Diagnosing Breast Cancer with Microwave Technology: remaining challenges and potential solutions with machine learning.

Bárbara Oliveira,Martin O’Halloran,Daniela Godinho,Edward Jones,Raquel Conceição,Martin Glavin

doi:10.3390/diagnostics8020036

Bárbara Oliveira, Martin O’Halloran + Show 4 more

Open Access

https://doi.org/10.3390/diagnostics8020036

Copy DOI

Abstract

Currently, breast cancer often requires invasive biopsies for diagnosis, motivating researchers to design and develop non-invasive and automated diagnosis systems. Recent microwave breast imaging studies have shown how backscattered signals carry relevant information about the shape of a tumour, and tumour shape is often used with current imaging modalities to assess malignancy. This paper presents a comprehensive analysis of microwave breast diagnosis systems which use machine learning to learn characteristics of benign and malignant tumours. The state-of-the-art, the main challenges still to overcome and potential solutions are outlined. Specifically, this work investigates the benefit of signal pre-processing on diagnostic performance, and proposes a new set of extracted features that capture the tumour shape information embedded in a signal. This work also investigates if a relationship exists between the antenna topology in a microwave system and diagnostic performance. Finally, a careful machine learning validation methodology is implemented to guarantee the robustness of the results and the accuracy of performance evaluation.

Highlights

With the algorithms implemented in Stage 3, this study aims to understand: (1) if the angular distance between the transmit and receive antennas in a channel determines its predictive power; (2) if the distance between the tumour and the channel has an impact on diagnostic performance; (3) how to better use the information from each channel while adhering to best machine learning practices
Microwave breast diagnosis systems could play a key role in further establishing microwave breast imaging and diagnosis as a tool for continuous and safe breast cancer monitoring
While diagnosis of breast tumours as benign or malignant could theoretically be performed through a number of avenues, shape and spiculation at the margin of a tumour are widely accepted as markers for malignancy, and previous studies have already demonstrated how backscattered signals are influenced by the shape of a tumour

Summary

Introduction

Microwave Breast Imaging (MBI) for breast cancer detection has seen significant academic and commercial development in recent years. The development of automated breast diagnosis systems is relevant to the clinical environment, considering recent reports showing minimal benefit of continuous mammographic screening in terms of long-term survival rates [13,14]. Many automated breast diagnosis systems have been proposed, and usually integrate signal or image pre-processing and segmentation, and diagnosis through machine learning [15,16]. Such systems have proved useful in aiding clinicians diagnose

Objectives

Methods

Results

Conclusion