Microwave system for breast tumor detection: experimental concept evaluation

A time‐domain microwave imaging system with improved image quality is presented for early detection of breast cancer. The time‐domain system has the advantage of short scan time compared with the frequency‐domain system but suffers from low signal‐to‐noise ratio (SNR) due to high attenuation of the RF signal within the breast tissue. Signal averaging of repeated measurements can improve the SNR, but this method also smears the signal peaks due to the signal‐averaged jitter, which adversely affects the reconstructed image quality. An image enhancement method using Gaussian bandpass filtering (BPF) is proposed for a time‐domain imaging system. The results from the proposed method were compared with those from the deconvolution method, which is typically used to remove the effects of signal‐averaged timing jitter in short pulse measurements. Simulation results with a 3D hemispherical homogeneous breast model containing tumour cells revealed improvement of the reconstructed image quality with the proposed method in the presence of timing jitter. The signal‐to‐mean ratio was increased from 6.46 to 14.15 with BPF, whereas deconvolution resulted in an increase to 9.36. Simulation results with a non‐homogeneous realistic breast model also showed image enhancement with the proposed method.

A novel focused active microwave system is investigated for detecting tumors in the breast. In contrast to X-ray and ultrasound modalities, the method reviewed here exploits the breast-tissue physical properties unique to the microwave spectrum, namely, the translucent nature of normal breast tissues and the high dielectric contrast between malignant tumors and surrounding lesion-free normal breast tissues. The system uses a pulsed confocal technique and time-gating to enhance the detection of tumors while suppressing the effects of tissue heterogeneity and absorption. Using published data for the dielectric properties of normal breast tissues and malignant tumors, we have conducted a two-dimensional (2-D) finite-difference time-domain (FDTD) computational electromagnetics analysis of the system. The FDTD simulations showed that tumors as small as 2 mm in diameter could be robustly detected in the presence of the background clutter generated by the heterogeneity of the surrounding normal tissue. Lateral spatial resolution of the tumor location was found to be about 0.5 cm.

Early detection of breast cancer is known to be a key factor in the successful treatment of the disease. Here, we present a detection technique complementary to the currently used modalities (primarily mammography, ultrasound and magnetic resonance imaging). Our time-domain breast cancer detection system transmits microwave-range pulses into the breast and records the scattering off of the breast in order to detect malignancies. This method is made possible by an intrinsic contrast in the dielectric parameters, specifically the relative permittivity and conductivity, of the healthy and malignant breast tissues over the microwave frequency range. The long-term goal of our work is to develop a system that can be used periodically to monitor for unusual changes in breast tissues; for instance, healthy breasts would be scanned, and follow-up scans at regular intervals would detect any small changes in breast tissue composition that could indicate the presence of a malignant growth. At that point, the patient would be referred to see a doctor for further investigation of the abnormal results. Such a system would compare each new scan with previous ones to determine the level of tissue changes, and would be used by patients at home. We report feasibility and performance tests for our initial system, conducted with breast phantoms made up of tissue-mimicking materials (unique skin, fat, gland and tumor mixtures). We initiated the system testing with simple homogeneous phantoms, consisting solely of adipose tissue. Then, we extended our tests to cases of increasing complexity by adding a skin layer and varying percentages of glandular structures and tumor sizes. In order to optimize the experimental system, we performed tests with multiple antenna arrangements, tumor sizes and locations. This work shows that there are specific antenna arrangements that are advantageous for tumor detection and demonstrates the capabilities of our time-domain microwave breast tumor detection system.

The innovative knowledgebase system enables the healthcare system to address patient’s customized requirements. Developing and implementing this system with context-aware modifications is a challenge for developing countries, where it can change the healthcare delivery landscape from diagnosis to treatment. Although many studies have been conducted on the development of a knowledge-based system for early breast cancer detection using a rule-based approach, they lack contextual factors such as cultural aspects and socioeconomic features. This paper proposes a contextually aware machine learning (ML)-based knowledge base system for early breast cancer detection in the Pakistani context. Established risk and preventive breast cancer factors were identified from an extensive literature review and validated by academic and industry experts. Contextual factors were investigated with the assistance of field experts, and predictions from the ML algorithm were duly validated. The nested if-then rule, based on established risk and preventive factors, was combined with ML prediction to form the foundation of the proposed knowledgebase system. This research will improve the performance of healthcare systems in terms of accuracy and cost-effectiveness.

