Advertisement

A markerless beam’s eye view tumor tracking algorithm based on unsupervised deformable registration learning framework of VoxelMorph in medical image with partial data

Published:December 18, 2022DOI:https://doi.org/10.1016/j.ejmp.2022.12.002

      Highlights

      • Propose a markerless BEV tumor radiotherapy tracking algorithm.
      • Requires no additional equipment; free of secondary trauma and additional dose.
      • Achieve the alignment of MLC occlusion images with template matching method.
      • End-to-end unsupervised framework for non-rigid registration of partial image data.
      • Processes inferior quality MV X-ray images with high noise level and low contrast.

      Abstract

      Purpose

      To propose an unsupervised deformable registration learning framework-based markerless beam's eye view (BEV) tumor tracking algorithm for the inferior quality megavolt (MV) images with occlusion and deformation.

      Methods

      Quality assurance (QA) plans for thorax phantom were delivered to the linear accelerator with artificially treatment offsets. Electronic portal imaging device (EPID) images (682 in total) and corresponding digitally reconstructed radiograph (DRR) were gathered as the moving and fixed image pairs, which were randomly divided into training and testing set in a ratio of 0.7:0.3 to train a non-rigid registration model with Voxelmorph. Simultaneously, 533 pairs of patient images from 21 lung tumor plans were acquired for tumor tracking investigation to offer quantifiable tumor motion data. Tracking error and image similarity measures were employed to evaluate the algorithm’s accuracy.

      Results

      The tracking algorithm can handle image registration with non-rigid deformation and losses ranging from 10 % to 80 %. The tracking errors in the phantom study were below 3 mm in about 86.8 % of cases, and below 2 mm in about 80 % of cases. The normalized mutual information (NMI) changes from 1.182 ± 0.024 to 1.198 ± 0.024 (p < 0.005). The patient target had an average translation of 3.784 mm and a maximum comprehensive displacement value of 7.455 mm. NMI of patient images changes from 1.209 ± 0.027 to 1.217 ± 0.026 (p < 0.005), indicating the presence of non-negligible non-rigid deformation.

      Conclusions

      The study provides a robust markerless tumor tracking algorithm for multi-modal, partial data and inferior quality image processing.

      Keywords

      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'

      Subscribe:

      Subscribe to Physica Medica: European Journal of Medical Physics
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect

