Automatic segmentation of male pelvic anatomy on computed tomography images: a comparison with multiple observers in the context of a multicentre clinical trial

Radiation Oncology

Martin A Ebert¹  Garry Grogan²  John P Geraghty² 

[1] School of Physics, University of Western Australia, Perth, Western Australia, Australia;Department of Radiation Oncology, Sir Charles Gairdner Hospital, Perth, Western Australia, Australia

关键词: Quality assurance;    Automation;    Clinical trials;    Volume;    Segmentation;    Computed tomography;   

Others  :  1154049
DOI  :  10.1186/1748-717X-8-106

 received in 2013-03-12, accepted in 2013-04-19,  发布年份 2013

【 摘 要 】

Background

This study investigates the variation in segmentation of several pelvic anatomical structures on computed tomography (CT) between multiple observers and a commercial automatic segmentation method, in the context of quality assurance and evaluation during a multicentre clinical trial.

Methods

CT scans of two prostate cancer patients (‘benchmarking cases’), one high risk (HR) and one intermediate risk (IR), were sent to multiple radiotherapy centres for segmentation of prostate, rectum and bladder structures according to the TROG 03.04 “RADAR” trial protocol definitions. The same structures were automatically segmented using iPlan software for the same two patients, allowing structures defined by automatic segmentation to be quantitatively compared with those defined by multiple observers. A sample of twenty trial patient datasets were also used to automatically generate anatomical structures for quantitative comparison with structures defined by individual observers for the same datasets.

Results

There was considerable agreement amongst all observers and automatic segmentation of the benchmarking cases for bladder (mean spatial variations < 0.4 cm across the majority of image slices). Although there was some variation in interpretation of the superior-inferior (cranio-caudal) extent of rectum, human-observer contours were typically within a mean 0.6 cm of automatically-defined contours. Prostate structures were more consistent for the HR case than the IR case with all human observers segmenting a prostate with considerably more volume (mean +113.3%) than that automatically segmented. Similar results were seen across the twenty sample datasets, with disagreement between iPlan and observers dominant at the prostatic apex and superior part of the rectum, which is consistent with observations made during quality assurance reviews during the trial.

Conclusions

This study has demonstrated quantitative analysis for comparison of multi-observer segmentation studies. For automatic segmentation algorithms based on image-registration as in iPlan, it is apparent that agreement between observer and automatic segmentation will be a function of patient-specific image characteristics, particularly for anatomy with poor contrast definition. For this reason, it is suggested that automatic registration based on transformation of a single reference dataset adds a significant systematic bias to the resulting volumes and their use in the context of a multicentre trial should be carefully considered.

【 授权许可】

   
2013 Geraghty et al.; licensee BioMed Central Ltd.

【 图 表 】

Figure 1.

Figure 2.

