Computer-assisted spine surgery planning based on quantitative image analysis (J2-7118)

General information

Title
Computer-assisted spine surgery planning based on quantitative image analysis
Period
Jan 1, 2016 -- Dec 31, 2018
Range
1.63 FTE
Activity
2.06 - Engineering Sciences and Technologies / Systems and Cybernetics / Biomedical Tehnics

Abstract

Present-day sedentary lifestyle and problems with overweight and aging are reflected in an increased number of spine and spine-related diseases. Surgical interventions are the preferred treatment for patients with acute disorders or traumatic spine injuries, and are inevitable when other approaches, e.g. painkillers, physiotherapy or bracing, are not effective. Vertebral fixation is a surgical procedure for treating conditions such as scoliosis, kyphosis, spondylolisthesis and other degenerative conditions that cause instability of the spine. As spinal instability may cause damage to the spinal cord and nerve roots, the aim of vertebral fixation is to reduce vertebral mobility and, as a result, avoid such damage. The vertebral fixation procedure is based on anchoring two (or more) vertebrae to each other by metal fixation devices such as rods, plates and/or screws. One of the most widely used fixation techniques is pedicle screw placement, which consists of inserting screws through vertebral pedicles from the posterior side so that they reach the interior of the vertebral body, and then attaching a stabilizing rod to the exterior part of the screws on each side of the vertebra. As there is a limited visibility of anatomical structures during surgery, it is important to gain a mental conceptualization and reconstruction of the three-dimensional (3D) anatomy of spinal structures that are hidden from direct view. For a safe pedicle screw placement, the spine surgeon has to perform proper surgery planning by taking into account pedicle morphology (shape and structure), and choosing the appropriate size (diameter and length) and insertion trajectory (entry point and inclination) of each pedicle screw, which has proved valuable for reducing the risk of screw misplacement. As a result, preoperative surgery planning has become essential for pedicle screw placement, during which the surgeon on the basis of 3D spine images accurately studies the anatomy of the treated patient. Preoperative imaging is therefore necessary for reliable surgery planning, and computed tomography (CT) proved to be the imaging modality of choice for assessing the bone structures forming the spine and vertebrae. Modern computer software allows the surgeon to manually plan the size and insertion trajectory of each pedicle screw by navigating through 3D images and manipulating with 3D models of the spine, vertebrae and screws in 3D space. However, human capability of surgery planning by proper interpretation of medical images is limited due to our nonsystematic search patterns and natural biological variability of human anatomy, while the presence of noise may conceal actual geometrical relationships between anatomical structures. As a result, surgery planning is a relatively time-consuming task, and its reliability and accuracy depend on the subjective interpretation of the surgeon. Our research objective is to achieve a quantitative, objective and reproducible evaluation of vertebral morphology from 3D spine images, especially of the morphology of pedicles and vertebral bodies for the purpose of preoperative planning of vertebral fixation by pedicle screw placement procedures. For this purpose, we will develop new computer-assisted techniques that will be based on automated processing and quantitative analysis of 3D spine images, and as such contribute to an improved planning of pedicle screw placement procedures as well as of other spine surgical procedures that require an accurate evaluation of vertebral morphology. The topic of this research project is therefore to develop and validate new techniques for automated computer-assisted quantitative analysis of 3D images for spine surgery planning.

Researchers

Phases of the project and their realization

2016
Implementation of the 3D vertebral body model
Completed
Application of the 3D vertebral body model to CT images of normal and pathological vertebrae
Completed
Collection of vertebral body modeling results according to the proposed validation methodology
Completed
Exploring the possibilities for improving/upgrading the 3D vertebral body model and the application process
Completed
Design and development of the 3D pedicle model
Completed
Implementation of the 3D pedicle model
Completed
Application of the 3D pedicle model to CT images of normal and pathological vertebrae
Completed
Collection of pedicle modeling results according to the proposed validation methodology
Completed
Exploring the possibilities for improving/upgrading the 3D pedicle model and the application process
Completed
Dissemination of the results - publishing in journals with a high impact factor
Completed
2017
Final exact definition of the validation methodology for evaluating the results of vertebral body and pedicle modeling
Completed
Application of the 3D vertebral body model and 3D pedicle model to CT images of normal and pathological vertebrae
Completed
Collection of the final results obtained by parametric modeling of the vertebral body and pedicles in 3D according to the defined validation methodology
Completed
Evaluation of vertebral morphology in 3D (vertebral body and pedicles)
Completed
Comparison of existing and newly developed techniques for quantitative evaluation of vertebral morphology
Completed
Design and development of the 3D pedicle screw model
Completed
Application of automated pedicle screw planning by using the developed 3D pedicle screw model to CT images of normal and pathological vertebrae
Completed
Collection of automated pedicle screw planning results according to the proposed validation methodology
Completed
Exploring the possibilities for improving/upgrading the 3D pedicle screw model and the process of automated pedicle screw planning
Completed
Final exact definition of the validation methodology for evaluating the results of pedicle screw modelling and automated pedicle screw planning
Completed
Dissemination of the results - publishing in journals with a high impact factor
Completed
2018
Collection of preoperative CT spine images of patients that underwent surgery by pedicle screw placement with corresponding manually determined preoperative surgery plans
Completed
Application of the automated pedicle screw planning process by using the developed 3D pedicle model to selected CT spine images of patients that underwent surgery by pedicle screw placement
Completed
Collection of the final automated pedicle screw planning results according to the defined validation methodology
Completed
Comparison of preoperative plans for pedicle screw placement that were obtained manually and by the proposed automated pedicle screw planning process
Completed
Comparison of existing and newly developed techniques for automated pedicle screw planning
Completed
Exploring the possibilities for improving/upgrading the developed techniques for evaluation of vertebral morphology in 3D and automated pedicle screw planning
Completed
Dissemination of the results - publishing in journals with a high impact factor
Completed

