Computer Vision Techniques for Breast Tumor Ultrasound Analysis

Abstract

In this paper, we present a new approach to the segmentation and analysis of solid breast nodules in ultrasonography. We have applied computer vision techniques to segment the nodules and analyze a series of diagnostic criteria which can help discriminate malignant and benignant tumors. The segmentation is carried out in a semiautomatic way, whereas the analysis of the diagnostic criteria involves several computational methods. The techniques which we propose have provided quite satisfactory results and show the usefulness of image processing in the diagnosis through medical imaging.

Rapid image stitching and computer‐aided detection for multipass automated breast ultrasound

“…The CAD system's use as a reference opinion for improving accuracy and reliability of diagnosis has attracted much interest among researchers over the past decade. Previous studies focused on two key areas: The detection of the tumor region 22,[38][39][40][41] and the classification of breast masses. [42][43][44][45][46] For boundary extraction of breast masses, Cary et al 38 used leak properties to grow a manually drawn seed region close to the tumor boundary.…”

Trust at Hand: A Multi-touch Tabletop Interface for Collaborative Training of Machine Learning Models in the Medical Domain

“…Due to the visual nature of the system we chose to tailor the system around image classification. This in itself is an active area of research in the domain of medicine, with work exploring a variety of applications such as tumor detection [113] and x-ray processing [114]. Based on this work we derived a set of criteria for our sample.…”

Computer-Aided Lesion Diagnosis in Automated 3-D Breast Ultrasound Using Coronal Spiculation

“…As computation of such features generally requires an accurate delineation of the abnormality, segmentation is an important step in lesion classification algorithms. Various segmentation methods have been proposed to segment lesions in different modalities [14], [28]- [30]. In ultrasound, inhomogeneity of intensities inside lesions, background structure, partially undefined boundaries, and different posterior acoustic behavior of lesions, make lesion segmentation a difficult task.…”