Automatic segmentation of head and neck CT images for radiotherapy treatment planning using multiple atlases, statistical appearance models, and geodesic active contours

Fritscher, Karl D.; Peroni, M.; Zaffino, Paolo; Spadea, Maria Francesca; Schubert, Rainer; Sharp, G

doi:10.1118/1.4871623

Cited by 115 publications

(107 citation statements)

References 33 publications

Supporting

Mentioning

101

Contrasting

Unclassified

Order By: Relevance

“…Some methods use the output of label fusion to simply initialize a subsequent algorithm, for instance, to determine the bounding box where a segmentation method is applied (van Rikxoort et al, 2007a), to start the evolution of an active contour (Fritscher et al, 2014; Hollensen et al, 2010), or to fit a smooth contour to the object boundary (Nouranian et al, 2014). Other MAS algorithms rely on applying heavy post-processing to the label fusion output, for example by employing an error detection and correction classifier (Yushkevich et al 2010, who use AdaBoost), deriving features to drive a subsequent voxel-wise segmentation method, based for example on level sets (Gholipour et al, 2012; Schreibmann et al, 2014), random forests (Han, 2013), support vector machines (Hao et al, 2014), patch-based techniques (Wang et al, 2014e), or a graph-cut-based method (Candemir et al, 2014; Lee et al, 2014b).…”

Section: Survey Of Methodological Developmentsmentioning

confidence: 99%

“…Likewise, interest in radiotherapy treatment planning has been the main driver of applications in head, neck, and thoracic CT segmentation (Han et al, 2008; Wang et al, 2014c), which have mainly focused on segmenting tumors (Ramus and Malandain, 2010), organs at risk (e.g, the parotid glands, Fritscher et al 2014; Gorthi et al 2010; Han et al 2010; Hollensen et al 2010; Yang et al 2010 or mediastinal lymph nodes, Liu et al 2014) and lymph node metastases (Sjöberg et al, 2013; Teng et al, 2010). MAS has also been used in abdominal imaging, despite the relatively poor performance of image registration in this domain (e.g., compared with brain MRI) due to the shifting of organs within the abdominal cavity.…”

Section: Survey Of Applicationsmentioning

confidence: 99%

See 1 more Smart Citation

Multi-atlas segmentation of biomedical images: A survey

Iglesias

Sabuncu

2015

Medical Image Analysis

589

457

View full text Add to dashboard Cite

Multi-atlas segmentation (MAS), first introduced and popularized by the pioneering work of Rohlfing, Brandt, Menzel and Maurer Jr (2004), Klein, Mensh, Ghosh, Tourville and Hirsch (2005), and Heckemann, Hajnal, Aljabar, Rueckert and Hammers (2006), is becoming one of the most widely-used and successful image segmentation techniques in biomedical applications. By manipulating and utilizing the entire dataset of “atlases” (training images that have been previously labeled, e.g., manually by an expert), rather than some model-based average representation, MAS has the flexibility to better capture anatomical variation, thus offering superior segmentation accuracy. This benefit, however, typically comes at a high computational cost. Recent advancements in computer hardware and image processing software have been instrumental in addressing this challenge and facilitated the wide adoption of MAS. Today, MAS has come a long way and the approach includes a wide array of sophisticated algorithms that employ ideas from machine learning, probabilistic modeling, optimization, and computer vision, among other fields. This paper presents a survey of published MAS algorithms and studies that have applied these methods to various biomedical problems. In writing this survey, we have three distinct aims. Our primary goal is to document how MAS was originally conceived, later evolved, and now relates to alternative methods. Second, this paper is intended to be a detailed reference of past research activity in MAS, which now spans over a decade (2003 – 2014) and entails novel methodological developments and application-specific solutions. Finally, our goal is to also present a perspective on the future of MAS, which, we believe, will be one of the dominant approaches in biomedical image segmentation.

show abstract

Section: Survey Of Methodological Developmentsmentioning

confidence: 99%

Section: Survey Of Applicationsmentioning

confidence: 99%

Multi-atlas segmentation of biomedical images: A survey

Iglesias

Sabuncu

2015

Medical Image Analysis

589

457

View full text Add to dashboard Cite

show abstract

“…Our proposed editing method was evaluated on three challenging data sets: 1) prostate data set [32] including 73 CT images with dimension 512 × 512 × (61~81) voxel 3 and spacing 0.94 × 0.94 × 3.00 mm 3 , 2) brainstem data set [33] including 40 head & neck CT images with spacing approximately 1.0 × 1.0 × (2.5~3.0) mm 3 , and 3) hippocampus data set 1 including 35 brain MR images with dimension 256 × 256 × 287 voxel 3 and spacing 1.0 × 1.0 × 1.0 mm 3 (Fig 5). Although several automatic methods [17, 33–35] have been proposed to address these segmentation problems, inaccurate results were often obtained due to low tissue contrast and large shape and appearance variations.…”

