Automatic segmentation of the spine by means of a probabilistic atlas with a special focus on ribs suppression. Preliminary results

Abstract

Spine is a structure commonly involved in several prevalent diseases. In clinical diagnosis, therapy, and surgical intervention, the identification and segmentation of the vertebral bodies are crucial steps. However, automatic and detailed segmentation of vertebrae is a challenging task, especially due to the proximity of the vertebrae to the corresponding ribs and other structures such as blood vessels. In this study, to overcome these problems, a probabilistic atlas of the spine, including cervical, thoracic and lumbar vertebrae has been built to introduce anatomical knowledge in the segmentation process, aiming to deal with overlapping gray levels and the proximity to other structures. From a set of 3D images manually segmented by a physician (training data), a 3D volume indicating the probability of each voxel of belonging to the spine has been developed, being necessary the generation of a probability map and its deformation to adapt to each patient. To validate the improvement of the segmentation using the atlas developed in the testing data, we computed the Hausdorff distance between the manually-segmented ground truth and an automatic segmentation and also between the ground truth and the automatic segmentation refined with the atlas. The results are promising, obtaining a higher improvement especially in the thoracic region, where the ribs can be found and appropriately eliminated.

Anatomy-Aware Inference of the 3D Standing Spine Posture from 2D Radiographs

“…Besides statistical shape models [10,21,23,24], other probabilistic models, such as probabilistic atlas [25], graph models [26], Hidden Markov Models [27] and hierarchical models [28][29][30] have also been proposed. For instance, in [25], the authors proposed a probabilistic atlas of the spine.…”

“…Besides statistical shape models [10,21,23,24], other probabilistic models, such as probabilistic atlas [25], graph models [26], Hidden Markov Models [27] and hierarchical models [28][29][30] have also been proposed. For instance, in [25], the authors proposed a probabilistic atlas of the spine. By co-registering 21 CT scans, a probability map is created which can be used to segment and detect the vertebrae with a special focus on ribs suppression.…”

Automatic segmentation of the spine by means of a probabilistic atlas with a special focus on ribs suppression

“…The first type, normally known as statistical atlas, is usually constructed by using techniques of Principal Component Analysis (PCA) using as input data the spatial coordinates of a set of relevant points (landmarks) chosen either manually or automatically . The second type, probabilistic atlases, utilizes techniques based on mathematical morphology and probabilistic models …”

“…30,31 The second type, probabilistic atlases, utilizes techniques based on mathematical morphology and probabilistic models. [32][33][34] With both approaches, considerable research effort has been directed toward the construction of brain atlases. 35 A variety of methods have been also proposed for the segmentation of other organs or anatomical structures by means of atlases.…”

OncoSpineSeg: A Software Tool for a Manual Segmentation of Computed Tomography of the Spine on Cancer Patients

“…These methods are known as active appearance models [20]. Using an atlas-based segmentation is also a way to introduce anatomical information related to the position of an organ [21][22][23]. The raw data values that are stored in an image are not always sufficiently informative, especially in the case of organs whose limit surface is not clearly defined in terms of signal value.…”