Bibliographics records

1.
Dejan Knez, Tomaž Vrtovec: Computerized three-dimensional pedicle morphometry from computed tomography images of the thoracic spine. Image Analysis & Stereology, 39:101-109, 2020 [COBISS-ID:18613763 ] [doi:10.5566/ias.2349 ]
2.
Dejan Knez, Imad S. Nahle, Tomaž Vrtovec, Stefan Parent, Samuel Kadoury: Computer-assisted pedicle screw trajectory planning using CT-inferred bone density: a demonstration against surgical outcomes. Medical Physics, 46(8):3543-3554, 2019 [COBISS-ID:12506964 ] [doi:10.1002/mp.13585 ]
3.
Hugo Jobidon-Lavergne, Samuel Kadoury, Dejan Knez, Carl-Éric Aubin: Biomechanically driven intraoperative spine registration during navigated anterior vertebral body tethering. Physics in Medicine and Biology, 64(11):115008, 2019 [COBISS-ID:12506708 ] [doi:10.1088/1361-6560/ab1bfa ]
4.
William Mandel, Olivier Turcot, Dejan Knez, Stefan Parent, Samuel Kadoury: Prediction outcomes for anterior vertebral body growth modulation surgery from discriminant spatiotemporal manifolds. International Journal of Computer Assisted Radiology and Surgery, 14:1565-1575, 2019 [COBISS-ID:12598868 ] [doi:10.1007/s11548-019-02041-w ]
5.
Dejan Knez, Janez Mohar, Robert Cirman, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Variability analysis of manual and computer-assisted preoperative thoracic pedicle screw placement planning. Spine, 43(21):1487-1495, 2018 [COBISS-ID:11993172 ] [doi:10.1097/BRS.0000000000002659 ]
6.
Dejan Knez: Computer-assisted design and analysis of pedicle screw placement based on medical images of the spine. Doctoral Dissertation, University of Ljubljana, Faculty of Electrical Engineering (supervisor: T. Vrtovec), 2018 [COBISS-ID:12080212 ]
7.
William Mandel, Olivier Turcot, Dejan Knez, Stefan Parent, Samuel Kadoury: Spatiotemporal manifold prediction model for anterior vertebral body growth modulation surgery in idiopathic scoliosis. 21st International Conference on Medical Image Computing and Computer Assisted Intervention - MICCAI 2018, A.F. Frangi et al. (Eds.), Sep 16-20, Granada, Spain, Lecture Notes in Computer Science 11073:206-213, 2018 [COBISS-ID:12228180 ] [doi:10.1007/978-3-030-00937-3_24 ]
8.
Dejan Knez, Imad S. Nahle, Tomaž Vrtovec, Stefan Parent, Samuel Kadoury: Computer-assisted pedicle screw placement planning: towards clinical practice. 15th IEEE Symposium on Biomedical Imaging: From Nano to Macro - ISBI 2018, Apr 4-7, Washington, D.C., USA, A. Amini, S. Acton (Eds.), 249-252, 2018 [COBISS-ID:12005204 ] [doi:10.1109/ISBI.2018.8363566 ]
9.
Imad S. Nahle, Dejan Knez, Tomaž Vrtovec, Stefan Parent, Samuel Kadoury: Computer-assisted pedicle screw planning from pre-operative CT: a quantitative comparison with post-operative 3D resonstructions. 47th Annual Meeting of the Quebec Scoliosis Society, Oct 19-21, Baie-Saint-Paul, Quebec, Canada, 2017 [COBISS-ID:11869268 ]
10.
Dejan Knez, Janez Mohar, Robert Cirman, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Analysis of manual and computer-assisted preoperative pedicle screw placement planning. 17th Annual Meeting of the International Society for Computer Assisted Orthopaedic Surgery - CAOS 2017, Jun 14-17, Aachen, Germany, 2017 [COBISS-ID:11811412 ]
11.
Dejan Knez, Imad S. Nahle, Stefan Parent, Samuel Kadoury, Tomaž Vrtovec: Quantitative comparison of computer-assisted pedicle screw placement plans with actually inserted pedicle screws. Annual Spine Experts Group Meeting and Slovenia 2017 Spine Symposium, Nov 24-25, Ljubljana, Slovenia, 2017 [COBISS-ID:11907412 ]
12.
Janez Mohar, Dejan Knez, Robert Cirman, Rihard Trebše, Rene Mihalič, Tomaž Vrtovec: Comparison of automatically and manually obtained pedicle screw plans in deformities of the thoracic spine. Orthopaedic Proceedings, 99-B(Supp 5):11, 2017 [COBISS-ID:1539322564 ]
13.
Dejan Knez, Janez Mohar, Robert Cirman, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Analysis of manual and computer-assisted pre-operative pedicle screw placement planning. Orthopaedic Proceedings, 99-B(Supp 20):17, 2017 [COBISS-ID:11908180 ]
