Radiation Physics

PV QA 3 - Poster Viewing Q&A 3

TU_16_3272 - Evaluating the Accuracy of Commercial Deformable Image Registration Software for Real Patient Images Using Anthropomorphic Modeling

Tuesday, October 23
1:00 PM - 2:30 PM
Location: Innovation Hub, Exhibit Hall 3

Evaluating the Accuracy of Commercial Deformable Image Registration Software for Real Patient Images Using Anthropomorphic Modeling
B. Guo1, L. Qiu1, J. D. Donaghue2, S. H. Hsu3, and P. Xia1; 1Department of Radiation Oncology, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, 2Moll Cancer Center, Fairview Hospital, Cleveland Clinic, Cleveland, OH, 3Radiation Oncology, Montefiore Medical Center, BRONX, NY

Purpose/Objective(s): Deformable image registration (DIR) is widely used nowadays to register images acquired throughout the radiation oncology treatment course for target delineation, dose accumulation, and adaptive planning. However, the accuracy of commercial DIR software remains questionable and there lacks a “ground truth” to benchmark DIR software for real patient images. We introduced a new method based on anthropomorphic modeling to provide accurate DIR for head and neck patients and evaluated the accuracy of commercial DIR software.

Materials/Methods: Four head and neck patients from the Cancer Imaging Archive, Collection Head-Neck Cetuximab were selected to demonstrate the concept. Each patient had two PETCTs and a planning CT taken with the patient at different positions. Two patients had arms up in PETCT and down in planning CT. The bony structures in each CT were automatically segmented into individual bone groups (e.g. each vertebra) using image processing, clustering, and pattern matching. Model-based DIR rigidly registers each bone group using Iterative Closest Point algorithm then deforms soft tissue with the bones using finite element method. Model-based DIR can be considered as the "truth" of DIR at the bones. The accuracy of rigid registrations and four commercial DIR software were evaluated.

Results: The mean absolute errors of DIR for multiple rigid registration (the best of three rigid registrations aligning to base of skull, mid-neck, and lower neck/shoulder respectively), DIR software A (constrained, intensity based, free-form algorithm), B (hybrid algorithm designed for head and neck), C (multi-resolution B-Spline algorithm) and D (hybrid algorithm using image intensity and region of interest constraints) were listed in Table 1. Table 1. Mean absolute errors of multiple rigid registration and four DIR software for head and neck patient images
Mean absolute error (cm) Head Jaw Spine Clavicle Scapulae Humerus (Arms up/down) Humerus (Arm down) Sternum Ribs
Multiple rigid registrations 0.26 0.51 0.72 1.89 2.14 7.92 2.43 0.61 0.77
Software A 0.57 0.25 0.61 1.59 1.50 6.55 1.58 0.35 0.65
Software B 0.64 0.30 0.63 1.16 1.37 6.83 0.92 0.30 0.78
Software C 0.43 0.35 0.85 1.47 1.51 5.22 1.17 0.56 1.07
Software D 0.66 0.32 0.74 1.31 1.88 5.84 1.68 0.45 0.85

Conclusion: Anthropomorphic modeling provides the truth of DIR for bones in head and neck patient images which can be used to measure the accuracy of commercial DIR software. For the patients studied, the mean DIR errors were on the order of millimeters except for large posture change (arms up/down). The performance of DIR software was comparable with each other and with multiple rigid registrations despite the difference in the algorithm used. There is not an algorithm that outperforms others in all regions.

Author Disclosure: B. Guo: None. J.D. Donaghue: None. S. Hsu: Research Grant; Varian Medical Systems. P. Xia: Employee; Cleveland Clinic. Research Grant; Philips Healthcare.

Send Email for Bingqi Guo


Assets

TU_16_3272 - Evaluating the Accuracy of Commercial Deformable Image Registration Software for Real Patient Images Using Anthropomorphic Modeling



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Evaluating the Accuracy of Commercial Deformable Image Registration Software for Real Patient Images Using Anthropomorphic Modeling