Consistency of three-dimensional planning target volumes across physicians and institutions

Ketting, Case H.; Austin-Seymour, Mary; Kalet, Ira J.; Unger, Jonathan; Hummel, Sharon; Jacky, Jonathan

doi:10.1016/s0360-3016(96)00459-2

Cited by 66 publications

(23 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[6][7][8][9][10][11] This may in fact be the most important source of variation in some cases. Intraobserver variability has also been studied, although usually, its magnitude is less than that of interobserver variation.…”

Section: Discussionmentioning

confidence: 99%

CT slice index and thickness: Impact on organ contouring in radiation treatment planning for prostate cancer

Berthelet¹

2003

J. Appl. Clin. Med. Phys.

View full text Add to dashboard Cite

Objective: To assess the impact of CT slice index and thickness ͑3 mm versus 5 mm͒ on ͑i͒ prostate volume, dimensions, and isocenter coordinates, ͑ii͒ bladder and rectal volumes, and ͑iii͒ DRR quality, in the treatment of prostate cancer. Methods: 16 patients with prostate cancer underwent two planning CT-scans using 3 and 5 mm slice index/thickness. Prostate, bladder, and rectum were outlined on all scans. Prostate isocenter coordinates, maximum dimensions, and volumes were compared along with bladder and rectal volumes. Bladder volumes and maximum diameters were further investigated using a second observer. A comparative analysis of DRR quality was conducted as well as a dosimetric analysis using DVH. Results: The differences in measurements of prostate volume, isocenter coordinates and maximum dimensions between the 3 and 5 mm scans, were small and not statistically significant. Similar finding was seen for rectal volume. However, bladder volume was always larger on the 3 mm scan ͑mean differenceϭ27.9 cc; SEϭ4.8 cc; 95% CI: 17.7-38.2 cc; pϽ0.001) and the findings were reproduced with the second observer ͑mean differenceϭ31.9 cc; SEϭ4.7 cc; 95% CI: 21.9-41.9 cc; p Ͻ0.001). The differences in volume are caused by a slight increase in ͑1͒ the measurement of the longitudinal dimensions on the 3 mm scans, and ͑2͒ the slice by slice measured bladder area on the 3 mm scans. The latter is due to partial volume effect. The 3 mm DRR were slightly better than the 5 mm DRR. The bladder DVH differed significantly in some patients. Conclusion: Bladder volume is significantly larger on the 3 mm scans. Differences in contoured areas may be accounted for, in part, by the partial volume effect.

show abstract

Section: Discussionmentioning

confidence: 99%

CT slice index and thickness: Impact on organ contouring in radiation treatment planning for prostate cancer

Berthelet¹

2003

J. Appl. Clin. Med. Phys.

View full text Add to dashboard Cite

show abstract

“…There was very substantial variation between individual clinicians. Ketting et al (13) reported variability in PTV de® nition for nasopharyngeal and lung cancer, but inter-physician and inter-radiotherapy centre variability were not studied separately. Three institutions participated in the study.…”

Section: Discussionmentioning

confidence: 99%

“…The PTV was drawn on three randomly designated CT slices and the volumes calculated are therefore not comparable to our results. However, analysis of variance showed that both physician (from different centres) and clinical case were highly signi® cant factors affecting PTV (13). Fiorino et al (12) studied inter-physician variability in CTVs in prostate cancer, using three-dimensional treatment planning tools.…”

Section: Discussionmentioning

confidence: 99%

“…In more recent studies (12± 15) the inter-radiotherapy centre variation has been investigated, but inter-physician (physician to physician) variation has not been considered separate from this (13). Inter-physician variability in PTVs and CTVs for the radiotherapy of prostate cancer using three-dimensional treatment planning systems has been evaluated in the most recent studies (14,16).…”

mentioning

confidence: 99%

See 1 more Smart Citation

Quality Assurance in Radiotherapy of Breast Cancer Variability in Planning Target Volume Delineation

