Classification of breast tumor models with a prototype microwave imaging system.

PURPOSE: The assessment of the size and shape of breast tumors is of utter importance to the correct diagnosis and staging of breast cancer. In this paper, we classify breast tumor models of varying sizes and shapes using signals collected with a monostatic ultra-wideband radar microwave imaging prototype system with machine learning algorithms specifically tailored to the collected data. METHODS: A database comprising 13 benign and 13 malignant tumor models with sizes between 13 and 40 mm was created using dielectrically representative tissue mimicking materials. These tumor models were placed inside two breast phantoms: a homogeneous breast phantom and a breast phantom with clusters of fibroglandular mimicking tissue, accounting for breast heterogeneity. The breast phantoms with tumors were imaged with a monostatic microwave imaging prototype system, over a 1-6 GHz frequency range. The classification of benign and malignant tumors embedded in the two breast phantoms was completed, and tumor classification was evaluated with Principal Component Analysis as a feature extraction method, and tuned Naïve Bayes (NB), decision trees (DT), and k-nearest neighbours (kNN) as classifiers. We further study which antenna positions are better placed to classify tumors, discuss the feature extraction method and optimize classification algorithms, by tuning their hyperparameters, to improve sensitivity, specificity and the receiver operating characteristic curve, while ensuring maximum generalization and avoiding overfitting and data contamination. We also added a realistic synthetic skin response to the collected signals and examined its global effect on classification of benign vs malignant tumors. RESULTS: In terms of global classification performance, kNN outperformed DT and NB machine learning classifiers, achieving a classification accuracy of 96.2% when classifying between benign and malignant tumor phantoms in a homogeneous breast phantom (both when the skin artifact is and is not considered). CONCLUSIONS: We experimentally classified tumor models as benign or malignant with a microwave imaging system, and we showed a methodology that can potentially assess the shape of breast tumors, which will give further insight into the correct diagnosis and staging of breast cancer.

Experimental feasibility of multistatic holography for breast microwave radar image reconstruction.

PURPOSE: The goal of this study was to assess the experimental feasibility of circular multistatic holography, a novel breast microwave radar reconstruction approach, using experimental datasets recorded using a preclinical experimental setup. The performance of this approach was quantitatively evaluated by calculating the signal to clutter ratio (SCR), contrast to clutter ratio (CCR), tumor to fibroglandular response ratio (TFRR), spatial accuracy, and reconstruction time. METHODS: Five datasets were recorded using synthetic phantoms with the dielectric properties of breast tissue in the 1-6 GHz range using a custom radar system developed by the authors. The datasets contained synthetic structures that mimic the dielectric properties of fibroglandular breast tissues. Four of these datasets the authors covered an 8 mm inclusion that emulated a tumor. A custom microwave radar system developed at the University of Manitoba was used to record the radar responses from the phantoms. The datasets were reconstructed using the proposed multistatic approach as well as with a monostatic holography approach that has been previously shown to yield the images with the highest contrast and focal quality. RESULTS: For all reconstructions, the location of the synthetic tumors in the experimental setup was consistent with the position in the both the monostatic and multistatic reconstructed images. The average spatial error was less than 4 mm, which is half the spatial resolution of the data acquisition system. The average SCR, CCR, and TFRR of the images reconstructed with the multistatic approach were 15.0, 9.4, and 10.0 dB, respectively. In comparison, monostatic images obtained using the datasets from the same experimental setups yielded average SCR, CCR, and TFRR values of 12.8, 4.9, and 5.9 dB. No artifacts, defined as responses generated by the reconstruction method of at least half the energy of the tumor signatures, were noted in the multistatic reconstructions. The average execution time of the images formed using the proposed approach was 4 s, which is one order of magnitude faster than the current state-of-the-art time-domain multistatic breast microwave radar reconstruction algorithms. CONCLUSIONS: The images generated by the proposed method show that multistatic holography is capable of forming spatially accurate images in real-time with signal to clutter levels and contrast values higher than other published monostatic and multistatic cylindrical radar reconstruction approaches. In comparison to the monostatic holographic approach, the images generated by the proposed multistatic approach had SCR values that were at least 50% higher. The multistatic images had CCR and TFRR values at least 200% greater than those formed using a monostatic approach.

Real time breast microwave radar image reconstruction using circular holography: a study of experimental feasibility.

