Automatic volume delineation in oncological PET

Summary

Aim: Evaluation of a dedicated software tool for automatic delineation of 3D regions of interest in oncological PET. Patients, methods: The applied procedure encompasses segmentation of user-specified subvolumes within the tomographic data set into separate 3D ROIs, automatic background determination, and local adaptive thresholding of the background corrected data. Background correction and adaptive thresholding are combined in an iterative algorithm. Nine experienced observers used this algorithm for automatic delineation of a total of 37 ROIs in 14 patients. Additionally, the observers delineated the same ROIs also manually (using a freely chosen threshold for each ROI) and the results of automatic and manual ROI delineation were compared. Results: For the investigated 37 ROIs the manual delineation shows a strong interobserver variability of (26.8 ± 6.3)% (range: 15% to 45%) while the corresponding value for automatic delineation is (1.1 ± 1.0)% (range: <0.1% to 3.6%). The fractional deviation of the automatic volumes from the observer-averaged manual ones is (3.7 ± 12.7)%. Conclusion: The evaluated software provides results in very good agreement with observer-averaged manual evaluations, facilitates and accelerates the volumetric evaluation, eliminates the problem of interobserver variability and appears to be a useful tool for volumetric evaluation of oncological PET in clinical routine.

Zusammenfassung

Ziel: Evaluierung eines für die automatische Abgrenzung von 3D-ROIs bestimmten Verfahrens in der onkologischen PET. Patienten, Methoden: Das Verfahren umfasst die Segmentierung benutzerdefinierter Teilvolumina innerhalb des tomographischen Datensatzes in getrennte 3D-ROIs, automatische Untergrundbestimmung und die Anwendung eines lokaladaptiven Schwellwertverfahrens auf die untergrundkorrigierten Bilddaten. Untergrundkorrektur und Schwellwertanalyse sind hierbei Teil eines iterativen Algorithmus. Neun erfahrene Personen nutzten diesen Algorithmus zur automatischen Abgrenzung von 37 ROIs bei 14 Patienten. Zusätzlich grenzte jeder Nutzer die gleichen Strukturen manuell ab (mit frei wählbarem Schwellwert für jede ROI). Die Resultate der automatischen und manuellen ROI-Abgrenzung wurden verglichen. Resultate: Für die untersuchten 37 ROIs zeigt die manuelle Auswertung eine starke Interobserver- Variabilität von (26.8 ± 6.3)% (Spannbreite: 15% bis 45%), der entsprechende Wert für die automatische Abgrenzung beträgt (1.1 ± 1.0)% (Spannbreite: 0.1% bis 3.6%). Die prozentuale Abweichung der automatisch bestimmten Volumina von den manuell bestimmten ist (3.7 ± 12.7)%. Schlussfolgerung: Die untersuchte Software liefert Resultate, die sehr gut mit der nutzergemittelten manuellen Auswertung überein stimmen. Sie erleichtert und beschleunigt die volumetrische Auswertung, eliminiert das Problem der Interobserver- Variabilität und scheint ein nützliches Werkzeug für die volumetrische Auswertung der onkologischen PET in der klinischen Routine zu sein.

Keywords

3D ROI quantification - PET - volume delineation - background correction - interobserver variability

Schlüsselwörter

3D-ROI - quantitative PET - Volumenabgrenzung - Untergrundkorrektur - Interobservervariabilität

