Assessing Patient-Specific Microwave Breast Imaging in Clinical Case Studies

Microwave breast imaging has seen increasing use in clinical investigations in the past decade with over eight systems having being trialled with patients. The majority of systems use radar-based algorithms to reconstruct the image shown to the clinician which requires an estimate of the dielectric properties of the breast to synthetically focus signals to reconstruct the image. Both simulated and experimental studies have shown that, even in simplified scenarios, misestimation of the dielectric properties can impair both the image quality and tumour detection. Many methods have been proposed to address the issue of the estimation of dielectric properties, but few have been tested with patient images. In this work, a leading approach for dielectric properties estimation based on the computation of many candidate images for microwave breast imaging is analysed with patient images for the first time. Using five clinical case studies of both healthy breasts and breasts with abnormalities, the advantages and disadvantages of computational patient-specific microwave breast image reconstruction are highlighted.

An Effective Mammogram Classification Using Hot Based Tree And Hot Based Cnn

Breast cancer is the subsequent leading cause of cancer-related deceases among women. Initial exposure stimulates enhanced visualization and saves survives. These days, the exact grouping classification of breast cancer images is a difficult errand. There are much research works delivering various strategies and algorithms for this specific errand of medical image processing. To build up an exact characterization, this paper presents a viable classification of mammogram images utilizing HOT based classification tree and HOT based convolutional neural network (CNN). The input breast image is at first taken from the database and pre-processed by RGB to grayscale conversion and normalization methodology. In this way Histogram of Oriented Texture (HOT) Descriptor is extorted from the pre-processed images. At long last images are classified as typical or irregular utilizing HOT based classification tree and HOT based CNN. The exploratory results show that the introduced method outperforms the existing strategies concerning various performance assessments like accuracy, sensitivity, specificity, mean absolute error, AUC score, kappa statistics, and Root mean square error

Breast Tumor Detection in MR Images Based on Density

Breast cancer is dangerous in women. It is generally found after the symptoms appear. Detecting the breast cancer at an early stage and understanding the treatment are the most important strategies to prevent death from cancer. Generally, for detection of breast cancer, breast Magnetic Resonance Image (MRI) takes place. It is one of the best approaches to detect tumor in women. In this research paper, a combination of selection methods for seed region growing image segmentation is suggested to detect breast tumor. The suggested method has been divided into following parts: First, the pre-processing of breast image is performed. Second, the automatic threshold for binarization process is calculated. Third, the number of seed points and its position in the breast image are determined automatically using density of pixels value. Fourth, a method for calculation of threshold value is proposed for the purpose of region creation in seed region growing. For the evaluation purpose, the proposed method was applied and tested on the RIDER MRI breast dataset from National Biomedical Imaging Archive (NBIA). After the test was performed, it was observed that proposed algorithm gives 90% accuracy, 88% True Negative Fraction, 91% True Positive Fraction, 10% Misclassification Rate, 94% Precision and 86% Relative Overlap which is better than other existing methods. It not only gives better evaluation measure but also provides segmentation method for multiple tumor detection.

ASSESSMENT OF ASYMMETRY IN BILATERAL STATIC FRONTAL BREAST THERMOGRAMS USING DIFFERENCE IMAGE AND RADIOMIC FEATURES

Asymmetry analysis of bilateral thermogram images is an important preliminary approach for breast cancer detection. The purpose of this work is to develop an automated algorithm to detect and classify symmetric and asymmetric bilateral static frontal breast thermograms (N=63). The images are pre-processed using anisotropic diffusion filter for removal of noise. Further, segmentation of complete breast region is carried out using level set segmentation without re-initialization. The bifurcation point is computed from the intersection point of interior inframammary curves attained using polynomial curve fitting on the boundary pixels. The obtained breast region is sliced vertically along this bifurcation point to obtain right and left breast sections. Image subtraction is performed between right breast image and flipped left breast image to obtain the difference image. The obtained difference image is sharpened and 144 texture features such as first-order statistical, co- occurrence, run length and laws energy features are extracted and Absolute Difference (AD) between symmetric and asymmetric subjects for each feature is computed. The features for which the value of AD is greater than 0.1 is considered as substantial features. Twenty four substantial features are obtained and are given as an input to Least Square Support Vector Machine (LSSVM) to automate the classification. The results shows that the maximum segmentation overlap measure obtained is 98.3%. The classification accuracy obtained using LSSVM with Radial Basis Function (RBF) is 95.65% and sensitivity, specificity and Area Under the Curve (AUC) are 100%, 90.9% and 0.9545 respectively. Thus the proposed methodology appears to be effective in detecting asymmetric heat patterns and hence can be deployed in thermal screening systems.

Transfer Learning in Breast Cancer Diagnoses via Ultrasound Imaging

Transfer learning is a machine learning approach that reuses a learning method developed for a task as the starting point for a model on a target task. The goal of transfer learning is to improve performance of target learners by transferring the knowledge contained in other (but related) source domains. As a result, the need for large numbers of target-domain data is lowered for constructing target learners. Due to this immense property, transfer learning techniques are frequently used in ultrasound breast cancer image analyses. In this review, we focus on transfer learning methods applied on ultrasound breast image classification and detection from the perspective of transfer learning approaches, pre-processing, pre-training models, and convolutional neural network (CNN) models. Finally, comparison of different works is carried out, and challenges—as well as outlooks—are discussed.