[go: up one dir, main page]

WO2025199473A1 - Procédé et système d'évaluation de précision de positionnement de vis pédiculaire - Google Patents

Procédé et système d'évaluation de précision de positionnement de vis pédiculaire

Info

Publication number
WO2025199473A1
WO2025199473A1 PCT/US2025/020972 US2025020972W WO2025199473A1 WO 2025199473 A1 WO2025199473 A1 WO 2025199473A1 US 2025020972 W US2025020972 W US 2025020972W WO 2025199473 A1 WO2025199473 A1 WO 2025199473A1
Authority
WO
WIPO (PCT)
Prior art keywords
pedicle screw
patient
image data
intraoperative
postoperative
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/US2025/020972
Other languages
English (en)
Inventor
Benjamin GROISSER
Roger Widmann
Howard Hillstrom
Ankush THAKUR
M. Timothy HRESKO
Matthew Cunningham
Jessica HEYER
John Blanco
Doug MINTZ
Ryan BREIGHNER
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
New York Society for Relief of Ruptured and Crippled
Original Assignee
New York Society for Relief of Ruptured and Crippled
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by New York Society for Relief of Ruptured and Crippled filed Critical New York Society for Relief of Ruptured and Crippled
Publication of WO2025199473A1 publication Critical patent/WO2025199473A1/fr
Pending legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B17/00—Surgical instruments, devices or methods
    • A61B17/14—Surgical saws
    • A61B17/15—Guides therefor
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B17/00—Surgical instruments, devices or methods
    • A61B17/16—Instruments for performing osteoclasis; Drills or chisels for bones; Trepans
    • A61B17/17—Guides or aligning means for drills, mills, pins or wires
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B17/00—Surgical instruments, devices or methods
    • A61B17/16—Instruments for performing osteoclasis; Drills or chisels for bones; Trepans
    • A61B17/17—Guides or aligning means for drills, mills, pins or wires
    • A61B17/1739—Guides or aligning means for drills, mills, pins or wires specially adapted for particular parts of the body
    • A61B17/1757—Guides or aligning means for drills, mills, pins or wires specially adapted for particular parts of the body for the spine
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B17/00—Surgical instruments, devices or methods
    • A61B17/56—Surgical instruments or methods for treatment of bones or joints; Devices specially adapted therefor
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
    • A61B2034/101—Computer-aided simulation of surgical operations
    • A61B2034/102—Modelling of surgical devices, implants or prosthesis
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
    • A61B2034/101—Computer-aided simulation of surgical operations
    • A61B2034/105—Modelling of the patient, e.g. for ligaments or bones
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/10—Computer-aided planning, simulation or modelling of surgical operations
    • A61B2034/107—Visualisation of planned trajectories or target regions
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B34/00—Computer-aided surgery; Manipulators or robots specially adapted for use in surgery
    • A61B34/20—Surgical navigation systems; Devices for tracking or guiding surgical instruments, e.g. for frameless stereotaxis
    • A61B2034/2046—Tracking techniques
    • A61B2034/2065—Tracking using image or pattern recognition
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B90/00—Instruments, implements or accessories specially adapted for surgery or diagnosis and not covered by any of the groups A61B1/00 - A61B50/00, e.g. for luxation treatment or for protecting wound edges
    • A61B90/36—Image-producing devices or illumination devices not otherwise provided for
    • A61B90/37—Surgical systems with images on a monitor during operation
    • A61B2090/374—NMR or MRI
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B90/00—Instruments, implements or accessories specially adapted for surgery or diagnosis and not covered by any of the groups A61B1/00 - A61B50/00, e.g. for luxation treatment or for protecting wound edges
    • A61B90/36—Image-producing devices or illumination devices not otherwise provided for
    • A61B90/37—Surgical systems with images on a monitor during operation
    • A61B2090/376—Surgical systems with images on a monitor during operation using X-rays, e.g. fluoroscopy
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B90/00—Instruments, implements or accessories specially adapted for surgery or diagnosis and not covered by any of the groups A61B1/00 - A61B50/00, e.g. for luxation treatment or for protecting wound edges
    • A61B90/36—Image-producing devices or illumination devices not otherwise provided for
    • A61B90/37—Surgical systems with images on a monitor during operation
    • A61B2090/376—Surgical systems with images on a monitor during operation using X-rays, e.g. fluoroscopy
    • A61B2090/3762—Surgical systems with images on a monitor during operation using X-rays, e.g. fluoroscopy using computed tomography systems [CT]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B90/00—Instruments, implements or accessories specially adapted for surgery or diagnosis and not covered by any of the groups A61B1/00 - A61B50/00, e.g. for luxation treatment or for protecting wound edges
    • A61B90/36—Image-producing devices or illumination devices not otherwise provided for
    • A61B90/37—Surgical systems with images on a monitor during operation
    • A61B2090/376—Surgical systems with images on a monitor during operation using X-rays, e.g. fluoroscopy
    • A61B2090/3762—Surgical systems with images on a monitor during operation using X-rays, e.g. fluoroscopy using computed tomography systems [CT]
    • A61B2090/3764—Surgical systems with images on a monitor during operation using X-rays, e.g. fluoroscopy using computed tomography systems [CT] with a rotating C-arm having a cone beam emitting source
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30008—Bone
    • G06T2207/30012—Spine; Backbone

