Segmentation and recognition of breast ultrasound images based on an expanded U-Net

Abstract

This paper establishes a fully automatic real-time image segmentation and recognition system for breast ultrasound intervention robots. It adopts the basic architecture of a U-shaped convolutional network (U-Net), analyses the actual application scenarios of semantic segmentation of breast ultrasound images, and adds dropout layers to the U-Net architecture to reduce the redundancy in texture details and prevent overfitting. The main innovation of this paper is proposing an expanded training approach to obtain an expanded of U-Net. The output map of the expanded U-Net can retain texture details and edge features of breast tumours. Using the grey-level probability labels to train the U-Net is faster than using ordinary labels. The average Dice coefficient (standard deviation) and the average IOU coefficient (standard deviation) are 90.5% (±0.02) and 82.7% (±0.02), respectively, when using the expanded training approach. The Dice coefficient of the expanded U-Net is 7.6 larger than that of a general U-Net, and the IOU coefficient of the expanded U-Net is 11 larger than that of the general U-Net. The context of breast ultrasound images can be extracted, and texture details and edge features of tumours can be retained by the expanded U-Net. Using an expanded U-Net can quickly and automatically achieve precise segmentation and multi-class recognition of breast ultrasound images.

An improved semantic segmentation for breast lesion from dynamic contrast enhanced MRI images using deep learning

“…25 Guo et al have suggested an expanded UNet that included seven dropout layers in the original UNet to segment tumors from ultrasound images. 26 The Intersection over Union (IoU) coefficient of the expanded UNet was 11.0 units larger than the IoU coefficient of the original UNet.…”

Application of Machine Learning in Intelligent Medical Image Diagnosis and Construction of Intelligent Service Process

“…In order to find the pixels in the image that have the greatest impact on the deep learning results, the logical relationship between the input and the output is judged by studying the influence of the input layer changes on the output results. Using backpropagation [ 19 ] and combining gradients, network weights, or activations [ 42 , 43 ] track information, and the network output tracks its input or intermediate hidden layers. Reference [ 43 ] filtered gradients through an optimization process to further extract fine-grained regions for specific prediction evidence.…”

“…Using backpropagation [ 19 ] and combining gradients, network weights, or activations [ 42 , 43 ] track information, and the network output tracks its input or intermediate hidden layers. Reference [ 43 ] filtered gradients through an optimization process to further extract fine-grained regions for specific prediction evidence. The core of these methods is to find the most representative perturbations through detailed search or optimization.…”

Lumbar Spinal Stenosis Analysis with Deep Learning Based Decision Support Systems

“…In fact, the computer treats this color image as three 512x512 matrices for three different color channels. This is performed on 2D matrices by image processing methods such as convolution and pooling [14,15]. Blur, sharpen, edge detection, noise reduction, etc.…”