14.
Dejan Knez, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Computer-assisted screw size and insertion trajectory planning for pedicle screw placement surgery. IEEE Transactions on Medical Imaging, 35(6):1420-1430, 2016 [COBISS-ID:11250260 ] [doi:10.5566/ias.2349 ]
15.
Dejan Knez, Janez Mohar, Robert J. Cirman, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Determination of the pedicle screw size and trajectory in CT images of thoracic spinal deformities: a comparison between manual and computer-assisted preoperative planning. Zdravniški vestnik, 85(11-12):619-631, 2016 [COBISS-ID:11678804 ]
16.
Dejan Knez, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Automated determination of pedicle morphometry in the thoracic spine. 13th IEEE Symposium on Biomedical Imaging: From Nano to Macro - ISBI 2016, Apr 12-16, Prague, Czech Republic, J. Kybic, M. Sonka (Eds.), 664-667, 2016 [COBISS-ID:11351124 ] [doi:10.1109/ISBI.2016.7493354 ]
17.
Dejan Knez, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Automated pedicle screw size and trajectory planning by maximization of fastening strength. Computational Methods and Clinical Applications for Spine Imaging, Third International Workshop and Challenge, CSI 2015, held in conjunction with the 18th International Conference on Medical Image Computing and Computer Assisted Intervention - MICCAI 2015, Lecture Notes in Computer Science 9402:3-13, 2016 [COBISS-ID:11453012 ] [doi:10.1007/978-3-319-41827-8_1 ]
18.
Dejan Knez, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Avtomatska metoda načrtovanja velikosti in trajektorije pedikularnih vijakov. Računalniška obdelava slik in njena uporaba v Sloveniji - 11. strokovna konferenca - ROSUS 2016, Mar 17, Maribor, Slovenia, B. Potočnik (Ed.), 55-68, 2016 [COBISS-ID:11308372 ]
19.
Dejan Knez, Janez Mohar, Robert J. Cirman, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Manual and computer-assisted pedicle screw placement plans: a quantitative comparison. Proceedings of the 4th MICCAI Workshop & Challenge on Computational Methods and Clinical Applications for Spine Imaging - MICCAI-CSI2016, J. Yao et al. (Eds.), Oct 17, Athens, Greece, 78-88, 2016 [COBISS-ID:11578452 ]
20.
Dejan Knez, Janez Mohar, Robert J. Cirman, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Manual and computer-assisted pedicle screw placement plans: a quantitative comparison. Computational Methods and Clinical Applications for Spine Imaging, Fourth International Workshop and Challenge, CSI 2016, held in conjunction with the 19th International Conference on Medical Image Computing and Computer Assisted Intervention - MICCAI 2016, Lecture Notes in Computer Science 10182:105-115, 2016 [COBISS-ID:11708244 ] [doi:10.1007/978-3-319-55050-3_10 ]
21.
Dejan Knez, Janez Mohar, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Quantitative comparison between the straight-forward and anatomical insertion technique for pedicle screw placement. SPIE Medical Imaging 2016, Image Processing, Feb 27 - Mar 3, San Diego, USA, M.A. Styner, E.D. Angelini (Eds.), 9784:978439, 2016 [COBISS-ID:11316820 ] [doi:10.1117/12.2216028 ]
22.
Janez Mohar, Dejan Knez, Robert J. Cirman, Rihard Trebše, Rene Mihalič, Tomaž Vrtovec: Comparison of automatically and manually obtained pedicle screw plans in deformities of the thoracic spine. 29th Annual Congress of the International Society for Technology in Arthoplasty - ISTA 2016, Oct 5-8, Boston, USA, S.B. Murphy (Ed.), #4334, 2016 [COBISS-ID:11690836 ]
23.
Dejan Knez, Boštjan Likar, Franjo Pernuš, Tomaž Vrtovec: Automated pedicle screw size and trajectory planning by maximization of fastening strength. Proceedings of the 3rd MICCAI Workshop & Challenge on Computational Methods and Clinical Applications for Spine Imaging - MICCAI-CSI2015, T. Vrtovec et al. (Eds.), Oct 5, Munich, Germany, 3-13, 2015 [COBISS-ID:11182676 ]