Classification of malignant tumours in breast ultrasound using unsupervised machine learning approaches

Abstract

Traditional computer-aided diagnosis (CAD) processes include feature extraction, selection, and classification. Effective feature extraction in CAD is important in improving the classification’s performance. We introduce a machine-learning method and have designed an analysis procedure of benign and malignant breast tumour classification in ultrasound (US) images without a need for a priori tumour region-selection processing, thereby decreasing clinical diagnosis efforts while maintaining high classification performance. Our dataset constituted 677 US images (benign: 312, malignant: 365). Regarding two-dimensional US images, the oriented gradient descriptors’ histogram pyramid was extracted and utilised to obtain feature vectors. The correlation-based feature selection method was used to evaluate and select significant feature sets for further classification. Sequential minimal optimisation—combining local weight learning—was utilised for classification and performance enhancement. The image dataset’s classification performance showed an 81.64% sensitivity and 87.76% specificity for malignant images (area under the curve = 0.847). The positive and negative predictive values were 84.1 and 85.8%, respectively. Here, a new workflow, utilising machine learning to recognise malignant US images was proposed. Comparison of physician diagnoses and the automatic classifications made using machine learning yielded similar outcomes. This indicates the potential applicability of machine learning in clinical diagnoses.

Incorporating the Breast Imaging Reporting and Data System Lexicon with a Fully Convolutional Network for Malignancy Detection on Breast Ultrasound

“…Previous related studies used image segmentation [ 3 , 4 ] or lesion texture [ 5 , 6 ] to generate a pattern or model for malignant classification. In addition, several studies incorporated established significant features of the whole image into a deep learning network for malignant or benign tumor classification [ 7 , 8 , 9 , 10 , 11 ]. While all these previous studies had a classification accuracy of over 85% and showed good preliminary performance, providing only the benign and/or malignant classification of an image is insufficient for clinical practice.…”

A novel bi-modal extended Huber loss function based refined mask RCNN approach for automatic multi instance detection and localization of breast cancer

“…The performance of EHL-RMRCNN is compared with other existing method which are Shen et al, 22 Chougrad et al, 18 Ragab et al, 19 Muduli et al, 32 Ragab et al, 30 Chiang et al, 16 Ciritsis et al, 25 Wang et al, 27 Shia et al, 33 Raza et al, 31 Chiao et al, 23 Min et al, 28 and He et al 14 Tables 6 and 7 exhibits the results of the comparative qualitative analysis. Table 6 shows the comparative study between existing CNN models and proposed method for mammogram and ultrasound breast images.…”

“…Muduli et al 32 2021 Deep CNN Mammograms and Breast US AC-90.68%, SE-92.72% and SP-88.21%. Shia et al 33 2021 Unsupervised ML Ultrasound SE-0.86 and SP-0.85.…”

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

“…Similarly, Ref. [25] proposed an effective method for extracting and selecting features. It was suggested that clinical efforts be reduced without segmentation; effective feature selection can improve classification.…”

“…Following that, by combining learned weights with it, the classification was carried out using minimal sequential optimization. The achieved classification sensitivity was 81.64%, with a specificity of 87.76% [25]. A deep learning-based study used privately collected ultrasonic images to feed shape and orientation scores to the quantitative morphological score.…”