Fast Automatic Multi-atlas Segmentation of the Prostate from 3D MR Images.
ABSTRACT A fast fully automatic method of segmenting the prostate from 3D MR scans is presented, incorporating dynamic multi-atlas label fusion. The diffeomorphic demons method is used for non-rigid registration and a comparison of alternate metrics for atlas selection is presented. A comparison of results from an average shape atlas and the multi-atlas approach is provided. Using the same clinical dataset and manual contours from 50 clinical scans as Klein et al. (2008) a median Dice similarity coefficient of 0.86 was achieved with an average surface error of 2.00mm using the multi-atlas segmentation method.
- SourceAvailable from: Fabrice Meriaudeau[Show abstract] [Hide abstract]
ABSTRACT: Prostate segmentation aids in prostate volume estimation, multi-modal image registration, and to create patient specific anatomical models for surgical planning and image guided biopsies. However, manual segmentation is time consuming and suffers from inter-and intra-observer variabilities. Low contrast images of trans rectal ultrasound and presence of imaging artifacts like speckle, micro-calcifications, and shadow regions hinder computer aided automatic or semi-automatic prostate segmentation. In this paper, we propose a prostate segmentation approach based on building multiple mean parametric models derived from principal component analysis of shape and posterior probabilities in a multi-resolution framework. The model parameters are then modified with the prior knowledge of the optimization space to achieve optimal prostate segmentation. In contrast to traditional statistical models of shape and intensity priors, we use posterior probabilities of the prostate region determined from random forest classification to build our appearance model, initialize and propagate our model. Furthermore, multiple mean models derived from spectral clustering of combined shape and appearance parameters are applied in parallel to improve segmentation accuracies. The proposed method achieves mean Dice similarity coefficient value of 0.91±0.09 for 126 images containing 40 images from the apex, 40 images from the base and 46 images from central regions in a leave-one-patient-out validation framework. The mean segmentation time of the procedure is 0.67±0.02s.Medical image analysis 04/2013; 17(6):587-600. · 3.09 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: To perform a comparative study assessing potential benefits of endorectal-balloons (ERB) in post-prostatectomy patients. Ten retrospective post-prostatectomy patients treated without ERB and ten prospective patients treated with the ERB in situ were recruited. All patients received IMRT and IGRT using kilovoltage cone-beam computed tomography (kVCBCT). kVCBCT datasets were registered to the planning dataset, recontoured and the original plan recalculated on the kVCBCTs to recreate anatomical conditions during treatment. The imaging, structure and dose data were imported into in-house software for the assessment of geometric variation and cumulative equivalent uniform dose (EUD) in the two groups. The difference in location (ΔCOV) for the bladder between planning and each CBCT was similar for each group. The range of mean ΔCOV for the rectum was 0.15-0.58cm and 0.15-0.59cm for the non-ERB and ERB groups. For superior-CTV and inferior-CTV the difference between planned and delivered D95% (mean±SD) for the non-ERB group was 2.1±6.0Gy and -0.04±0.20Gy. While for the ERB group the difference in D95% was 8.7±12.6Gy and 0.003±0.104Gy. The use of ERBs in the post-prostatectomy setting did improve geometric reproducibility of the target and surrounding normal tissues, however no improvement in dosimetric stability was observed for the margins employed.Radiotherapy and Oncology 09/2013; · 4.52 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Clinical validation and quantitative evaluation of computed tomography (CT) image autosegmentation using Smart Probabilistic Image Contouring Engine (SPICE). CT images of 125 treated patients (32 head and neck [HN], 40 thorax, 23 liver, and 30 prostate) in 7 independent institutions were autosegmented using SPICE and computational times were recorded. The number of structures autocontoured were 25 for the HN, 7 for the thorax, 3 for the liver, and 6 for the male pelvis regions. Using the clinical contours as reference, autocontours of 22 selected structures were quantitatively evaluated using Dice Similarity Coefficient (DSC) and Mean Slice-wise Hausdorff Distance (MSHD). All 40 autocontours were evaluated by a radiation oncologist from the institution that treated the patients. The mean computational times to autosegment all the structures using SPICE were 3.1 to 11.1 minutes per patient. For the HN region, the mean DSC was >0.70 for all evaluated structures, and the MSHD ranged from 3.2 to 10.0 mm. For the thorax region, the mean DSC was 0.95 for the lungs and 0.90 for the heart, and the MSHD ranged from 2.8 to 12.8 mm. For the liver region, the mean DSC was >0.92 for all structures, and the MSHD ranged from 5.2 to 15.9 mm. For the male pelvis region, the mean DSC was >0.76 for all structures, and the MSHD ranged from 4.8 to 10.5 mm. Out of the 40 autocontoured structures reviews by experts, 25 were scored useful as autocontoured or with minor edits for at least 90% of the patients and 33 were scored useful autocontoured or with minor edits for at least 80% of the patients. Compared with manual contouring, autosegmentation using SPICE for the HN, thorax, liver, and male pelvis regions is efficient and shows significant promise for clinical utility.International journal of radiation oncology, biology, physics 11/2013; 87(4):809-16. · 4.59 Impact Factor