PURPOSE: The purpose of this paper is to assess the experimental feasibility of a novel breast microwave radar reconstruction approach, circular holography, using realistic experimental datasets recorded using a preclinical experimental setup. The performance of this approach was quantitatively evaluated by calculating the signal to noise ratio, contrast to noise ratio, spatial accuracy, and reconstruction time. METHODS: Six datasets were recorded, three corresponding to fatty cases and three containing synthetic dense tissue structures. Five of these datasets contained an 8 mm inclusion that emulated a malignant lesion. The data were acquired from synthetic phantoms that mimic the dielectric properties of breast tissues in the 1-6 GHz range using a custom experimental breast microwave radar system. The spatial accuracy and signal to noise ratio of the reconstructed was calculated for all the reconstructed images. The contrast to noise ratio of the reconstructed images corresponding to the datasets containing fibroglandular tissue regions was determined. This was done to evaluate the ability of the circular holographic method to provide images in which the responses from tumors can be distinguished from adjacent dense tissue structures. The execution time required to form the images was also measured to evaluate the data throughput of the holographic approach. RESULTS: For all the reconstructed datasets, the location of the synthetic tumors in the experimental setup was consistent with its position in the reconstructed image. The average spatial error was 2.2 mm, which is less than half the spatial resolution of the data acquisition system. The average signal to noise ratio of the reconstructed images containing an artificial malignant lesion was 8.5 dB, while the average contrast to noise ratio was 6.7 dB. The reconstructed images presented no artifacts. The average execution time of the images formed using the proposed approach was 5 ms, which is six orders of magnitude faster than current state of the art breast microwave radar (BMR) reconstruction algorithms. CONCLUSIONS: The results show that circular holography is capable of forming accurate images with signal to noise levels higher than 8 dB in quasi real time. Compared to BMR reconstruction algorithms tested on datasets containing dense tissue structures, the holographic approach generated images of similar spatial accuracy with higher signal to noise ratios and an acceleration factor of one order of magnitude.

Electrical impedance tomography reconstruction using a monotonicity approach based on a priori knowledge.

Electrical Impedance Tomography(EIT) has been proposed as an alternative modality for breast imaging. Current EIT reconstruction algorithms are based in optimization procedures that aim to minimize the difference between the recorded data and a set of candidate scenarios. However, these methods produce images with diffused edges, as sharp structures are penalized by current regularization techniques. In this paper, a novel edge preserving EIT reconstruction method is proposed. This technique uses a priori information obtained from Breast Microwave Radar images to estimate the location of the dense breast regions. Then, the monotonicity of the impedance matrix of the collected data is used to reconstruct a profile of the tissue distribution in the breast region. The proposed method yielded promising results when applied to numeric phantoms generated from Magnetic Resonance Imaging datasets.

Real time MRI prostate segmentation based on wavelet multiscale products flow tracking.

Currently, prostate cancer is the third leading cause of cancer-related deaths among men in North America. As with many others types of cancer, early detection and treatment greatly increases the patient's chance of survival. Combined Magnetic Resonance Imaging and Spectroscopic Imaging (MRI/MRSI) techniques have became a reliable tool for early stage prostate cancer detection. Nevertheless, their performance is strongly affected by the determination of the region of interest (ROI) prior to data acquisition process. The process of executing prostate MRI/MRSI techniques can be significantly enhanced by segmenting the whole prostate. A novel method for segmentation of the prostate in MRI datasets is presented. This method exploits the different behavior presented by signal singularities and noise in the wavelet domain in order to accurately detect the borders around the prostate. The prostate contour is then traced by using a set of spatially variant rules that are based on prior knowledge about the general shape of the prostate. The proposed method yielded promising results when applied to clinical datasets.

A comparison of interpolation methods for breast microwave radar imaging.

In recent years, Breast Microwave Imaging (BMI) has shown its potential as a promising breast cancer detection technique. This imaging technology is based on the electrical characteristic differences that exist between normal and malignant breast tissues at the microwave frequency range. A promising image formation technique for BMI radar based approaches is wavefront reconstruction. In this approach, the image quality and execution time of this image formation technique is strongly affected by the interpolation method that is used. In this paper, a performance study between three popular interpolation techniques, nearest neighbor, linear and cubic splines, for breast microwave radar imaging is presented. The performance of the evaluated techniques was assessed using numeric phantoms obtained from Magnetic Resonance Imaging (MRI) data sets. The results of this study indicate that linear interpolation techniques are the most suitable choices based on their computational cost, and the focal quality and signal to noise of their resulting images.

