Breast Density Classification Using Multiresolution Local Quinary Patterns in Mammograms

Abstract

Abstract. This paper presents a method for breast density classification using local quinary patterns (LQP) in mammograms. LQP operators are used to capture the texture characteristics of the fibroglandular disk region (F GDroi) instead of the whole breast region as the majority of current studies have done. To maximise the local information, a multiresolution approach is employed followed by dimensionality reduction by selecting dominant patterns only. Subsequently, the Support Vector Machine classifier is used to perform the classification and a stratified ten-fold cross-validation scheme is employed to evaluate the performance of the method. The proposed method produced competitive results up to 85.6% accuracy which is comparable with the state-of-the-art in the literature. Our contributions are two fold: firstly, we show the role of the fibroglandular disk area in representing the whole breast region as an important region for more accurate density classification and secondly we show that the LQP operators can extract discriminative features comparable with the other popular techniques such as local binary patterns, textons and local ternary patterns (LTP).

Breast Density Classification Using Local Quinary Patterns with Various Neighbourhood Topologies

“…In this paper, we present a fully automatic method for breast density classification using local pattern information extracted using Local Quinary Patterns (LQP) operators from the fibro-glandular region instead of from the whole breast area as all of the studies in the literature have done. In particular, this paper extends our previous work in [14] by investigating the various neighbourhood topologies and different dominant local patterns. The contributions of our study are:…”

“…(11,16) do not provide significant improvement on the classification accuracy (Acc = 79.65%) due to insufficient textural information in LQP small (1,8) and over-representation of the actual textural information from LQP large (11,16). We further tested LQP small el (1, 3, 10) + LQP medium el (7,14,14) + LQP large el (15,19,18) and achieved an accuracy of 78.62%. The results in Figure 10 also indicate that the parameter n plays an important role in the performance of the method.…”

“…2. We have extended the evaluation results for the circle topology, which was originally presented in [14] covering different aspects such as parameter selection, orientations and quantitative comparison. 3.…”

Implementation of an ontological reasoning to support the guideline-based management of primary breast cancer patients in the DESIREE project

“…The objective is to develop a web-based software ecosystem dedicated to the personalized, collaborative, and multidisciplinary management of primary breast cancer, from diagnosis, to therapy, and follow-up. The DESIREE platform offers some image-based diagnostic decision support modalities involving mammogram-based breast density classification [26], fully automated breast boundary and pectoral muscle segmentation [27], and breast mass classification using ensemble convolutional neural networks [28]. Research works on predictive modeling have also been conducted, e.g., to predict the esthetic outcome of Breast Conservative Therapy considering mechanical forces due to gravity, breast density and tissue distribution, and the inflammation induced by radiotherapy and the wound healing [29].…”