images generated by the proposed neurocomputational models. The generated images demonstrate that the proposed neurocomputational models can be successfully utilized for radar-based microwave imaging, especially for breast imaging.

Keywords: breast imaging; circular synthetic aperture radar (CSAR); convolutional neural networks (CNNs); deep neural networks (DNNs); inverse scattering.

Figure 1

Simulation setup for two-dimensional breast tumor imaging (The red arcs from the antenna to the imaging field represent the propagating wave, the gray arrows the scattered field, and the red arrows the backscattered field).

Figure 2

(a) Fantom fabrication, dielectric constant measurement of (b) healthy and (c) tumor phantoms and (d) microwave measurement setup for two-dimensional breast tumor imaging.

Figure 3

Dielectric constants of the fabricated phantoms between 1 GHz and 10 GHz.

Figure 4

The proposed RV-DNN model for microwave medical imaging.

Figure 5

The proposed RV-CNN model for microwave medical imaging (Input data is shown in red color).

Figure 6

The proposed MWINet model for microwave medical imaging (Input data is shown in red color).

Figure 7

Mean squared error (MSE) curves for training and validation phases of the proposed RV-DNN model.

Figure 8

Mean squared error (MSE) curves for training and validation phases of the proposed RV-CNN model.

Figure 9

Accuracy curves for training and validation phases of the proposed RV-MWINet model.

Figure 10

Accuracy curves for training and validation phases of the proposed CV-MWINet model.

Figure 11

Comparison of samples of microwave images generated by the proposed neurocomputational models for train data ((a,g,m) are ground truth images).

Figure 11

Comparison of samples of microwave images generated by the proposed neurocomputational models for train data ((a,g,m) are ground truth images).

Figure 12

Comparison of samples of microwave images generated by the proposed neurocomputational models for test data ((a,g,m) are ground truth images).

Figure 13

Comparison of samples of microwave images generated by the proposed CV-MWINet model for measurement data (metal screw in fine dust).

Figure 14

Comparison of samples of microwave images generated by the proposed CV-MWINet model for measurement data (tumor phantom in healthy phantom).