The physical basis for breast tumor detection with microwave imaging is the contrast in dielectric properties of normal and malignant breast tissues. Confocal microwave imaging involves illuminating the breast with an ultra-wideband pulse from a number of antenna locations, then synthetically focusing reflections from the breast. The detection of malignant tumors is achieved by the coherent addition of returns from these strongly scattering objects. In this paper, we demonstrate the feasibility of detecting and localizing small (<1 cm) tumors in three dimensions with numerical models of two system configurations involving synthetic cylindrical and planar antenna arrays. Image formation algorithms are developed to enhance tumor responses and reduce early- and late-time clutter. The early-time clutter consists of the incident pulse and reflections from the skin, while the late-time clutter is primarily due to the heterogeneity of breast tissue. Successful detection of 6-mm-diameter spherical tumors is achieved with both planar and cylindrical systems, and similar performance measures are obtained. The influences of the synthetic array size and position relative to the tumor are also explored.

We have previously reported investigations of a novel pulsed confocal microwave system for detection of breast tumors (Hagness et al., 1997). In this paper, we extend this work by identifying a frequency window for optimum operation of the tumor detection system. The finite difference time-domain (FDTD) method is used to study a simplified 2-D breast-tissue geometry adjacent to an elliptical reflector antenna. First, power density results are presented to demonstrate the focusing abilities of the reflector antenna for sinusoid excitation at 3, 6 and 9 GHz. Second, a tumor located at the in-breast focus of the antenna is included in the geometry in order to observe the pulse response of the detector as a function of the tumor size at center frequencies of 3, G and 9 GHz.

Breast cancer is the second most common type of cancer among women in the USA, and it is very common to appear in its invasive form. Detecting its presence in the early stages can potentially aid in the mortality rate depletion since at that point large tumours are highly unlikely to have developed. Technological advances of the last decades have provided advanced tools that employ machine learning for early detection. Common techniques include tumour imaging using special equipment that in most cases is not widely accessible. In order to overcome this limitation, new techniques that employ blood-based biomarkers are being explored. In the current work machine learning algorithms are exploited for the development of a decision support system for breast cancer using easily obtainable user information, age, body mass index, glucose and resistin. The explored algorithms include Logistic Regression, Naive Bayes, Support Vector Machine and Gradient Boosting Classification, all of which are used for the classification of new patients based on a dataset that includes information from previous breast cancer incidents. The results depict that the optimal algorithm based on the current methodology and implementation is the Gradient Boosting Classification which exhibits the highest prediction scores. In order to ensure wide accessibility, a mobile application is developed. The user can easily provide the required information for the prediction to the application and obtain the results rapidly.

In screening X-ray mammography two different views are captured and analysed of both breasts. In the four X-ray images some special signs are looked for. First the individual images are analysed independently, but good result can be achieved only if joint analysis of the images is also done. This paper proposes a simple procedure for the joint analysis of the breast's two views. The procedure is based upon the experiences of radiologists: masses and calcifications should emerge on both views, so if no matching is found, the given object is a false positive hit. First a reference system is evolved to find corresponding regions on the two images. Calcification clusters obtained in individual images are matched in “2.5 D” provided by the reference system. Masses detected in individual images are further examined with texture segmentation. The proposed approach can significantly reduce the number of false positive hits both in calcification and in mass detection.

<i>Objective:</i> This paper assesses the diagnostic performance of deep learning methods for tumour detection in breast microwave sensing (BMS). <i>Methods:</i> A convolutional neural network (CNN) was used to predict the presence of a cancerous lesion in data from experimental scans of MRI-derived phantoms. An experimental dataset containing data from 1257 scans was used. The CNN was compared to a similarly sized dense neural network (DNN) and logistic regression classifier. <i>Results:</i> The CNN was able to exploit the sinogram data structure to achieve diagnostic performance significantly better than random classification, while neither the DNN nor logistic regression classifiers could generalize to unseen test data. The area under the curve of the receiver operating characteristic curve of the CNN classifier was estimated to be between (78 <inline-formula><tex-math notation="LaTeX">$\pm$</tex-math></inline-formula> 3)&#x0025; and (90 <inline-formula><tex-math notation="LaTeX">$\pm$</tex-math></inline-formula> 3)&#x0025;, where the upper estimate was obtained when the testing set was constrained to consist of phantoms with breast volumes that are within the volume bounds of the training set and when the tumour was located at the same vertical position as the system antennas. <i>Conclusion:</i> The results obtained in this investigation demonstrate the potential of combining deep learning and BMS systems for breast cancer detection. <i>Impact:</i> This paper provides an estimate of the diagnostic performance of an air-based BMS system using deep learning methods for automatic tumour detection on a sizable experimental dataset. The performance was found to be comparable to that of AI-assisted mammography and marks a first step toward larger-scale investigations in BMS.

