Simultaneous segmentation and correspondence improvement using statistical modes Sinha a , Austin Reiter a , Simon Leonard a , Masaru Ishii b , Ayu yushi i Si Gregory D. Hager a , Russell H. Taylor a 14 th February, 2017 a Dept. of Computer Science, the Johns Hopkins University 1 b Dept. of Otolaryngology-Head and Neck Surgery, Johns Hopkins Medical Institutions Orlando, Florida
Functional Endoscopic Sinus Surgery Frontal sinus • What is it? Ethmoid sinuses • Minimally invasive procedure • Chronic sinusitis, nasal polyps, etc. • 600,000 procedures in the US per year [1] • 5-7% result in minor complications [2] Maxillary sinuses • ~ 1% result in major complications [2] WHY? Image from Toffelcenter.com [1] Bhattacharyya, N.: Ambulatory sinus and nasal surgery in the United States: Demographics and perioperative outcomes. The Laryngoscope. 120, 635-638 (2010) 2 [2] Dalziel, Kim; Stein, Ken; Round, Ali; Garside, Ruth; Royle, P.: Endoscopic sinus surgery for the excision of nasal polyps: A systematic review of safety and effectiveness. American Journal of Rhinology. 20(5), 506-519 (2006)
Sinuses & Nasal Airway: Complex structures with thin boundaries Fovea eth thmoidalis is: separates the ethmoid cells from the anterior Boundary between the cranial fossa sinuses and the orbit 0.5 mm [3] Thickness: ~ ~ 0.5 0.91 mm [4] Thickness: ~ ~ 0.91 [3] Kainz, J. and Stammberger , H., “The roof of the anterior ethmoid: A place of least resistance in the skull base,” American Journal of Rhinology 3(4), 191-199 (1989). 3 [4] Tao, H., Ma, Z., Dai, P., and Jiang, L., “Computer -aided three-dimensional reconstruction and measurement of the optic canal and intracanalicular structures ,” The Laryngoscope 109(9), 1499-1502 (1999).
Enhanced Endoscopic Navigation Labeled Template Pre-op CT Reg egis istration De Deformable (ICP [13] /IMLP [14] /IMLOP [15] / V-IMLOP [16] /etc.) Reg egis istration Str tructure fr from mot otion [13] P.J. Besl, H.D. McKay, A method for registration of 3-D shapes, IEEE Transactions on Pattern Analysis and Machine Intelligence, 1992 Intra-op video [14] Billings SD, Boctor EM, Taylor RH. Iterative Most-Likely Point Registration (IMLP): A Robust Algorithm for Computing Optimal Shape Alignment. PLOS ONE 10(3): e0117688, 2015 [15] Seth D. Billings, RH Taylor. Iterative Most Likely Oriented Point Registration. MICCAI, Boston, Proceedings, Part I. Vol. 8673: pp. 178-185, 2014 4 [16] Seth D. Billings, A Sinha, A. Reiter, S Leonard, M Ishii, GD Hager, RH Taylor. Anatomically constrained Video- CT registration via the V-IMLOP algorithm. MICCAI, Athens, Proceedings, Part III. Vol. 9902: pp. 133-141, 2016
Segmentation & Statistics Statistics Se Set t of of CTs 5
Our paper: Better segmentation & statistics Seg egmen entation Statis istic ics Mes esh Qualit ality Be Before Before Be Aft fter Be Before Aft fter Aft fter 6
Statistical Shape Model (SSM) [5] 𝑜 𝑡 𝑊 = 1 𝑊 1 𝑊 𝑗 𝑜 𝑡 𝑗=1 𝑊 2 𝑜 𝑡 Σ = 1 𝑗 − 𝑊 𝑈 (𝑊 𝑗 − 𝑊 𝑊) 𝑜 𝑡 𝑊 3 𝑗=1 𝜇 1 Σ = [𝑛 1 ⋯ 𝑛 𝑜 𝑡 ] 𝑛 1 ⋯ 𝑛 𝑜 𝑡 𝑈 ⋱ 𝜇 𝑜 𝑡 𝑊 𝑜 𝑡 [5] Cootes, T., Taylor, C., Cooper, D., and Graham, J., "Active shape models-their training and application," 7 Computer Vision and Image Understanding 61(1), 38-59 (1995).