      References

        • Boda-Heggemann J.
        • Lohr F.
        • Wenz F.
        • Flentje M.
        • Guckenberger M.
        KV cone-beam CT-based IGRT: a clinical review.
        Strahlenther Onkol. 2011; 187: 284-291https://doi.org/10.1007/s00066-011-2236-4
        • Vergalasova I.
        • Cai J.
        A modern review of the uncertainties in volumetric imaging of respiratory-induced target motion in lung radiotherapy.
        Med Phys. 2020; 47: e988-e1008https://doi.org/10.1002/mp.14312
        • Loi M.
        • Magallon-Baro A.
        • Suker M.
        • Eijck C.V.
        • Sharma A.
        • Hoogeman M.
        • et al.
        Pancreatic cancer treated with SBRT: effect of anatomical interfraction variations on dose to organs at risk.
        Radiother Oncol. 2019; 134: 67-73https://doi.org/10.1016/j.radonc.2019.01.020
        • Sande E.P.S.
        • Acosta Roa A.M.
        • Hellebust T.P.
        Dose deviations induced by respiratory motion for radiotherapy of lung tumors: Impact of CT reconstruction, plan complexity, and fraction size.
        J Appl Clin Med Phy. 2020; 21: 68-79https://doi.org/10.1002/acm2.12847
        • Herk M.V.
        Errors and margins in radiotherapy.
        Semin Radiat Oncol. 2013; 11: 52-64https://doi.org/10.1053/j.semradonc.2003.10.003
        • Shimohigashi Y.
        • Toya R.
        • Saito T.
        • Kono Y.
        • Doi Y.
        • Fukugawa Y.
        • et al.
        Impact of four-dimensional cone-beam computed tomography on target localization for gastric mucosa-associated lymphoid tissue lymphoma radiotherapy: reducing planning target volume.
        Radiat Oncol. 2021; 16: 14https://doi.org/10.1186/s13014-020-01734-w
        • Bezjak A.
        • Rumble R.B.
        • Rodrigues G.
        • Hope A.
        • Warde P.
        Intensity-modulated radiotherapy in the treatment of lung cancer.
        Clin Oncol-UK. 2012; 24: 508-520https://doi.org/10.1016/j.clon.2012.05.007
        • Morgenthaler J.
        • Köhler C.
        • Budach V.
        • Sehouli J.
        • Stromberger C.
        • Besserer A.
        • et al.
        Long-term results of robotic radiosurgery for non brachytherapy patients with cervical cancer.
        Strahlenther Onkol. 2021; 197: 474-486https://doi.org/10.1007/s00066-020-01685-x
        • Gustafsson C.
        • Korhonen J.
        • Persson E.
        • Gunnlaugsson A.
        • Nyholm T.
        • Olsson L.E.
        Registration free automatic identification of gold fiducial markers in MRI target delineation images for prostate radiotherapy.
        Med Phys. 2017; 44: 5563-5574https://doi.org/10.1002/mp.12516
        • Patel R.
        • Panfil J.
        • Campana M.
        • Block A.M.
        • Harkenrider M.M.
        • Surucu M.
        • et al.
        Markerless motion tracking of lung tumors using dual-energy fluoroscopy.
        Med Phys. 2015; 42: 254-262https://doi.org/10.1118/1.4887903
        • Bhagat N.
        • Fidelman N.
        • Durack J.C.
        • Collins J.
        • Gordon R.L.
        • LaBerge J.M.
        • et al.
        Complications associated with the percutaneous insertion of fiducial markers in the thorax.
        Cardiovasc Inter Rad. 2010; 33: 1186-1191https://doi.org/10.1007/s00270-010-9949-0
        • Seides B.J.
        • Egan Ill J.P.
        • French K.D.
        • Kovitz K.L.
        • Desai N.R.
        Fiducial marker placement for stereotactic body radiation therapy via convex probe endobronchial ultrasound: a case series and review of literature.
        J Thorac Dis. 2018; 10: 1972-1983
        • Roeske J.C.
        • Mostafavi H.
        • Haytmyradov M.
        • Wang A.
        • Morf D.
        • Cortesi L.
        • et al.
        Characterization of markerless tumor tracking using the on-board imager of a commercial linear accelerator equipped with fast-kV switching dual-energy imaging.
        Adv Radiat Oncol. 2020; 5: 1006-1013https://doi.org/10.1016/j.adro.2020.01.008
        • SaKata Y.
        • Hirai R.
        • Kobuna K.
        • Tanizawa A.
        • Mori S.
        A machine learning-based real-time tumor tracking system for fluoroscopic gating of lung radiotherapy.
        Phys Med Biol. 2020; 65085014https://doi.org/10.1088/1361-6560/ab79c5
      1. Winston JP, Best K, Plusquellic L, Thoma P, Gilley DB. Patient exposure and dose guide - 2003. 1st ed. Frankfort: Conference of Radiation Control Program Directors, Inc; 2003.

