3D Deep Learning for Efficient and Robust Landmark Detection in Volumetric Data
Recently, deep learning has demonstrated great success in computer vision with the capability to learn powerful image features from a large training set. However, most of the published work has been confined to solving 2D problems, with a few limited exce
- PDF / 563,794 Bytes
- 8 Pages / 439.363 x 666.131 pts Page_size
- 46 Downloads / 215 Views
Abstract. Recently, deep learning has demonstrated great success in computer vision with the capability to learn powerful image features from a large training set. However, most of the published work has been confined to solving 2D problems, with a few limited exceptions that treated the 3D space as a composition of 2D orthogonal planes. The challenge of 3D deep learning is due to a much larger input vector, compared to 2D, which dramatically increases the computation time and the chance of over-fitting, especially when combined with limited training samples (hundreds to thousands), typical for medical imaging applications. To address this challenge, we propose an efficient and robust deep learning algorithm capable of full 3D detection in volumetric data. A two-step approach is exploited for efficient detection. A shallow network (with one hidden layer) is used for the initial testing of all voxels to obtain a small number of promising candidates, followed by more accurate classification with a deep network. In addition, we propose two approaches, i.e., separable filter decomposition and network sparsification, to speed up the evaluation of a network. To mitigate the over-fitting issue, thereby increasing detection robustness, we extract small 3D patches from a multi-resolution image pyramid. The deeply learned image features are further combined with Haar wavelet features to increase the detection accuracy. The proposed method has been quantitatively evaluated for carotid artery bifurcation detection on a head-neck CT dataset from 455 patients. Compared to the state-ofthe-art, the mean error is reduced by more than half, from 5.97 mm to 2.64 mm, with a detection speed of less than 1 s/volume.
1 Introduction There are many applications of automatic anatomical landmark detection in medical image analysis. For example, they can be used to align an input volume to a canonical plane on which physicians routinely perform diagnosis and quantification [1, 2]; A detected vascular landmark provides a seed point for automatic vessel centerline extraction and lumen segmentation [3]. Various landmark detection methods have been proposed in the literature. Most of the state-of-the-art algorithms [1–3] apply machine learning on a set of handcrafted image features. However, in practice, we found some landmark detection problems (e.g., carotid artery bifurcation landmarks in this work) are still too challenging to be solved with the current technology. Recently, deep learning [4] has demonstrated great success in computer vision with the capability to learn powerful image features from a large training set. However, several c Springer International Publishing Switzerland 2015 N. Navab et al. (Eds.): MICCAI 2015, Part I, LNCS 9349, pp. 565–572, 2015. DOI: 10.1007/978-3-319-24553-9_69
566
Y. Zheng et al.
challenges are present in applying deep learning to 3D landmark detection. Normally, the input to a neural network classifier is an image patch, which increases dramatically in size from 2D to 3D. For example, a patch of 32
Data Loading...