Definitions

  • the present invention relates to a method and system for assessing the accuracy of vertebral pedicle screw placement.
  • Pedicle screws and other bone screws are the primary bone anchors for spine hardware constructs used in spine fusion and spine deformity correction surgery. Insertion of bone anchors in a safe manner is a primary goal of spine surgery.
  • Conventional methods for determining pedicle screw accuracy and pedicle screw breach include visual analysis. Such conventional methods are fraught with systematic biases as well as observer biases that affect the reliability of the data. Further, during conventional methods accuracy and breach cannot be determined until after surgery, rather than during surgery where inaccurate placement may be corrected.
  • Well-documented complications associated with bone anchor placement include neurological complications, vascular complications, and complications associated with pulmonary (pleura, lung, trachea) and gastrointestinal (esophagus) systems. Most of these complications relate to errors in the three-dimensional (3D) placement of the drill, tap or screw.
  • the system may be an orthopedic surgical system (e.g., automated and/or semi-automated orthopedic surgical system).
  • the automated orthopedic surgical system may include a processor configured to receive preoperative 3D image data of a planned pedicle screw placement in a patient’s vertebra; obtain intraoperative or postoperative 3D image data of a pedicle screw placement in the patient’s vertebra; determine an accuracy measurement of the pedicle screw placement in the patient’s vertebra based on the preoperative 3D image data and the intraoperative and/or postoperative 3D image data.
  • the accuracy measurement is based on comparing a planned pedicle screw tip position and intraoperative or postoperative 3D image data of a pedicle screw tip; a planned pedicle screw midportion position and intraoperative or postoperative 3D image data of a pedicle screw midportion; and/or a planned pedicle screw tail position and intraoperative or postoperative 3D image data of a pedicle screw tail.
  • the extra-vertebral anatomic structure can be a nerve, an artery, a vein, or a visceral organ such as, for example, a spinal cord, a phrenic nerve, a vagus nerve, an aorta, a vena cava, an esophagus, a lung, a trachea, and a bronchus.
  • a method for assessing pedicle screw placement in a patient includes receiving a preoperative 3D image of a vertebrae of a patient; determining a desired trajectory and position for insertion of the pedicle screw into the vertebrae of the patient; inserting the pedicle screw into the vertebral pedicle of the patient; receiving at least one of an intraoperative 3D image or postoperative 3D image of the vertebrae of the patient having the pedicle screw inserted therein; and determining an accuracy measurement of a pedicle screw position in the patient’s vertebra based on the desired trajectory and position for insertion of the pedicle screw into the vertebrae of the patient and the at least one intraoperative 3D image or postoperative 3D image.
  • a method for inserting a vertebral pedicle screw in a patient in need thereof.
  • the method includes obtaining a preoperative 3D image of a vertebral pedicle in a patient in need thereof; planning a desired path for insertion of the screw into the patient’s vertebral pedicle; inserting a vertebral pedicle screw into the patient’s vertebral pedicle; obtaining an intraoperative or postoperative 3D image of the vertebral pedicle having the pedicle screw inserted therein; and determining, using a computer, an accuracy measurement of the pedicle screw compared to the planned desired path for insertion of the pedicle screw into the patient’s vertebral pedicle based on the preoperative 3D image and the intraoperative or postoperative 3D image.
  • the method includes providing an accuracy measurement based on a distance from a pedicle screw to an extra- vertebral anatomic structure of the patient obtained from the intraoperative or postoperative 3D image.
  • An orthopedic surgical system comprising a processor, characterized by the processor being configured to: receive preoperative 3D image data of a planned pedicle screw placement in a patient’s vertebra; receive at least one of an intraoperative 3D image data of the pedicle screw placement in the patient’s vertebra or a postoperative 3D image data of the pedicle screw placement in the patient’s vertebra; and determine an accuracy measurement of a pedicle screw position in the patient’s vertebra based on the preoperative 3D image data and the at least one of the intraoperative 3D image data or the postoperative 3D image data.