        • Murphy M.J.
        • Balter J.
        • Balter S.
        • BenComo Jr, J.A.
        • Das I.J.
        • Jiang S.B.
        • et al.
        The management of imaging dose during image-guided radiotherapy: report of the AAPM Task Group 75.
        Med Phys. 2007; 34: 4041-4063https://doi.org/10.1118/1.2775667
        • Fang R.
        • Yang J.Z.
        • Du W.L.
        • Court L.
        Automatic detection of graticule isocenter and scale from kV and MV images.
        J Appl Clin Med Phys. 2019; 20: 18-28https://doi.org/10.1002/acm2.12558
        • Chen L.
        • Bai S.
        • Li G.J.
        • Li Z.B.
        • Xiao Q.
        • Bai L.
        • et al.
        Accuracy of real-time respiratory motion tracking and time delay of gating radiotherapy based on optical surface imaging technique.
        Radiat Oncol. 2020; 15: 170https://doi.org/10.1186/s13014-020-01611-6
        • Naumann P.
        • Batista V.
        • Farnia B.
        • Fischer J.
        • Liermann J.
        • Tonndorf-Martini E.
        • et al.
        Feasibility of optical surface-guidance for position verification and monitoring of stereotactic body radiotherapy in deep-inspiration breath-hold.
        Front Oncol. 2020; 10573279https://doi.org/10.3389/fonc.2020.573279
        • Seppenwoolde Y.
        • Berbeco R.I.
        • Nishioka S.
        • Shirato H.
        • Heijmen B.
        Accuracy of tumor motion compensation algorithm from a robotic respiratory tracking system: a simulation study.
        Med Phys. 2007; 34: 2774-2784https://doi.org/10.1118/1.2739811
        • Nakamura M.
        • Sawada A.
        • Mukumoto N.
        • Takahashi K.
        • Mizowaki T.
        • Kokubo M.
        • et al.
        Effect of audio instruction on tracking errors using a four-dimensional image-guided radiotherapy system.
        J Appl Clin Med Phys. 2013; 14: 255-264https://doi.org/10.1120/jacmp.v14i5.4488
        • Li W.Z.
        • Liang Z.W.
        • Cao Y.
        • Cao T.T.
        • Quan H.
        • Yang Z.Y.
        • et al.
        Estimating intrafraction tumor motion during fiducial-based liver stereotactic radiotherapy via an iterative closest point (ICP) algorithm.
        Radia Oncol. 2019; 14: 185https://doi.org/10.1186/s13014-019-1401-2
        • Moutrie V.
        • Kairn T.
        • Rosenfeld A.
        • Charles P.H.
        Use of a megavoltage electronic portal imaging device to identify prosthetic materials.
        Australas Phys Eng Sci Med. 2015; 38: 93-100https://doi.org/10.1007/s13246-015-0327-8
        • Arimura H.
        • Egashira Y.
        • Shioyama Y.
        • Nakamura K.
        • Yoshidome S.
        • Anai S.
        • et al.
        Computerized method for estimation of the location of a lung tumor on EPID cine images without implanted markers in stereotactic body radiotherapy.
        Phys Med Biol. 2009; 54: 665-677https://doi.org/10.1088/0031-9155/54/3/013
        • Rottmann J.
        • Aristophanous M.
        • Chen A.
        • Court L.
        • Berbeco R.
        A multi-region algorithm for markerless beam's-eye view lung tumor tracking.
        Phys Med Biol. 2010; 55: 5585-5598https://doi.org/10.1088/0031-9155/55/18/021
        • Mesin L.
        Balanced multi-image demons for non-rigid registration of magnetic resonance images.
        Magn Reson Imaging. 2020; 74: 128-138https://doi.org/10.1016/j.mri.2020.09.013
        • Deo R.C.
        Machine learning in medicine.
        Circulation. 2015; 132: 1920-1930https://doi.org/10.1161/CIRCULATIONAHA.115.001593
        • Avanzo M.
        • Porzio M.
        • Lorenzon L.
        • Milan L.
        • Sghedoni R.
        • Russo G.
        • et al.
        Artificial intelligence applications in medical imaging: A review of the medical physics research in Italy.
        Phys Med. 2021; 83: 221-241https://doi.org/10.1016/j.ejmp.2021.04.010
        • Zanca F.
        • Hernandez-Giron I.
        • Avanzo M.
        • Guidi G.
        • Crijns W.
        • Diaz O.
        • et al.
        Expanding the medical physicist curricular and professional programme to include Artificial Intelligence.
        Phys Med. 2021; 83: 174-183https://doi.org/10.1016/j.ejmp.2021.01.069
