Gianfranco Loi1, Marco Fusella2, Eleonora Lanzi3, Elisabetta Cagni4, Cristina Garibaldi5, Giuseppina Iacoviello6, Francesco Lucio7, Enrico Menghi8, Roberto Miceli9, Lucia C Orlandini10,11, Antonella Roggio2, Federica Rosica12, Michele Stasi13, Lidia Strigari14, Silvia Strolin14, Christian Fiandra15. 1. Department of Medical Physics, University Hospital "Maggiore della Carità", Novara, Italy. 2. Medical Physics Department, Veneto Institute of Oncology IOV IRCCS, Padua, Italy. 3. R&D Department, Tecnologie Avanzate, Turin, Italy. 4. Department of Medical Physics, S. Maria Nuova Hospital, Reggio Emilia, Italy. 5. Unit of Radiation Research, European Institute of Oncology, Milano, Italy. 6. ARNAS-Civico Hospital, Palermo, Italy. 7. Department of Medical Physics, "Santa Croce e Carle" Hospital, Cuneo, Italy. 8. Medical Physics Department, Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori (IRST) IRCCS, Meldola, FC, Italy. 9. Department of Diagnostic Imaging, Molecular Imaging, Interventional Radiology and Radiotherapy, Tor Vergata General Hospital, Rome, Italy. 10. Medical Physics Unit, Centro Oncologico Fiorentino, Firenze, Italy. 11. Radiation Oncology Department, Sichuan Cancer Hospital, Chengdu, China. 12. Department of Medical Physics, Ospedale Civile Giuseppe Mazzini, Teramo, Italy. 13. SC Fisica sanitaria, A.O. Ordine Mauriziano di Torino, Turin, Italy. 14. Laboratory of Medical Physics and Expert Systems, Regina Elena National Cancer Institute, Rome, Italy. 15. Department of Oncology, University of Turin, Turin, Italy.
Abstract
PURPOSE: To investigate the performance of various algorithms for deformable image registration (DIR) to propagate regions of interest (ROIs) using multiple commercial platforms. METHODS AND MATERIALS: Thirteen institutions participated in the study with six commercial platforms: RayStation (RaySearch Laboratories, Stockholm, Sweden), MIM (Cleveland, OH, USA), VelocityAI and Smart Adapt (Varian Medical Systems, Palo Alto, CA, USA), Mirada XD (Mirada Medical Ltd, Oxford, UK), and ABAS (Elekta AB, Stockholm, Sweden). The DIR algorithms were tested on synthetic images generated with the ImSimQA package (Oncology Systems Limited, Shrewsbury, UK) by applying two specific Deformation Vector Fields (DVF) to real patient data-sets. Head-and-neck (HN), thorax, and pelvis sites were included. The accuracy of the algorithms was assessed by comparing the DIR-mapped ROIs from each center with those of reference, using the Dice Similarity Coefficient (DSC) and Mean Distance to Conformity (MDC) metrics. Statistical inference on validation results was carried out in order to identify the prognostic factors of DIR performances. RESULTS: DVF intensity, anatomic site and participating center were significant prognostic factors of DIR performances. Sub-voxel accuracy was obtained in the HN by all algorithms. Large errors, with MDC ranging up to 6 mm, were observed in low-contrast regions that underwent significant deformation, such as in the pelvis, or large DVF with strong contrast, such as the clinical tumor volume (CTV) in the lung. Under these conditions, the hybrid DIR algorithms performed significantly better than the free-form intensity based algorithms and resulted robust against intercenter variability. CONCLUSIONS: The performances of the systems proved to be site specific, depending on the DVF type and the platforms and the procedures used at the various centers. The pelvis was the most challenging site for most of the algorithms, which failed to achieve sub-voxel accuracy. Improved reproducibility was observed among the centers using the same hybrid registration algorithm.
PURPOSE: To investigate the performance of various algorithms for deformable image registration (DIR) to propagate regions of interest (ROIs) using multiple commercial platforms. METHODS AND MATERIALS: Thirteen institutions participated in the study with six commercial platforms: RayStation (RaySearch Laboratories, Stockholm, Sweden), MIM (Cleveland, OH, USA), VelocityAI and Smart Adapt (Varian Medical Systems, Palo Alto, CA, USA), Mirada XD (Mirada Medical Ltd, Oxford, UK), and ABAS (Elekta AB, Stockholm, Sweden). The DIR algorithms were tested on synthetic images generated with the ImSimQA package (Oncology Systems Limited, Shrewsbury, UK) by applying two specific Deformation Vector Fields (DVF) to real patient data-sets. Head-and-neck (HN), thorax, and pelvis sites were included. The accuracy of the algorithms was assessed by comparing the DIR-mapped ROIs from each center with those of reference, using the Dice Similarity Coefficient (DSC) and Mean Distance to Conformity (MDC) metrics. Statistical inference on validation results was carried out in order to identify the prognostic factors of DIR performances. RESULTS: DVF intensity, anatomic site and participating center were significant prognostic factors of DIR performances. Sub-voxel accuracy was obtained in the HN by all algorithms. Large errors, with MDC ranging up to 6 mm, were observed in low-contrast regions that underwent significant deformation, such as in the pelvis, or large DVF with strong contrast, such as the clinical tumor volume (CTV) in the lung. Under these conditions, the hybrid DIR algorithms performed significantly better than the free-form intensity based algorithms and resulted robust against intercenter variability. CONCLUSIONS: The performances of the systems proved to be site specific, depending on the DVF type and the platforms and the procedures used at the various centers. The pelvis was the most challenging site for most of the algorithms, which failed to achieve sub-voxel accuracy. Improved reproducibility was observed among the centers using the same hybrid registration algorithm.
Authors: Molly M McCulloch; Brian M Anderson; Guillaume Cazoulat; Christine B Peterson; Abdallah S R Mohamed; Stefania Volpe; Hesham Elhalawani; Houda Bahig; Bastien Rigaud; Jason B King; Alexandra C Ford; Clifton D Fuller; Kristy K Brock Journal: Phys Med Biol Date: 2019-09-05 Impact factor: 3.609
Authors: Yunfei Hu; Mikel Byrne; Ben Archibald-Heeren; Kenton Thompson; Andrew Fong; Marcel Knesl; Amy Teh; Eve Tiong; Richard Foster; Paul Melnyk; Michelle Burr; Amelia Thompson; Jiy Lim; Luke Moore; Fiona Gordon; Rylie Humble; Anna Hardy; Saul Williams Journal: J Med Radiat Sci Date: 2019-10-28