Pitkänen

2001

Acta Oncologica

View full text Add to dashboard Cite

“…Most methods in routine clinical practice are user-guided slice-by-slice contouring tools that require well trained users to achieve acceptable results for 3D RTP. Other flaws of current segmentation methods that tend toward sub-optimal treatment planning include intra-and inter-user variabilities [Dowsett 1992;Leunens 1993;Valley 1993;Kagawa 1997;Roach 1996;Algan 1995;Ketting 1997aKetting , 1997bRasch 1999], the lack of practical approaches that fully consider all three spatial dimensions, and the inability to deal with ambiguous surface localization.The development of automatic 3D segmentation methods is motivated by several considerations including economic pressure to improve efficiency and contain costs, and the clinical need to improve accuracy and reproducibility in order to steer user-directed planning decisions and inverse treatment planning algorithms consistently in the right direction. Deformable shape models is a general class that is showing great promise for automatic segmentation of normal anatomical structures.…”

mentioning

confidence: 99%

Department of Radiation Oncology

2015

Juntendo Medical Journal

View full text Add to dashboard Cite

Purpose: A controlled observer study was conducted to compare a method for automatic image segmentation with conventional user-guided segmentation of right and left kidneys from planning CT images. Materials and Methods:Deformable shape models called m-reps were used to automatically segment right and left kidneys from twelve target CT images and the results were compared to careful manual segmentations performed by two human experts. M-rep models were trained based on manual segmentations from a collection of images that did not include the targets. Segmentation using m-reps began with interactive initialization to position the kidney model over the target kidney in the image data. Fully automatic segmentation proceeded through two stages at successively smaller spatial scales. At the first stage a global similarity transformation of the kidney model was computed to position the model closer to the target kidney. The similarity transformation was followed by large scale deformations based on principal geodesic analysis (PGA). During the second stage each medial atom comprising the m-rep model was deformed in turn and iteratively. The transformations and deformations at both stages were driven by optimizing an objective function with two terms. One term computed the geometric typicality of the currently deformed m-rep via the probability distribution of transformations and deformations derived from PGA of the training segmentations. The second term computed a model-to-image match term based on the goodness of match of the trained intensity template for the currently deformed m-rep with the corresponding intensity data in the target image. Human and m-rep segmentations were compared using quantitative metrics provided in a toolset called VALMET. Metrics reported in this paper include 1) percent volume overlap; 2) mean surface distance between two segmentations; and 3) maximum surface separation (Hausdorff distance). Results:Averaged over all kidneys the the mean surface separation was 0.12 cm, t he mean Hausdorff distance was 0.99 cm, and the mean volume overlap for human segmentations was 88.8%. Between human and m-rep segmentations the mean surface separation was 0.18-0.19 cm, the mean Hausdorff distance was 1.14-1.25 cm, and the mean volume overlap was 82-83%,. Conclusion:Overall in this study the best m-rep kidney segmentations were at least as good as careful manual slice-by-slice segmentations, and the worst performance was no worse than humans in the typical clinical setting. The mean surface separations for human-m-rep segmentations were slightly larger that for humanhuman segmentations but still in the sub-voxel range, and volume overlap and maximum surface separation were slightly better for human-human comparisons. These results were expected because of experimental factors that favored comparison of the human-human segmentations. In particular m-rep agreement with humans appears to have been limited largely by fundamental differences between manual slice-by-slice and true three-dimensional seg...

show abstract

Consistency of three-dimensional planning target volumes across physicians and institutions

Cited by 66 publications

References 25 publications

CT slice index and thickness: Impact on organ contouring in radiation treatment planning for prostate cancer

CT slice index and thickness: Impact on organ contouring in radiation treatment planning for prostate cancer

Quality Assurance in Radiotherapy of Breast Cancer Variability in Planning Target Volume Delineation

Department of Radiation Oncology

Contact Info

Product

Resources

About