        • Zhao Q.
        • Ma Y.K.
        • Zhu C.
        • Yao C.F.
        • Feng B.L.
        • Dai F.
        Image stitching via deep homography estimation.
        Neurocomputing. 2021; 450: 219-229https://doi.org/10.1016/j.neucom.2021.03.099
        • Papadomanolaki M.
        • Christodoulidis S.
        • Karantzalos K.
        • Vakalopoulou M.
        Unsupervised multistep deformable registration of remote sensing imagery based on deep learning.
        Remote Sens (Basel). 2021; 13: 1294https://doi.org/10.3390/rs13071294
        • Yang X.
        • Kwitt R.
        • Styner M.
        • Niethammer M.
        Quicksilver: Fast predictive image registration - a deep learning approach.
        Neuroimage. 2017; 158: 378-396https://doi.org/10.1016/j.neuroimage.2017.07.008
        • Wolfs C.J.A.
        • Canters R.A.M.
        • Verhaegen F.
        Identification of treatment error types for lung cancer patients using convolutional neural networks and EPID dosimetry.
        Radiother Oncol. 2020; 153: 243-249https://doi.org/10.1016/j.radonc.2020.09.048
        • Nie X.
        • Li G.
        Real-time 2D MR cine from beam eye's view with tumor-volume projection to ensure beam-to-tumor conformality for MR-guided radiotherapy of lung cancer.
        Front Oncol. 2022; 12898771https://doi.org/10.3389/fonc.2022.898771
        • Balakrishnan G.
        • Zhao A.
        • Sabuncu M.R.
        • Guttag J.
        • Dalca A.V.
        VoxelMorph: A learning framework for deformable medical image registration.
        IEEE Trans Med Imaging. 2019; 38: 1788-1800https://doi.org/10.1109/TMI.2019.2897538
        • Fanizzi A.
        • Basile T.M.A.
        • Losurdo L.
        • Bellotti R.
        • Bottigli U.
        • Dentamaro R.
        • et al.
        A machine learning approach on multiscale texture analysis for breast microcalcification diagnosis.
        BMC Bioinf. 2020; 21: 91https://doi.org/10.1186/s12859-020-3358-4
        • Belmont B.
        • Kessler R.
        • Theyyunni N.
        • Fung C.
        • Huang R.
        • Cover M.
        • et al.
        Continuous inferior vena cava diameter tracking through an iterative kanade–lucas–tomasi-based algorithm.
        Ultrasound Med Bio. 2018; 44: 2793-2801https://doi.org/10.1016/j.ultrasmedbio.2018.07.022
        • Lee H.
        • Cho A.
        • Lee S.
        • Whang M.
        Vision-Based measurement of heart rate from ballistocardiographic head movements using unsupervised clustering.
        Sensors. 2019; 19: 3263https://doi.org/10.3390/s19153263
        • Shieh C.C.
        • Caillet V.
        • Dunbar M.
        • Keall P.J.
        • Booth J.T.
        • Hardcastle N.
        • et al.
        A bayesian approach for three-dimensional markerless tumor tracking using kV imaging during lung radiotherapy.
        Phys Med Biol. 2017; 62: 3065-3080https://doi.org/10.1088/1361-6560/aa6393
        • Hazelaar C.
        • Dahele M.
        • Mostafavi H.
        • Weide L.V.D.
        • Slotman B.
        • Verbakel W.
        Markerless positional verification using template matching and triangulation of kV images acquired during irradiation for lung tumors treated in breath-hold.
        Phys Med Biol. 2018; 63115005https://doi.org/10.1088/1361-6560/aac1a9
        • Liu G.
        • Hu F.L.
        • Ding X.F.
        • Li X.Q.
        • Shao Q.H.
        • Wang Y.N.
        • et al.
        Simulation of dosimetry impact of 4DCT uncertainty in 4D dose calculation for lung SBRT.
        Radiat Oncol. 2019; 14: 1https://doi.org/10.1186/s13014-018-1191-y
        • Yang T.J.
        • Tang Q.
        • Li L.
        • Song J.K.
        • Zhu C.H.
        • Tang L.
        Nonrigid registration of medical image based on adaptive local structure tensor and normalized mutual information.
        J Appl Clin Med Phys. 2019; 20: 99-110https://doi.org/10.1002/acm2.12612
        • Wu J.Z.
        • Lei P.
        • Shekhar R.
        • Li H.L.
        • Suntharalingam M.
        • D’Souza W.D.
        Do tumors in the lung deform during normal respiration? An image registration investigation.
        Int J Radiat Oncol Biol Phys. 2009; 75: 268-275https://doi.org/10.1016/j.ijrobp.2009.03.008