Expert-driven label fusion in multi-atlas-based segmentation of the prostate using weighted atlases

Langerak, Thomas; Heide, Uulke A. van der; Kotte, Antj Alexis; Berendsen, Floris G; Vulpen, Marco van; Pluim, Jpw Josien

doi:10.1007/s11548-013-0836-4

Cited by 5 publications

(3 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These semi‐automatic techniques described in the previous section are very powerful since they utilize a spectrum of anatomical information, that is, size, repetitive form of geometry, organ‐dependent features, shape, etc. The early efforts include atlas‐based segmentation of brain 41–43 and head and neck organs on T1w MRI using subject‐specific atlases, 44 multi‐atlas approach, and label fusion 45,46 . In addition, mathematical modeling using gradients, region, and shape‐based constraints have proven to be successful for segmentation of selective abdomen organs at risk 47 and the pancreas on T2w and T1w volumetric interpolated breath‐hold examination (VIBE) MR images 48,49 .…”

Section: Segmentationmentioning

confidence: 99%

“…The early efforts include atlas-based segmentation of brain [41][42][43] and head and neck organs on T1w MRI using subject-specific atlases, 44 multi-atlas approach, and label fusion. 45,46 In addition, mathematical modeling using gradients,region,and shape-based constraints have proven to be successful for segmentation of selec-tive abdomen organs at risk 47 and the pancreas on T2w and T1w volumetric interpolated breath-hold examination (VIBE) MR images. 48,49 While atlas-based segmentation is a useful tool, the quality of the output is strongly dependent on registration accuracy and quality of the atlas itself.…”

Section: Automatic Segmentationmentioning

confidence: 99%

See 1 more Smart Citation

Multi‐parametric magnetic resonance imaging for radiation treatment planning

et al. 2022

View full text Add to dashboard Cite

In recent years, multi-parametric magnetic resonance imaging (MpMRI) has played a major role in radiation therapy treatment planning. The superior soft tissue contrast, functional or physiological imaging capabilities, and the flexibility of site-specific image sequence development has placed MpMRI at the forefront. In this article, the present status of MpMRI for external beam radiation therapy planning is reviewed. Common MpMRI sequences, preprocessing, and quality assurance strategies are briefly discussed, and various image registration techniques and strategies are addressed. Image segmentation methods including automatic segmentation and deep learning techniques for organs at risk and target delineation are reviewed. Due to the advancement in MRI-guided online adaptive radiotherapy, treatment planning considerations addressing MRI only planning are also discussed. K E Y W O R D S MpMRI, MRI for treatment planning, MRI-guided radiotherapy 1 2 IMAGE SELECTION AND REGISTRATION Brief description of common MpMRI sequencesMpMRI sequences typically include multi-slice 2D and 3D images with T1 and T2 weighting and a variety of 2836

show abstract

Section: Segmentationmentioning

confidence: 99%

Section: Automatic Segmentationmentioning

confidence: 99%

Multi‐parametric magnetic resonance imaging for radiation treatment planning

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Langerak et al recently developed an expert-driven label fusion method, which incorporates the expert knowledge on the region-dependent requirement of the accuracy levels into the label fusion (Langerak et al 2013). By including this expert knowledge, the segmentation accuracy was increased in the regions of the vulnerable tissue with a decreased accuracy in less clinically relevant regions.…”

Section: Introductionmentioning

confidence: 99%

A novel geometry-dosimetry label fusion method in multi-atlas segmentation for radiotherapy: a proof-of-concept study

Chang

Tian²,

et al. 2017

Phys. Med. Biol.

View full text Add to dashboard Cite

Multi-atlas segmentation (MAS) has been widely used to automate the delineation of organs at risk (OARs) for radiotherapy. Label fusion is a crucial step in MAS to cope with the segmentation variabilities among multiple atlases. However, most existing label fusion methods do not consider the potential dosimetric impact of the segmentation result. In this proof-of-concept study, we propose a novel geometry-dosimetry label fusion method for MAS-based OAR auto-contouring, which evaluates the segmentation performance in terms of both geometric accuracy and the dosimetric impact of the segmentation accuracy on the resulting treatment plan. Differently from the original selective and iterative method for performance level estimation (SIMPLE), we evaluated and rejected the atlases based on both Dice similarity coefficient and the predicted error of the dosimetric endpoints. The dosimetric error was predicted using our previously developed geometry-dosimetry model. We tested our method in MAS-based rectum auto-contouring on 20 prostate cancer patients. The accuracy in the rectum sub-volume close to the planning tumor volume (PTV), which was found to be a dosimetric sensitive region of the rectum, was greatly improved. The mean absolute distance between the obtained contour and the physician-drawn contour in the rectum sub-volume 2 mm away from PTV was reduced from 3.96 mm to 3.36 mm on average for the 20 patients, with the maximum decrease found to be from 9.22 mm to 3.75 mm. We also compared the dosimetric endpoints predicted for the obtained contours with those predicted for the physician-drawn contours. Our method led to smaller dosimetric endpoint errors than the SIMPLE method in 15 patients, comparable errors in 2 patients, and slightly larger errors in 3 patients. These results indicated the efficacy of our method in terms of considering both geometric accuracy and dosimetric impact during label fusion. Our algorithm can be applied to different tumor sites and radiation treatments, given a specifically trained geometry-dosimetry model.

show abstract

Conditional Generative Adversarial Networks with Multi-scale Discriminators for Prostate MRI Segmentation

Liu

et al. 2020

Neural Process Lett

View full text Add to dashboard Cite

Expert-driven label fusion in multi-atlas-based segmentation of the prostate using weighted atlases

Cited by 5 publications

References 18 publications

Multi‐parametric magnetic resonance imaging for radiation treatment planning

Multi‐parametric magnetic resonance imaging for radiation treatment planning

A novel geometry-dosimetry label fusion method in multi-atlas segmentation for radiotherapy: a proof-of-concept study

Conditional Generative Adversarial Networks with Multi-scale Discriminators for Prostate MRI Segmentation

Contact Info

Product

Resources

About