Correspondence Improvement [8] Project shape onto the modes 𝑈 (𝑊 𝑗 − 𝑐 𝑗 = 𝑛 𝑗 𝑊) Compute estimate shape 𝑜 𝑡 𝑊 = 𝑊 + 𝑐 𝑗 𝑛 𝑗 𝑗=1 Move vertices of original shape along the surface toward the corresponding vertex on estimated shape [8] [8] Seshamani, S., Chintalapani, G., and Taylor, R., "Iterative refinement of point correspondences for 3D 8 statistical shape models," in Medical Image Computing and Computer-Assisted Intervention, 417-425 (2011).
Assumption • High accuracy segmentations • Segmentation improvement • E.g.: Using gradient vector flow (GVF) snakes [6][7] • Use gradient in corresponding CT image • Move mesh vertices toward structure boundaries • Correspondences between shapes • Lost during segmentation improvement [6] Xu, C. and Prince, J. L., "Gradient vector flow: A new external force for snakes," in Computer Vision and Pattern Recognition, IEEE Computer Society Conference on, 66-71 (1997). 9 [7] Xu, C. and Prince, J., “Snakes , shapes, and gradient vector ow," Image Processing, IEEE Transactions on, 7, 359- 369 (Mar 1998).
Simultaneous segmentation and correspondence improvement 10
Constrained segmentation improvement • Using GVF • Move vertices toward large gradients in image to obtain new surface, 𝜚 • Estimate 𝜚 using pre-existing SSM • Slide vertices on 𝜚 along the surface toward corresponding vertices on estimated shape 11
Simultaneous segmentation and correspondence improvement 𝑄 = 5 𝑅 = 3 5 iterations 12
Results From 52 publicly available CTs [9][10][11][12] [9] Bosch, W. R., Straube, W. L., Matthews, J. W., and Purdy, J. A., “Data from head-neck-cetuximab. The cancer imaging archive.," (2015). [10] Beichel, R. R., Ulrich, E. J., Bauer, C., Wahle, A., Brown, B., Chang, T., Plichta, K. A., Smith, B. J., Sunderland, J. J., Braun, T., Fedorov, A., Clunie, D., Onken, M., Riesmeier, J., Pieper, S., Kikinis, R., Graham, M. M., Casavant, T. L., Sonka, M., and Buatti, J. M., “Data from qin-headneck. The cancer imaging archive.," (2015). [11] Fedorov, A., Clunie, D., Ulrich, E., Bauer, C., Wahle, A., Brown, B., Onken, M., Riesmeier, J., Pieper, S., Kikinis, R., Buatti, J., and Beichel, R. R., “DICOM for quantitative imaging biomarker development: a standards based approach to sharing clinical data and structured pet/ct analysis results in head and neck cancer research," PeerJ 4(e2057) (2016). [12] Clark, K., Vendt, B., Smith, K., Freymann, J., Kirby, J., Koppel, P., Moore, S., Phillips, S., Matt, D., Pringle, M., 13 Tarbox, L., and Prior, F., \The cancer imaging archive (tcia): Maintaining and operating a public information repository," Journal of Digital Imaging 26(6), 1045{1057 (2013).
Results: Segmentation Red contour: Segmentation via label transfer using deformable registration Green contour: Improved Blue contour: segmentation Hand-labeled using our gold standard method 14
Results: Segmentation Segmentation errors compared against hand-segmented gold standard computed using the Hausdorff distance metric. Mea ean Err Error ± St Std. De Dev. (m (mm) Max Err Error (m (mm) Deformable registration 0.3327 ± 0.3147 2.338 GVF 0.1135 ± 0.1316 1.1548 GVF + SSM (our method) 0.09 0.0985 ± 0.12 0.128 1.0364 1.03 1 0.8 0.6 0.4 0.2 Truth De Deformable Reg egistr tratio ion GVF VF GVF VF+SSM (ou (our meth thod) 0 15
Results: Correspondence 16
Results: Mesh Quality Segmentation improved using our method Segmentation improved using GVF Triangle Quality 17
Conclusion • Our method improves segmentation while maintaining correspondences • Demonstrate improved segmentation and correspondence Our shape model contains more accurate information Our shape model is able to estimate a new shape accurately 18
Thank you! Facult lty Fundin ing Russ Taylor NIH R01R01-EB015530: Enhanced Navigation Greg Hager for Endoscopic Sinus Surgery through Video Masaru Ishii Analysis (PI: Hager) Austin Reiter Johns Hopkins University internal funds Simon Leonard Framework Rob Grupp cisst Developers 19
Recommend
More recommend