Section: Resultsmentioning

confidence: 99%

“…Although several automatic methods [17, 33–35] have been proposed to address these segmentation problems, inaccurate results were often obtained due to low tissue contrast and large shape and appearance variations. To evaluate editing performances, we first used one of the state-of-the-art automatic methods to generate the initial segmentation, and then applied our editing method to the half of the results with the largest errors.…”

Section: Resultsmentioning

confidence: 99%

Multiatlas-Based Segmentation Editing With Interaction-Guided Patch Selection and Label Fusion

Park

Gao

Shen

2016

IEEE Trans. Biomed. Eng.

View full text Add to dashboard Cite

We propose a novel multi-atlas based segmentation method to address the segmentation editing scenario, where an incomplete segmentation is given along with a set of existing reference label images (used as atlases). Unlike previous multi-atlas based methods, which depend solely on appearance features, we incorporate interaction-guided constraints to find appropriate atlas label patches in the reference label set and derive their weights for label fusion. Specifically, user interactions provided on the erroneous parts are first divided into multiple local combinations. For each combination, the atlas label patches well-matched with both interactions and the previous segmentation are identified. Then, the segmentation is updated through the voxel-wise label fusion of selected atlas label patches with their weights derived from the distances of each underlying voxel to the interactions. Since the atlas label patches well-matched with different local combinations are used in the fusion step, our method can consider various local shape variations during the segmentation update, even with only limited atlas label images and user interactions. Besides, since our method does not depend on either image appearance or sophisticated learning steps, it can be easily applied to general editing problems. To demonstrate the generality of our method, we apply it to editing segmentations of CT prostate, CT brainstem, and MR hippocampus, respectively. Experimental results show that our method outperforms existing editing methods in all three data sets.

show abstract

“…Finally, multi-atlas based and learning-based methods have been heavily used in many other segmentation problems, such as CT head and neck segmentation 29 and tooth segmentation. 30 In our future work, we also plan to evaluate our method in those applications.…”

Section: Discussionmentioning

confidence: 99%

Nonlocal atlas‐guided multi‐channel forest learning for human brain labeling

Gao

et al. 2016

Medical Physics

View full text Add to dashboard Cite

Purpose: It is important for many quantitative brain studies to label meaningful anatomical regions in MR brain images. However, due to high complexity of brain structures and ambiguous boundaries between different anatomical regions, the anatomical labeling of MR brain images is still quite a challenging task. In many existing label fusion methods, appearance information is widely used. However, since local anatomy in the human brain is often complex, the appearance information alone is limited in characterizing each image point, especially for identifying the same anatomical structure across different subjects. Recent progress in computer vision suggests that the context features can be very useful in identifying an object from a complex scene. In light of this, the authors propose a novel learning-based label fusion method by using both low-level appearance features (computed from the target image) and high-level context features (computed from warped atlases or tentative labeling maps of the target image). Methods: In particular, the authors employ a multi-channel random forest to learn the nonlinear relationship between these hybrid features and target labels (i.e., corresponding to certain anatomical structures). Specifically, at each of the iterations, the random forest will output tentative labeling maps of the target image, from which the authors compute spatial label context features and then use in combination with original appearance features of the target image to refine the labeling. Moreover, to accommodate the high inter-subject variations, the authors further extend their learning-based label fusion to a multi-atlas scenario, i.e., they train a random forest for each atlas and then obtain the final labeling result according to the consensus of results from all atlases. Results: The authors have comprehensively evaluated their method on both public LONI_LBPA40 and IXI datasets. To quantitatively evaluate the labeling accuracy, the authors use the dice similarity coefficient to measure the overlap degree. Their method achieves average overlaps of 82.56% on 54 regions of interest (ROIs) and 79.78% on 80 ROIs, respectively, which significantly outperform the baseline method (random forests), with the average overlaps of 72.48% on 54 ROIs and 72.09% on 80 ROIs, respectively. Conclusions: The proposed methods have achieved the highest labeling accuracy, compared to several state-of-the-art methods in the literature. C 2016 American Association of Physicists in Medicine. [http://dx

show abstract

Automatic segmentation of head and neck CT images for radiotherapy treatment planning using multiple atlases, statistical appearance models, and geodesic active contours

Cited by 115 publications

References 33 publications

Multi-atlas segmentation of biomedical images: A survey

Multi-atlas segmentation of biomedical images: A survey

Multiatlas-Based Segmentation Editing With Interaction-Guided Patch Selection and Label Fusion

Nonlocal atlas‐guided multi‐channel forest learning for human brain labeling

Contact Info

Product

Resources

About