label image

Detection of vertebral fractures in CT using 3D Convolutional Neural Networks


Since our task is detection and not segmentation, correctly predicting only a sufficient amount of voxels around the vertebra centroid is needed to detect normal or fractured vertebrae in an image. We leverage this observation to construct 3D label images for our training database in a semi-automated fashion. First, radiologist S.R. created a text file with annotations for every vertebra present in the field of view as described in section 2. Next, J.N. enriched these labels with 3D centroid coordinates by manually localizing every vertebra centroid in the image using MeVisLab [8]. This step required an average of less than two minutes per image in our dataset. Finally, we extended the method described by Glocker et al. [6] to automatically generate 3D label images from these sparse annotations.

Multi-resolution neural networks for tracking seismic horizons from few training images Machine Learning

Detecting a specific horizon in seismic images is a valuable tool for geological interpretation. Because hand-picking the locations of the horizon is a time-consuming process, automated computational methods were developed starting three decades ago. Older techniques for such picking include interpolation of control points however, in recent years neural networks have been used for this task. Until now, most networks trained on small patches from larger images. This limits the networks ability to learn from large-scale geologic structures. Moreover, currently available networks and training strategies require label patches that have full and continuous annotations, which are also time-consuming to generate. We propose a projected loss-function for training convolutional networks with a multi-resolution structure, including variants of the U-net. Our networks learn from a small number of large seismic images without creating patches. The projected loss-function enables training on labels with just a few annotated pixels and has no issue with the other unknown label pixels. Training uses all data without reserving some for validation. Only the labels are split into training/testing. Contrary to other work on horizon tracking, we train the network to perform non-linear regression, and not classification. As such, we propose labels as the convolution of a Gaussian kernel and the known horizon locations that indicate uncertainty in the labels. The network output is the probability of the horizon location. We demonstrate the proposed computational ingredients on two different datasets, for horizon extrapolation and interpolation. We show that the predictions of our methodology are accurate even in areas far from known horizon locations because our learning strategy exploits all data in large seismic images.

Learning Discriminative Multilevel Structured Dictionaries for Supervised Image Classification Machine Learning

Sparse representations using overcomplete dictionaries have proved to be a powerful tool in many signal processing applications such as denoising, super-resolution, inpainting, compression or classification. The sparsity of the representation very much depends on how well the dictionary is adapted to the data at hand. In this paper, we propose a method for learning structured multilevel dictionaries with discriminative constraints to make them well suited for the supervised pixelwise classification of images. A multilevel tree-structured discriminative dictionary is learnt for each class, with a learning objective concerning the reconstruction errors of the image patches around the pixels over each class-representative dictionary. After the initial assignment of the class labels to image pixels based on their sparse representations over the learnt dictionaries, the final classification is achieved by smoothing the label image with a graph cut method and an erosion method. Applied to a common set of texture images, our supervised classification method shows competitive results with the state of the art.