Atlas based segmentation and mapping of organs at risk from planning CT for the development of voxel-wise predictive models of toxicity in prostate radiotherapy

Published on Sep 24, 2010 in MICCAI (Medical Image Computing and Computer-Assisted Intervention)
· DOI :10.1007/978-3-642-15989-3_6
Oscar Acosta21
Estimated H-index: 21
(University of Rennes),
Jason Dowling18
Estimated H-index: 18
(CSIRO: Commonwealth Scientific and Industrial Research Organisation)
+ 4 AuthorsPascal Haigron21
Estimated H-index: 21
(University of Rennes)
The prediction of toxicity is crucial to managing prostate cancer radiotherapy (RT). This prediction is classically organ wise and based on the dose volume histograms (DVH) computed during the planning step, and using for example the mathematical Lyman Normal Tissue Complication Probability (NTCP) model. However, these models lack spatial accuracy, do not take into account deformations and may be inappropiate to explain toxicity events related with the distribution of the delivered dose. Producing voxel wise statistical models of toxicity might help to explain the risks linked to the dose spatial distribution but is challenging due to the difficulties lying on the mapping of organs and dose in a common template. In this paper we investigate the use of atlas based methods to perform the non-rigid mapping and segmentation of the individuals' organs at risk (OAR) from CT scans. To build a labeled atlas, 19 CT scans were selected from a population of patients treated for prostate cancer by radiotherapy. The prostate and the OAR (Rectum, Bladder, Bones) were then manually delineated by an expert and constituted the training data. After a number of affine and non rigid registration iterations, an average image (template) representing the whole population was obtained. The amount of consensus between labels was used to generate probabilistic maps for each organ. We validated the accuracy of the approach by segmenting the organs using the training data in a leave one out scheme. The agreement between the volumes after deformable registration and the manually segmented organs was on average above 60% for the organs at risk. The proposed methodology provides a way to map the organs from a whole population on a single template and sets the stage to perform further voxel wise analysis. With this method new and accurate predictive models of toxicity will be built.
📖 Papers frequently viewed together
76 Citations
4,519 Citations
58 Citations
#1Ingelise Jensen (AU: Aarhus University)H-Index: 1
#2Jesper Carl (AU: Aarhus University)H-Index: 14
Last. Jane Nielsen (AU: Aarhus University)H-Index: 6
view all 5 authors...
Abstract Dose escalation in prostate radiotherapy is limited by normal tissue toxicities. The aim of this study was to assess the impact of margin size on tumor control and side effects for intensity-modulated radiation therapy (IMRT) and 3D conformal radiotherapy (3DCRT) treatment plans with increased dose. Eighteen patients with localized prostate cancer were enrolled. 3DCRT and IMRT plans were compared for a variety of margin sizes. A marker detectable on daily portal images was presupposed f...
16 CitationsSource
#1Wilma D. Heemsbergen (NKI-AVL: Netherlands Cancer Institute)H-Index: 29
#2Abrahim Al-Mamgani (EUR: Erasmus University Rotterdam)H-Index: 27
Last. Joos V. Lebesque (NKI-AVL: Netherlands Cancer Institute)H-Index: 18
view all 6 authors...
Purpose To investigate the relationship between late urinary obstruction and the details of the dose distribution of irradiated prostate cancer patients, taking into account their baseline symptoms and acute complaints. Patients and Methods We selected patients from the Dutch multicenter trial randomized between 68 Gy and 78 Gy, for whom toxicity data and dose data were available ( n = 557). The absolute dose surface parameters of the delineated bladder were calculated. Next, we constructed thre...
85 CitationsSource
#1Marnix G. Witte (NKI-AVL: Netherlands Cancer Institute)H-Index: 18
#2Wilma D. Heemsbergen (NKI-AVL: Netherlands Cancer Institute)H-Index: 29
Last. Marcel van Herk (NKI-AVL: Netherlands Cancer Institute)H-Index: 75
view all 7 authors...
Purpose For prostate cancer patients at risk for subclinical spread of the disease, we investigated whether incidental dose outside the target was associated with tumor control. Methods and Materials We selected 352 intermediate-risk (mainly T2b-T3a) and high-risk (mainly T3b) patients treated in a randomized trial. Target volume was prostate (68–78 Gy) and seminal vesicles (50–78 Gy). Failure (clinical or biochemical) was evaluated at 4 years. To compare three-dimensional dose distributions, an...
35 CitationsSource
#1David A. Jaffray (U of T: University of Toronto)H-Index: 96
#2Patricia Lindsay (U of T: University of Toronto)H-Index: 19
Last. Wolfgang A. Tomé (UW: University of Wisconsin-Madison)H-Index: 62
view all 5 authors...