An improved wavefront reconstruction method for breast microwave imaging.

Breast cancer is the leading cause of cancer related deaths in women between the ages of 15 and 54, and the second cause of cancer death in women the 55 to 74 age range. In recent years, Breast Microwave Imaging (BMI) has shown its potential as a promising breast cancer detection technique. This imaging technology is based on the electrical characteristic differences that exist between normal and malignant breast tissues at the microwave frequency range. A novel reconstruction approach for the formation of 2D BMI models is proposed in this paper. This technique uses the phase differences introduced during the collection of target responses in order to determine the correct spatial location of the different structures that constitute the final image. The proposed method yielded promising results when applied to simulated data sets obtained from Magnetic Resonance Images (MRI).

A wavefront reconstruction method for 3-D cylindrical subsurface radar imaging.

In recent years, the use of radar technology has been proposed in a wide range of subsurface imaging applications. Traditionally, linear scan trajectories are used to acquire data in most subsurface radar applications. However, novel applications, such as breast microwave imaging and wood inspection, require the use of nonlinear scan trajectories in order to adjust to the geometry of the scanned area. This paper proposes a novel reconstruction algorithm for subsurface radar data acquired along cylindrical scan trajectories. The spectrum of the collected data is processed in order to locate the spatial origin of the target reflections and remove the spreading of the target reflections which results from the different signal travel times along the scan trajectory. The proposed algorithm was successfully tested using experimental data collected from phantoms that mimic high contrast subsurface radar scenarios, yielding promising results. Practical considerations such as spatial resolution and sampling constraints are discussed and illustrated as well.

Semi automatic MRI prostate segmentation based on wavelet multiscale products.

Currently, prostate cancer is the third leading cause of cancer-related deaths among men in North America. As with many others types of cancer, early detection and treatment greatly increases the patient's chance of survival. MRI prostate segmentation allows clinical personnel to design an accurate treatment plan. A novel method for MRI prostate imagery segmentation is proposed in this paper. This method exploits the different behavior presented by signal singularities and noise in the wavelet domain in order to accurately detect the borders around the prostate. The prostate contour is then traced by using a set of spatially variant rules that are based on prior knowledge about the general shape of the prostate. The proposed method yielded promising results when applied to real data.

Heart sound cancellation based on multiscale products and linear prediction.

This paper presents a novel method for Heart Sound (HS) cancellation from Lung Sound (LS) records. The method uses the multiscale product of the wavelet coefficients of the original signal to detect HS-included segments. Once the HS segments are identified, the method removes them from the wavelet coefficients at every level and estimates the created gaps by using a set of linear prediction filters. It is shown that if the segment to be predicted is stationary, a final record with no audible artifacts such as clicks can be reconstructed using this approach. The results were promising for HS removal from LS records and showed no hampering of the main components of the LS. The results were confirmed both qualitatively by listening to the reconstructed signal and quantitatively by spectral analysis.

3D breast microwave imaging based on wavefront reconstruction.

Currently, breast cancer is the leading cause of cancer death in women between the ages of 15 and 54, and the second cause of cancer death in women 55 to 74. In recent years, Breast Microwave Imagery (BMI) has shown its potential as a promising breast cancer detection technique. This imaging technology is based on the electrical characteristic differences that exist between normal and malignant breast tissues at the microwave frequency range. A novel reconstruction approach for the formation of 3D BMI models is proposed in this paper. This technique uses the phase differences introduced during the collection of target responses in order to determine the correct spatial location of the different scatterers that constitute the final image. The proposed method yielded promising results when applied to simulated data.

Hybrid microwave tomography technique for breast cancer imaging.

This paper addresses a two-dimensional inverse scattering method with a combination of tomography and radar methods for breast cancer detection. In order to rapidly construct high resolution images displaying the location, size, permittivity and conductivity of malignant tumors inside the body, the collected reflection from the scattered fields present in the scan area is segmented and their associated dielectric property maps are calculated. The dielectric profiles are obtained by using a technique that combines frequency domain finite difference time domain (FD)2TD analysis with genetic algorithm (GA) optimization. The applications of the proposed method can vary from medical imaging to nondestructive testing of materials and structures. The proposed technique yielded promising results when applied to simulated data.