A convolutional neural network for an automatic biopsy needle detection and segmentation on ultrasound images
The dataset contains a convolutional neural network (CNN) which was trained to segment a biopsy needle in 2D ultrasound images. Together with the network, source files of Matlab scripts and a few input images are provided. The files are enough to run experiments. See the `README' file for further instructions. The CNN structure was described in the article "An automatic biopsy needle detection and segmentation on ultrasound images using a convolutional neural network", currently under review in Ultrasonic Imaging. It is a fully automatic method that uses a CNN to detect a biopsy needle in 2D ultrasound images. In the paper, the CNN architecture is described, which was trained using the adaptive moment estimation optimizer. The needle is eventually located in the image by applying the Radon transform.