The actual distribution of radiation dose accumulated in normal tissues over the complete course of radiation therapy is, in general, poorly quantified. Differences in the patient anatomy between planning and treatment can occur gradually (e.g., tumor regression, resolution of edema) or relatively rapidly (e.g., bladder filling, breathing motion) and these undermine the accuracy of the planned dose distribution. Current efforts to maximize the therapeutic ratio require models that relate the tru...
116 CitationsSource
#1C. Fiorino (UniSR: Vita-Salute San Raffaele University)H-Index: 7
#2Tiziana RancatiH-Index: 27
Last. Riccardo ValdagniH-Index: 51
view all 3 authors...
Dose-volume modeling of late and acute toxicity in radiotherapy for prostate cancer is a rapidly evolving field of investigation. The availability of individual, 3-dimensional dose distribution and dose-volume histograms (DVHs) permits the quantitative assessment of dose-volume relations for specific endpoints by investigating the correlation between individual dose-volume data and clinical outcomes. These studies often entail a huge effort in collecting data from large populations that have bee...
36 CitationsSource
#1Raffaella CambriaH-Index: 16
#2Barbara Alicja Jereczek-Fossa (University of Milan)H-Index: 36
Last. Roberto Orecchia (University of Milan)H-Index: 71
view all 9 authors...
Purpose: To analyze the reliability of different methods used in evaluating the risk of late rectal toxicity. Patients and Methods: The treatment plans of 57 patients treated at the authors’ institute between September 1999 and September 2000 for localized prostate cancer using three-dimensional conformal radiotherapy (3D-CRT) were analyzed retrospectively. The expected rate of late rectal toxicity was analyzed (a) by means of the dose-volume histogram (DVH) constraints; (b) by calculating the n...
16 CitationsSource
Purpose To quantify adequate anisotropic clinical target volume (CTV)-to-planning target volume (PTV) margins for three different setup strategies used during prostate irradiation: (1) no setup corrections, (2) on-line corrections determined from bony anatomy, and (3) on-line corrections determined from gold markers. Method and Materials Three radiation oncologists independently delineated the CTV on computed tomography images of 30 prostate cancer patients. Eight repeat scans were acquired to a...
59 CitationsSource
#1Connor Kupchak (UWO: University of Western Ontario)H-Index: 10
#2Jerry J. Battista (UWO: University of Western Ontario)H-Index: 47
Last. Jacob Van Dyk (UWO: University of Western Ontario)H-Index: 15
view all 3 authors...
Commonly, the quality of treatment plans is judged by a dose-volume histogram (DVH) in regards to satisfying a series of dose-volume constraints. This paper presents a novel technique for mapping normal tissue complication probabilities (NTCP) onto regions of dose-volume space with statistical considerations of risk. Mapping is done for DVHs specific to one treatment technique for one disease site. In this study, the method is illustrated for simplified intensity modulated arc therapy of the pro...
14 CitationsSource
#1Tom Vercauteren (Mauna Kea Technologies)H-Index: 45
#2Xavier Pennec (IRIA: French Institute for Research in Computer Science and Automation)H-Index: 70
Last. Nicholas Ayache (IRIA: French Institute for Research in Computer Science and Automation)H-Index: 105
view all 4 authors...
We propose a non-parametric diffeomorphic image registration algorithm based on Thirion's demons algorithm. The demons algorithm can be seen as an optimization procedure on the entire space of displacement fields. The main idea of our algorithm is to adapt this procedure to a space of diffeomorphic transformations. In contrast to many diffeomorphic registration algorithms, our solution is computationally efficient since in practice it only replaces an addition of free form deformations by a few ...
295 CitationsSource
#1Matthias SöhnH-Index: 16
#2Markus AlberH-Index: 32
Last. Di Yan (Beaumont Hospital)H-Index: 69
view all 3 authors...
Purpose: The variability of dose-volume histogram (DVH) shapes in a patient population can be quantified using principal component analysis (PCA). We applied this to rectal DVHs of prostate cancer patients and investigated the correlation of the PCA parameters with late bleeding. Methods and Materials: PCA was applied to the rectal wall DVHs of 262 patients, who had been treated with a four-field box, conformal adaptive radiotherapy technique. The correlated changes in the DVH pattern were revea...
37 CitationsSource
Cited By23
#1James Coates (Harvard University)H-Index: 8
#2Giacomo Pirovano (MSK: Memorial Sloan Kettering Cancer Center)H-Index: 9
Last. Issam El NaqaH-Index: 58
view all 3 authors...
The power of predictive modeling for radiotherapy outcomes has historically been limited by an inability to adequately capture patient-specific variabilities; however, next-generation platforms together with imaging technologies and powerful bioinformatic tools have facilitated strategies and provided optimism. Integrating clinical, biological, imaging, and treatment-specific data for more accurate prediction of tumor control probabilities or risk of radiation-induced side effects are high-dimen...
