Radiation-Free Microwave Technology for Breast Lesion Detection Using Supervised Machine Learning Model

Abstract

Mammography is the gold standard technology for breast screening, which has been demonstrated through different randomized controlled trials to reduce breast cancer mortality. However, mammography has limitations and potential harms, such as the use of ionizing radiation. To overcome the ionizing radiation exposure issues, a novel device (i.e. MammoWave) based on low-power radio-frequency signals has been developed for breast lesion detection. The MammoWave is a microwave device and is under clinical validation phase in several hospitals across Europe. The device transmits non-invasive microwave signals through the breast and accumulates the backscattered (returned) signatures, commonly denoted as the S21 signals in engineering terminology. Backscattered (complex) S21 signals exploit the contrast in dielectric properties of breasts with and without lesions. The proposed research is aimed to automatically segregate these two types of signal responses by applying appropriate supervised machine learning (ML) algorithm for the data emerging from this research. The support vector machine with radial basis function has been employed here. The proposed algorithm has been trained and tested using microwave breast response data collected at one of the clinical validation centres. Statistical evaluation indicates that the proposed ML model can recognise the MammoWave breasts signal with no radiological finding (NF) and with radiological findings (WF), i.e., may be the presence of benign or malignant lesions. A sensitivity of 84.40% and a specificity of 95.50% have been achieved in NF/WF recognition using the proposed ML model.

Keywords: MammoWave’s dielectric breast response; X-ray free breast screening; non-invasive lesion detection; radiation-free technology; supervised machine learning.

Figure 1

Proposed flow chart of MammoWave signal classification for the breast lesion detection.

Figure 2

(a) MammoWave device designed by Umbria Bioengineering Technologies (UBT), Italy. (b) Patient’s breast examination procedure with MammoWave. (c) Pictorial view of the transmitting-receiving antenna positions of MammoWave measurement.

Figure 3

A Flowchart of the classification stages involved in the proposed ML experiment. $SV{M}_G$ and $SV{M}_Q$ have been trialled and compared, employing PCs of the complex $S_{21}$ data (raw data). The investigation continued with $SV{M}_G$, comparing its performance with $SV{M}_Q$. Subsequently, real parts and PCs obtained from the real parts of the complex $S_{21}$ have been employed in the second and third stages respectively to classify the NF-WF signals by $SV{M}_G$. The optimal performance attained in the third stage applying PCs of real parts of complex $S_{21}$ with $SV{M}_G$.

Figure 4

Example of significant variance achieved applying PCA on complex $S_{21}$ signals. (a) Significant variance achieved applying PCA on complex $S_{21}$ signals of a NF breast. (b) Significant variance achieved applying PCA on complex $S_{21}$ signals of a WF breast.

Figure 5

NF and WF signal classification results obtained from $SV{M}_Q$ and $SV{M}_G$ applying PCA over MammoWave’s complex $S_{21}$ data. (a) Accuracy obtained from $SV{M}_G$ varying PCs and validation data. (b) Accuracy obtained from $SV{M}_Q$ varying PCs and validation data. (c) Sensitivity obtained from $SV{M}_G$ varying PCs and validation data. (d) Sensitivity obtained from $SV{M}_Q$ varying PCs and validation data. (e) Specificity obtained from $SV{M}_G$ varying PCs and validation data. (f) Specificity obtained from $SV{M}_Q$ varying PCs and validation data. (g) $MCC$ obtained from $SV{M}_G$ varying PCs and validation data. (h) $MCC$ obtained from $SV{M}_Q$ varying PCs and validation data.

Figure 6

NF and WF signal prediction results (accuracy, sensitivity, specificity, and $MCC$) obtained using real parts of MammoWave’s complex $S_{21}$ signals applying $SV{M}_G$ over different amount of validation data.

Figure 7

Example of significant variance achieved applying PCA on real part of complex $S_{21}$ signals. (a) Significant variance achieved applying PCA on real part of complex $S_{21}$ signals of a NF breast. (b) Significant variance achieved applying PCA on real part of complex $S_{21}$ signals of a WF breast.

Figure 8

NF and WF signal classification results obtained from $SV{M}_G$ applying PCA over real-parts of MammoWave’s complex $S_{21}$ data. (a) Accuracy obtained from $SV{M}_G$ varying PCs and validation data. (b) Sensitivity obtained from $SV{M}_G$ varying PCs and validation data. (c) Specificity obtained from $SV{M}_G$ varying PCs and validation data. (d) $MCC$ obtained from $SV{M}_G$ varying PCs and validation data.

Figure 9

ROC curve obtained from $SV{M}_G$ employing principal components of real parts of $S_{21}$ signals for NF-WF signal classification.