Application of the migration method for radiotomography of breast cancer

A new method to visualize microwave image is presented to early non-invasive detection of breast cancer tumors. A new processing method to compute microwave images of heterogeneity in a biological environment is described here, as well as a new algorithm for accelerating the calculation of three-dimensional radio images. Sounding of synthetic phantoms with dielectric properties of breast tissue was carried out in the range of 2–8 GHz using a special radar system developed by the authors. Results show that it is possible to use this microwave imaging method to form 3D accurate images using hemispherical scan Images of tumor phantoms were obtained during probing in the 2–8 GHz range with a resolution of about 5–7 mm. According to the results of the reconstruction of three-dimensional radio images, it was revealed that the calculation time can be reduced by at least 2 times with an insignificant loss of quality.

Different Explanations for the Cosmic Microwave Background Radiation

In this research, the other reasonable explanations for the cosmic microwave background radiation is revealed. Due to the microwave resolution, it very roughly shows the image of galaxies in the universe. Moreover, the intensity measurement on each pixel of the image is the sum of the incident microwaves from different directions, so the microwave image cannot represent the microwave sources clearly far away from the Earth. Hence, we propose a simulation after removing several strongest microwave sources, the remaining microwave radiation sources can establish a very uniform intensity distribution over a range of several ten light years. On the other hand, Sloan Digital Sky Survey reveals 200 million galaxies in the universe and, in fact, only to eliminate the contributions of all galaxies from the microwave image is impossible. The way to further obtain the fine-scale structure by only removing the few strongest microwave sources as the foreground effect will keep the other contributions from all the rest galaxies and stars. Therefore, the Cosmic Microwave Background cannot be uniquely explained the radiation which was left after the initial formation of the universe. Moreover, it is the mainly residual radiation from the un-calculated galaxies and inaccurate estimation of the microwave source strength.

PySpark-Based Optimization of Microwave Image Reconstruction Algorithm for Head Imaging Big Data on High-Performance Computing and Google Cloud Platform

For processing large-scale medical imaging data, adopting high-performance computing and cloud-based resources are getting attention rapidly. Due to its low–cost and non-invasive nature, microwave technology is being investigated for breast and brain imaging. Microwave imaging via space-time algorithm and its extended versions are commonly used, as it provides high-quality images. However, due to intensive computation and sequential execution, these algorithms are not capable of producing images in an acceptable time. In this paper, a parallel microwave image reconstruction algorithm based on Apache Spark on high-performance computing and Google Cloud Platform is proposed. The input data is first converted to a resilient distributed data set and then distributed to multiple nodes on a cluster. The subset of pixel data is calculated in parallel on these nodes, and the results are retrieved to a master node for image reconstruction. Using Apache Spark, the performance of the parallel microwave image reconstruction algorithm is evaluated on high-performance computing and Google Cloud Platform, which shows an average speed increase of 28.56 times on four homogeneous computing nodes. Experimental results revealed that the proposed parallel microwave image reconstruction algorithm fully inherits the parallelism, resulting in fast reconstruction of images from radio frequency sensor’s data. This paper also illustrates that the proposed algorithm is generalized and can be deployed on any master-slave architecture.

On Microwave Breast Imaging with Ultrasound Spatial Priors

We investigate the use of ultrasound images as prior structural information (also known as ultrasound spatial priors) to guide microwave breast imaging so as to enhance its achievable complex permittivity images. In the main approach considered herein, the edges within the discretized<br>ultrasound compressibility image are fed as spatial priors<br>into a microwave imaging algorithm. It is shown that this<br>method requires minimal post-processing of the ultrasound<br>image and can enhance the achievable microwave image<br>accuracy. It is also demonstrated that small tumours can<br>still go undetected in microwave breast imaging using this<br>method if their edges are missing from the spatial priors.

On Microwave Breast Imaging with Ultrasound Spatial Priors

We investigate the use of ultrasound images as prior structural information (also known as ultrasound spatial priors) to guide microwave breast imaging so as to enhance its achievable complex permittivity images. In the main approach considered herein, the edges within the discretized<br>ultrasound compressibility image are fed as spatial priors<br>into a microwave imaging algorithm. It is shown that this<br>method requires minimal post-processing of the ultrasound<br>image and can enhance the achievable microwave image<br>accuracy. It is also demonstrated that small tumours can<br>still go undetected in microwave breast imaging using this<br>method if their edges are missing from the spatial priors.

An Improved Holographic Microwave Breast Imaging Based on Deep Neural Network

Microwave imaging offers excellent potential for breast cancer detection. Deep learning is state-of-the-art in biomedical imaging, which has been successfully applied for biomedical image classifications. This paper investigates a deep neural network (DNN) based classification method for identifying breast lesion in holographic microwave image (HMI). A computer model is developed to demonstrate the proposed method under practical consideration. Various experiments are carried out to evaluate the proposed DNN-based HMI for breast lesion classification. Results have shown that the proposed method could serve as a helpful imaging tool for automatically classifying different types of breast tissues.

Deep Convolutional Neural Networks for Breast Image Analysis on Holographic Microwave Imaging

This paper presents the development of a deep convolutional neural network (CNN) method namely super-solution CNN to produce a high-resolution microwave breast image from a low-resolution model, which helps to improve the accuracy and efficiency of breast lesion detection within microwave image. Various experiments are conducted to validate the proposed method. Experimental results show that the proposed approach has the potential to produce a high-resolution breast image with high-accuracy.