Recently, a new speckle noise reduction and contrast enhancement technique has been introduced that is motivated by the research in compressive sampling or sensing. Compressive sampling is based on the principle that a sparse signal such as ultrasound can be fully recovered when sampled below the Nyquist rate. This allows for a new noise reduction technique that preserves the high frequency and fine details while reducing the effects of speckle noise. This method improves the overall perceptual quality of the image for visualization and diagnosis by the radiologist. This paper examines how the improvement in SNR makes the method suitable as a preprocessor to improve a computer aided detection (CAD) system for breast cancer detection. Classical performance metrics such as false positive rates, false negative rates and receiver operator curves will be used to show the benefits of this approach. Initial experiments look promising for microcalcification detection, where the new method yields a false negative rate of 20 percent at a false positive rate of 0.5 percent while the traditional speckle reduction techniques yield a false negative rate of 60 percent at a false positive rate of 0.5 percent.

A three-dimensional microwave imaging algorithm using least-electrical-path (LEP) method is presented. The algorithm is designed for the ultra-wideband imaging system for breast cancer detection and localization. By considering the signal refraction on the borders of the imaged breast, this developed algorithm provides an accurate estimation for the pulse delay from a hypothesis point within the imaged region. It, thus, provides an accurate estimation of the tumor's position. The proposed method is validated using full-wave simulations.

Antenna system for breast cancer detection

The ability to discern malignant from benign tissue in excised human breast specimens in Breast Conservation Surgery (BCS) was evaluated using single frequency terahertz radiation. Terahertz (THz) images of the specimens in reflection mode were obtained by employing a gas laser source and mechanical scanning. The images were correlated with optical histological micrographs of the same specimens, and a mean discrimination of 73% was found for five out of six samples using Receiver Operating Characteristic (ROC) analysis. The system design and characterization is discussed in detail. The initial results are encouraging but further development of the technology and clinical evaluation is needed to evaluate its feasibility in the clinical environment.

In this paper, a hardware advancement for a microwave multi-static system for breast cancer detection is presented. In particular, we propose a metasurface-enhanced antenna capable of enhancing relevant “weak” signals backscattered by a tumour-mimicking target inserted in a breast phantom. Furthermore, we simulate a radar-based approach in CST Microwave Studio and use our previously developed Huygens principle-based algorithm to produce images of the target. Our results indicate that the proposed metasurface-enhanced antenna can improve target localization and decrease image artefacts.

Ultrawideband radar imaging is regarded as one of the most promising alternatives for breast cancer detection. A range of algorithms reported in literature shows satisfactory tumor detection capabilities. However, most of algorithms suffer significant deterioration or even fail when the early-stage artifact, including incident signals and skin-fat interface reflections, cannot be perfectly removed from received signals. Furthermore, fibro-glandular tissue poses another challenge for tumor detection, due to the small dielectric contrast between glandular and cancerous tissues. This paper introduces a novel Robust and Artifact Resistant (RAR) algorithm, in which a neighborhood pairwise correlation-based weighting is designed to overcome the adverse effects from both artifact and glandular tissues. In RAR, backscattered signals are time-shifted, summed, and weighted by the maximum combination of the neighboring pairwise correlation coefficients between shifted signals, forming the intensity of each point within an imaging area. The effectiveness was investigated using 3-D anatomically and dielectrically accurate finite-difference-time-domain numerical breast models. The use of neighborhood pairwise correlation provided robustness against artifact and enabled the detection of multiple scatterers. RAR is compared with four well-known algorithms: delay-and-sum, delay-multiply-and-sum, modified-weighted-delay-and-sum, and filtered-delay-and-sum. It has shown that RAR exhibits improved identification capability, robust artifact resistance, and high detectability over its counterparts in most scenarios considered, while maintaining computational efficiency. Simulated tumors in both homogeneous and heterogonous, from mildly to moderately dense breast phantoms, combining an entropy-based artifact removal algorithm, were successfully identified and localized. These results show the strong potential of RAR for breast cancer screening.