#1Oscar AcostaH-Index: 21
2 CitationsSource
#1Anjali Balagopal (UTSW: University of Texas Southwestern Medical Center)H-Index: 5
#2Samaneh Kazemifar (UTSW: University of Texas Southwestern Medical Center)H-Index: 5
Last. Steve B Jiang (UTSW: University of Texas Southwestern Medical Center)H-Index: 83
view all 7 authors...
: Accurate segmentation of prostate and surrounding organs at risk is important for prostate cancer radiotherapy treatment planning. We present a fully automated workflow for male pelvic CT image segmentation using deep learning. The architecture consists of a 2D organ volume localization network followed by a 3D segmentation network for volumetric segmentation of prostate, bladder, rectum, and femoral heads. We used a multi-channel 2D U-Net followed by a 3D U-Net with encoding arm modified with...
51 CitationsSource
#1Kensaku Mori (Nagoya University)H-Index: 42
#2Noboru Niki (University of Tokushima)H-Index: 20
Last. Hirohiko Kimura (University of Fukui)H-Index: 27
view all 11 authors...
This chapter shows applied technologies using computational anatomy (CA) models. CA systems based on clinical images assist physicians by providing useful information related to diagnostic and therapeutic procedures. Such systems include computer-aided diagnosis and computer-assisted surgery systems. A thorough understanding of anatomy is essential when designing these systems. It is important to understand how anatomical information extracted by a computer is used. In this chapter, we introduce...
#1James Coates (University of Oxford)H-Index: 8
#2Luis Souhami (MUHC: McGill University Health Centre)H-Index: 61
Last. Issam El Naqa (UM: University of Michigan)H-Index: 58
view all 3 authors...
Radiation therapy is a first-line treatment option for localized prostate cancer and radiation-induced normal tissue damage are often the main limiting factor for modern radiotherapy regimens. Conversely, under-dosing of target volumes in an attempt to spare adjacent healthy tissues limits the likelihood of achieving local, long-term control. Thus, the ability to generate personalized data-driven risk profiles for radiotherapy outcomes would provide valuable prognostic information to help guide ...
30 CitationsSource
#1Andrzej Skalski (AGH University of Science and Technology)H-Index: 12
#2Artur Kos (AGH University of Science and Technology)H-Index: 3
Last. Paweł Kukołowicz (Curie Institute)H-Index: 9
view all 5 authors...
In the paper a new method for prostate segmentation in computed tomography (CT) data is proposed. In the proposed approach, first, corresponding points of training data sets are found using point clouds generation by Marching Cubes algorithm and non-rigid Iterative Closest Points registration. After that, having the corresponding points available, the statistical model of the prostate is built by the Active Shape Model (ASM). As a feature vector histogram of image gradient (HoG) is utilized. Fin...
4 CitationsSource
Apr 16, 2015 in ISBI (International Symposium on Biomedical Imaging)
#1Maxime Guinin (University of Rouen)H-Index: 2
#2Su Ruan (University of Rouen)H-Index: 27
Last. Isabelle Gardin (University of Rouen)H-Index: 21
view all 6 authors...
Segmentation of organs at risk (OAR) in male pelvis is critical for planning prostate cancer radiotherapy. We are interested in femoral heads, rectum and bladder segmentation in magnetic resonance imaging (MRI) and computed tomography (CT) images in order to protect OARs during radiotherapy planning. The proposed methodology is based on superpixel algorithm in order to over-segment patient image by solving a local Eikonal function from initial seeds. Afterwards, the segmentation is obtained by c...
4 CitationsSource
#1Fabio Martínez (National University of Colombia)H-Index: 6
#2Eduardo Romero (National University of Colombia)H-Index: 21
Last. Oscar Acosta (University of Rennes)H-Index: 21
view all 7 authors...
Accurate segmentation of the prostate and organs at risk in computed tomography (CT) images is a crucial step for radiotherapy planning. Manual segmentation, as performed nowadays, is a time consuming process and prone to errors due to the a high intra- and inter-expert variability. This paper introduces a new automatic method for prostate, rectum and bladder segmentation in planning CT using a geometrical shape model under a Bayesian framework. A set of prior organ shapes are first built by app...
38 CitationsSource
#1Oscar Acosta (University of Rennes)H-Index: 21
#2Jason Dowling (CSIRO: Commonwealth Scientific and Industrial Research Organisation)H-Index: 18
Last. Pascal Haigron (University of Rennes)H-Index: 21
view all 6 authors...
In prostate cancer radiotherapy, the accurate identification of the prostate and organs at risk in planning computer tomography (CT) images is an important part of the therapy planning and optimization. Manually contouring these organs can be a time-consuming process and subject to intra- and inter-expert variability. Automatic identification of organ boundaries from these images is challenging due to the poor soft tissue contrast. Atlas-based approaches may provide a priori structural informati...
24 CitationsSource