An Automatic Localization Algorithm for Ultrasound Breast Tumors Based on Human Visual Mechanism.

Yuting Xie,Ke Chen,Jiangli Lin

doi:10.3390/s17051101

Abstract

Human visual mechanisms (HVMs) can quickly localize the most salient object in natural images, but it is ineffective at localizing tumors in ultrasound breast images. In this paper, we research the characteristics of tumors, develop a classic HVM and propose a novel auto-localization method. Comparing to surrounding areas, tumors have higher global and local contrast. In this method, intensity, blackness ratio and superpixel contrast features are combined to compute a saliency map, in which a Winner Take All algorithm is used to localize the most salient region, which is represented by a circle. The results show that the proposed method can successfully avoid the interference caused by background areas of low echo and high intensity. The method has been tested on 400 ultrasound breast images, among which 376 images succeed in localization. This means this method has a high accuracy of 94.00%, indicating its good performance in real-life applications.

Highlights

Breast cancer is one of the most common malignant tumors women face
We found above that the saliency of tumor area is greatly affected by high-intensity background areas and low-intensity background areas, respectively, in the intensity saliency map and black ratio areas and low-intensity background areas, respectively, in the intensity saliency map and black ratio saliency map
An automatic localization algorithm for US breast images is tested

Summary

Introduction

Breast cancer is one of the most common malignant tumors women face. It has the highest incidence and mortality of all diseases affecting women and even holds a rising trend. The early detection of suspicious lesions is very important for effective treatment of breast cancer. Ultrasound, mammography and magnetic resonance imaging (MRI) are general methods for the clinical detection of lesions, among which ultrasound (US) has been widely used due to its non-invasion, non-ionization radiation and non-injury [1,2]. The shortcomings of US breast images, such as low contrast, serious speckles, and low spatial resolution, make it difficult for doctors to read and analyze these suspicious lesions. With an increasing number of patients, doctors feel heavily burdened, resulting in a higher rate of misdiagnosis

Results

Discussion

Conclusion