Identification of Breast Malignancy by Marker-Controlled Watershed Transformation and Hybrid Feature Set for Healthcare

Abstract

Breast cancer is a highly prevalent disease in females that may lead to mortality in severe cases. The mortality can be subsided if breast cancer is diagnosed at an early stage. The focus of this study is to detect breast malignancy through computer-aided diagnosis (CADx). In the first phase of this work, Hilbert transform is employed to reconstruct B-mode images from the raw data followed by the marker-controlled watershed transformation to segment the lesion. The methods based only on texture analysis are quite sensitive to speckle noise and other artifacts. Therefore, a hybrid feature set is developed after the extraction of shape-based and texture features from the breast lesion. Decision tree, k-nearest neighbor (KNN), and ensemble decision tree model via random under-sampling with Boost (RUSBoost) are utilized to segregate the cancerous lesions from the benign ones. The proposed technique is tested on OASBUD (Open Access Series of Breast Ultrasonic Data) and breast ultrasound (BUS) images collected at Baheya Hospital Egypt (BHE). The OASBUD dataset contains raw ultrasound data obtained from 100 patients containing 52 malignant and 48 benign lesions. The dataset collected at BHE contains 210 malignant and 437 benign images. The proposed system achieved promising accuracy of 97% with confidence interval (CI) of 91.48% to 99.38% for OASBUD and 96.6% accuracy with CI of 94.90% to 97.86% for the BHE dataset using ensemble method.

Breast ultrasound tumour classification: A Machine Learning—Radiomics based approach

“…In one of the more recent works by Moon et al (2020), a convolutional neural network (CNN) ensemble based approach is presented that works on the combination of the original ultrasound image and its ground truth segmentation mask. It is important to note here that both Sadad et al (2020) and (Moon et al (2020) validated their proposed approaches on the BUSI dataset, that is used in this current work also, and the current work is able to outperform these state‐of‐the‐art approaches, in terms of almost all evaluation metrics under consideration. Moreover, almost none of the related work explicitly handles the problem of class imbalance that is typically present in this area, thereby resulting in less sensitive models.…”

“…More specifically, the authors in Moon et al (2020) recently proposed a CNN ensemble based approach that automatically extracts ROI features, in order to perform the benign versus malignant classification on the BUSI dataset, which was able to achieve the Accuracy, AUC and F 1‐score of 0.946, 0.97 and 0.911. In another recent work (Sadad et al, 2020), an experimental study was presented on the BUSI dataset, where the authors proposed the use of shape and texture based features from the ROIs. These quantitative ROI features were then utilized by an ensemble approach, which internally performed random undersampling based boosting (RUSBoost), achieving the accuracy, sensitivity and specificity of 0.966, 0.943 and 0.977.…”

“…Moreover, the researchers only explored fisher linear discriminant analysis based classification in this work. In a more recent ML‐radiomics based work (Sadad et al, 2020), the use of a hybrid feature set that consists of shape as well as texture based features, along with an ensemble classifier is proposed. Impressive classification performance was obtained with the help of such hybrid radiomic features, however the researchers did not explore the idea of selecting the most important predictive features for the classification process, which could have further improved the efficiency of the proposed approach.…”

Dilated Semantic Segmentation for Breast Ultrasonic Lesion Detection Using Parallel Feature Fusion

“…The Hilbert transform was used to reproduce B-mode features from raw images, followed by the marker-controlled watershed transformation to segment the breast cancer lesion [ 23 ]. The techniques used, which were solely focused on texture analysis, were very susceptible to speckle noise and other objects.…”

Internet of medical things embedding deep learning with data augmentation for mammogram density classification

“…BD classification is beneficial in BC risk assessment because high BD increases BC risk chances (Mughal, Muhammad, Sharif, Rehman, & Saba, 2018). Mammography screening is a valuable imaging modality to detect BD (Sadad, Munir, saba, & Hussain, 2018). Generally, a doctor advises mammography to analyze BD for additional BC risk because it is a low‐cost and helpful procedure to judge breast tumors' symptoms at the initial stage (Saba, Haseeb, Ahmed, & Rehman, 2020).…”