Parallel Delay Multiply and Sum Algorithm for Microwave Medical Imaging Using Spark Big Data Framework

Abstract

Microwave imaging systems are currently being investigated for breast cancer, brain stroke and neurodegenerative disease detection due to their low cost, portable and wearable nature. At present, commonly used radar-based algorithms for microwave imaging are based on the delay and sum algorithm. These algorithms use ultra-wideband signals to reconstruct a 2D image of the targeted object or region. Delay multiply and sum is an extended version of the delay and sum algorithm. However, it is computationally expensive and time-consuming. In this paper, the delay multiply and sum algorithm is parallelised using a big data framework. The algorithm uses the Spark MapReduce programming model to improve its efficiency. The most computational part of the algorithm is pixel value calculation, where signals need to be multiplied in pairs and summed. The proposed algorithm broadcasts the input data and executes it in parallel in a distributed manner. The Spark-based parallel algorithm is compared with sequential and Python multiprocessing library implementation. The experimental results on both a standalone machine and a high-performance cluster show that Spark significantly accelerates the image reconstruction process without affecting its accuracy.

Multistatic radar-based imaging in layered and dispersive media for biomedical applications

“…For example, the computation time is of paramount importance in the case of stroke detection as the patient seeks urgent help. Various algorithms are used to improve computational efficiency, such as hardware resources, including graphical processing units, field-programmable gate arrays [26] and software frameworks such as Spark and Hadoop [27]. For applications like AD monitoring, the computation time becomes less important than accuracy.…”

Design and Evaluation of Wearable Multimodal RF Sensing System for Vascular Dementia Detection

“…Future work will be done to optimize it further to obtain the exact dimension of the malignant target. The image quality and contrast can also be improved through delay and sum (DAS) [35], [36], delay multiply and sum (DMAS) [37]- [39] and channel-ranked delay and sum (CR-DAS) [40], [41] algorithms.…”

Next Generation Cognition-Aware Hearing Aid Devices With Microwave Sensors: Opportunities and Challenges

“…This data can be used to compare the diagnosis of the target patient and accurately reconcile the results. Recent research has been focused on utilizing a deep learning network [198], PySpark-based reconstruction [199] and parallel delay multiply and sum image reconstruction algorithm [200], which performed more efficiently than the delay multiple and sum algorithm. The experimental results from these researches have shown that the Spark significantly accelerates the imaging reconstruction process without affecting the accuracy.…”