【 参考文献 】

• [1]Hamilton CS, Ebert MA: Volumetric uncertainty in radiotherapy. Clin Oncol 2005, 17:456-464.
• [2]Jameson MG, Holloway LC, Vial PJ, Vinod SK, Metcalfe PE: A review of methods of analysis in contouring studies for radiation oncology. J Med Imag Radiat Oncol 2010, 54:401-410.
• [3]Matzinger O, Poortmans P, Giraud JY, Maingon P, Budiharto T, van den Bergh ACM, Davis JB, Musat E, Ataman F, Huyskens DP: Quality assurance in the 22991 EORTC ROG trial in localized prostate cancer: Dummy run and individual case review. Radioth Oncol 2009, 90:285-290.
• [4]Huyskens DP, Maingon P, Vanuytsel L, Remouchamps V, Roques T, Dubray B, Haas B, Kunz P, Coradi T, Buhlman R: A qualitative and a quantitative analysis of an auto-segmentation module for prostate cancer. Radioth Oncol 2009, 90:337-345.
• [5]Isambert A, Dhermain F, Bidault F, Commowick O, Bondiau PY, Malandain G, Lefkopoulos D: Evaluation of an atlas-based automatic segmentation software for the delineation of brain organs at risk in a radiation therapy clinical context. Radioth Oncol 2008, 87:93-99.
• [6]Sims R, Isambert A, Gregoire V, Bidault F, Fresco L, Sage J, Mills J, Bourhis J, Lefkopoulos D, Commowick O: A pre-clinical assessment of an atlas-based automatic segmentation tool for the head and neck. Radioth Oncol 2009, 93:474-478.
• [7]Choi HJ, Kim YS, Lee SH, Lee YS, Park G, Jung JH, Cho BC, Park SH, Ahn H, Kim C-S: Inter- and intra-observer variability in contouring of the prostate gland on planning computed tomography and cone beam computed tomography. Acta Oncol 2011, 50:539-546.
• [8]Fiorino C, Reni M, Bolognesi A, Cattaneo GM, Calandrino R: Intra- and inter-observer variability in contouring prostate and seminal vesicles: implications for conformal treatment planning. Radioth Oncol 1998, 47:285-292.
• [9]Rasch C, Barillot I, Remeijer P, Touw A, van Herk M, Lebesque JV: Definition of the prostate in CT and MRI: a multi-observer study. Int J Radiat Oncol Biol Phys 1999, 43:57-66.
• [10]Remeijer P, Rasch C, Lebesque JV, van Herk M: A general methodology for three-dimensional analysis of variation in target volume delineation. Med Phys 1999, 26:931-940.
• [11]Oliveira CM, Rodrigues PP: Automatic organ delineation of computed tomography images for radiotherapy planning in prostate cancer: An overview. Setubal: Scitepress; 2011.
• [12]TROG 03.04 - Randomised trial investigating the effect on survival and PSA control of different durations of adjuvant androgen deprivation in association with definitive radiation treatment for localised carcinoma of the prostate (RADAR). http://clinicaltrials.gov/ct2/show/NCT00193856 webcite
• [13]Kearvell R, Haworth A, Ebert MA, Murray J, Hooton B, Richardson S, Joseph DJ, Lamb D, Spry NA, Duchesne G, Denham JW: Quality improvements in prostate radiotherapy: Outcomes and impact of comprehensive quality assurance during the TROG 03.04 ‘RADAR’ trial. JMIRO 2013. In press
• [14]Baxter BS, Hitchner LE, Maguire GQ Jr: A standard format for digital image exchange. Madison WI: AAPM; 1982.
• [15]NEMA: Digital Imaging and Communications in Medicine (DICOM) Standard. In National Electrical Manufacturers Association. Edited by Association NEM. Washington DC: Office of Publications; 2001.
• [16]Blumhofer A, Achatz S, Braun R, Brett DJ: iPlan® Automatic Segmentation (Clinical White Paper). Brainlab AG; 2008.
• [17]Enke C, Soldberg TD: Clinical Validation of Automated Prostate Segmentation in iPlan Image. Omaha: University of Nebraska Medical Center; 2007.
• [18]Ebert MA, Haworth A, Kearvell R, Hooton B, Coleman R, Spry NA, Bydder S, Joseph DJ: Detailed review and analysis of complex radiotherapy clinical trial planning data: Evaluation and initial experience with the SWAN software system. Radioth Oncol 2008, 86:200-210.
• [19]Ebert MA, McDermott LN, Haworth A, van der Wath E, Hooton B: Tools to analyse and display variations in anatomical delineation. Aust Phys Eng Sci Med 2012, 35:159-164.
• [20]Dice LR: Measures of the amount of ecologic association between species. Ecology 1945, 26:297-302.
• [21]Chandra SS, Dowling JA, Shen K-K, Raniga P, Pluim JPW, Greer PB, Salvado O, Fripp J: Patient Specific Prostate Segmentation in 3-D Magnetic Resonance Images. IEEE Trans Med Imag 2012, 31:1955-1964.
• [22]Gao Y, Liao S, Shen D: Prostate segmentation by sparse representation based classification. Med Phys 2012, 39:6372-6387.
• [23]Liao S, Shen D: A Feature-Based Learning Framework for Accurate Prostate Localization in CT Images. IEEE Trans Imag Proc 2012, 21:3546-3559.