• References

• 1 Apostolova I, Renisch S, Opfer R. et al. FDG PET/CT in cancer therapy monitoring. Computer-assisted analysis of baseline together with up to two follow-ups. Nuklearmedizin 2010; 50: 83-92.
• 2 Aristophanous M, Penney BC, Martel MK, Pelizzar CA. A Gaussian mixture model for definition of lung tumor volumes in positron emission tomography. Med Phys 2007; 34: 4223-4235.
• 3 Belhassen S, Zaidi H. A novel fuzzy C-means algorithm for unsupervised heterogeneous tumor quantification in PET. Med Phys 2010; 37: 1309-1324.
• 4 Black QC, Grills IS, Kestin LL. et al. Defining a radiotherapy target with positron emission tomography. Int J Radiat Oncol Biol Phys 2004; 60: 1272-1282.
• 5 Boellaard R, Krak NC, Hoekstra OS, Lammertsma AA. Effects of noise, image resolution, and ROI definition on the accuracy of standard uptake values: a simulation study. J Nucl Med 2004; 45: 1519-1527.
• 6 Ciernik IF, Huser M, Burger C. et al. Automated functional image-guided radiation treatment planning for rectal cancer. Int J Radiat Oncol Biol Phys 2005; 62: 893-900.
• 7 Daisne J, Sibomana M, Bol A. et al. Tri-dimensional automatic segmentation of PET volumes based on measured source-to-background ratios: influence of reconstruction algorithms. Radiother Oncol 2003; 69: 247-250.
• 8 Denecke T, Rau B, Hoffmann KT. et al. Comparison of CT, MRI and FDG-PET in response prediction of patients with locally advanced rectal cancer after multimodal preoperative therapy: Is there a benefit in using functional imaging?. Eur Radiol 2005; 15: 1658-1666.
• 9 Drever L, Robinson DM, McEwan A, Roa W. A local contrast based approach to threshold segmentation for PET target volume delineation. Med Phys 2006; 33: 1583.
• 10 Erasmus JJ, Gladish GW, Broemeling L. et al. Interobserver and intraobserver variability in measurement of non–small-cell carcinoma lung lesions: Implications for assessment of tumor response. J Clin Oncol 2003; 21: 2574.
• 11 Erdi YE, Mawlawi O, Larson SM. etal. Segmentation of lung lesion volume by adaptive positron emission tomography image thresholding. Cancer 1997; 80 (Suppl. 12) 2505-2509.
• 12 Frings V, deLangen AJ, Smit EF. etal. Repeatability of metabolically active volume measurements with ¹⁸F-FDG and ¹⁸F-FLT PET in non-small cell lung cancer. J Nucl Med 2010; 51: 1870.
• 13 Geets X, Lee JA, Bol A. et al. A gradient-based method for segmenting FDG-PET images: methodology and validation. Eur J Nucl Med Mol Imaging 2007; 34: 1427-1438.
• 14 Hatt M, Lamare F, Boussion N. et al. Fuzzy hidden Markov chains segmentation for volume determination and quantitation in PET. Phys Med Biol 2007; 52: 3467-3491.
• 15 Hatt M, Visvikis D, Albarghach NM. et al. Prognostic value of ¹⁸F-FDG PET image-based parameters in oesophageal cancer and impact of tumour delineation methodology. Eur J Nucl Med Mol Imaging 2011; 38: 1-12.
• 16 Hatt M, Visvikis D, LeRest CC. Autocontouring Versus Manual Contouring. J Nucl Med 2011; 52: 658.
• 17 Hoffman E, Huang S, Phelps M. Quantification in positron emission computed tomography. 1. Effect of object size. J Comp Assist Tomogr 1979; 3: 299-308.
• 18 Hofheinz F, Dittrich S, Pötzsch C, vanden Hoff J. Effects of cold sphere walls in PET phantom measurements on the volume reproducing threshold. Phys Med Biol 2010; 55: 1099-1113.
• 19 Jentzen W, Freudenberg L, Eising EG. et al. Segmentation of PET volumes by iterative image thresholding. J Nucl Med 2007; 48: 108-114.
• 20 Kessler R, Ellis J, Eden M. Analysis of emission tomographic scan data: limitations imposed by resolution and background. J Comp Assist Tomogr 1984; 8: 514-522.
• 21 Kim J, Cai W, Eberl S D F. Real-time volume rendering visualization of dual-modality PET/CT images with interactive fuzzy thresholding segmentation. IEEE Trans Inf Technol Biomed 2007; 11: 16.
• 22 Krohn T, Kaiser HJ, Gagel B. et al. 3D volume and SUV analysis of oncological PET studies: a voxel-based image processing tool with NSCLC as example. Nuklearmedizin 2007; 46: 141-148.
• 23 Leunens G, Menten J, Weltens C. et al. Quality assessment of medical decision making in radiation oncology: variability in target volume delineation for brain tumours. Radiother Oncol 1993; 29: 169-175.
• 24 Li H, Thorstad WL, Biehl KJ. et al. A novel PET tumor delineation method based on adaptive region-growing and dual-front active contours. Med Phys 2008; 35: 3711-3721.
• 25 Nestle U, Kremp S, Schaefer-Schuler A. etal. Comparison of different methods for delineation of ¹⁸F-FDG PET-positive tissue for target volume definition in radiotherapy of patients with non-small cell lung cancer. J Nucl Med 2005; 46: 1342-1348.
• 26 Philips Imalytics;. http://www.imalytics.philips.com
• 27 Riegel AC, Berson AM, Destian S. et al. Variability of gross tumor volume delineation in head-and-neck cancer using CT and PET/CT fusion. Int J Radiat Oncol Biol Phys 2006; 65: 726-732.
• 28 ROVER. ROVER: ROI visualization, evaluation and image registration. ABX Radeberg. http://www.abx.de/rover; 2008
• 29 Van Baardwijk A, Baumert BG, Bosmans G. et al. The current status of FDG-PET in tumour volume definition in radiotherapy treatment planning. Cancer Treat Rev 2006; 32: 245-260.
• 30 Van Baardwijk A, Bosmans G, Boersma L. et al. PET-CT-based auto-contouring in non-small-cell lung cancer correlates with pathology and reduces interobserver variability in the delineation of the primary tumor and involved nodal volumes. Int J Radiat Oncol Biol Phys 2007; 68: 771-778.
• 31 Van Dalen JA, Hoffmann AL, Dicken V. et al. A novel iterative method for lesion delineation and volumetric quantification with FDG PET. Nucl Med Commun 2007; 28: 485-493.
• 32 Vauclin S, Doyeux K, Hapdey S. et al. Development of a generic thresholding algorithm for the delineation of ¹⁸FDG-PET-positive tissue: application to the comparison of three thresholding models. Phys Med Biol 2009; 54: 6901-6916.
• 33 Wu K, Ung YC, Hwang D. etal. Autocontouring and manual contouring: Which is the better method for target delineation using ¹⁸F-FDG PET/CT in non-small cell lung cancer?. J Nucl Med 2010; 51: 1517.
• 34 Yu H, Caldwell C, Mah K. et al. Automated radiation targeting in head-and-neck cancer using region-based texture analysis of PET and CT images. Int J Radiat Oncol Biol Phys 2009; 75: 618-625.