  • the region of interest comprises at least one of a nerve, an artery, a vein, a visceral organ, a spinal cord, a phrenic nerve, a vagus nerve, an aorta, a vena cava, an esophagus, a lung, a trachea, or a bronchus.
  • a method for assessing pedicle screw placement in a patient characterized by: receiving, via a processor, a preoperative 3D image of a vertebrae of a patient; determining a desired trajectory and position for insertion of the pedicle screw into the vertebrae of the patient; inserting the pedicle screw into the vertebral pedicle of the patient; receiving at least one of an intraoperative 3D image or a postoperative 3D image of the vertebrae of the patient having the pedicle screw inserted therein; and determining an accuracy measurement of a pedicle screw position in the patient’s vertebra based on the desired trajectory and position for insertion of the pedicle screw into the vertebrae of the patient and the at least one of the intraoperative 3D image or the postoperative 3D image.
  • the region of interest comprises at least one of a nerve, an artery, a vein, a visceral organ, a spinal cord, a phrenic nerve, a vagus nerve, an aorta, a vena cava, an esophagus, a lung, a trachea, or a bronchus.
  • the accuracy measurement of the pedicle screw placement in the patient’s vertebra is based on comparing at least one of a group consisting of a planned pedicle screw tip position and at least one of an intraoperative 3D image data or a postoperative 3D image data of a pedicle screw tip; a planned pedicle screw midportion position and at least one of an intraoperative 3D image data or a postoperative 3D image data of a pedicle screw midportion; and a planned pedicle screw tail position and at least one of an intraoperative 3D image data or a postoperative 3D image data of a pedicle screw tail.
  • FIGS. 1 A, 1 B illustrate example vertebrae with pedicle screws in place
  • FIG. 2 is a flowchart illustrating a registration method of an orthopedic surgical system in accordance with an exemplary embodiment of the subject disclosure
  • FIG. 3 is a flowchart illustrating an accuracy measurement method of an orthopedic surgical system in accordance with an exemplary embodiment of the subject disclosure
  • FIG. 4 is a flowchart illustrating an accuracy measurement method of an orthopedic surgical system in accordance with an exemplary embodiment of the subject disclosure
  • FIG. 5 is a flowchart illustrating a method of assessing pedicle screw position accuracy in accordance with an exemplary embodiment of the subject disclosure.
  • FIG. 6 is a schematic diagram of an orthopedic surgical system in accordance with an exemplary embodiment of the subject disclosure.
  • “Substantially” as used herein shall mean considerable in extent, largely but not wholly that which is specified, or an appropriate variation therefrom as is acceptable within the field of art. “Exemplary” as used herein shall mean serving as a non-limiting example.
  • Pre-Op 3D Imaging is an image or volumetric image obtained preoperatively and containing at least multiple levels of the vertebral column (spine) involved in a surgery.
  • Example Pre-Op 3D Imaging include a CT scan or an MRI scan.
  • Pre-Op Plan is a preoperative plan of screw trajectories defined in reference to the Pre-Op 3D imaging.
  • Pre-Op Plans can include the dimensions (shaft length, shaft diameter, etc.) and intended position of each screw.
  • Imaging is an intraoperative image or volumetric image containing at least multiple levels of the spine involved in a surgery. Examples include a CT scan, MRI scan, a fluoroscopy scan, 3D cone-beam CT, flat-panel X- ray detectors, and OEC 3D (GE Healthcare).
  • a “Fitted Screw” is a screw position stored as xyz coordinates of a screw tip and tail in a local coordinate system. Because screw structure is known a priori, only position is required for the screw detection technique of the present disclosure.
  • “Final Alignment” is a rigid-body transformation for each bone stored as a 4x4 homogeneous transformation matrix with orthonormal rotational components (six degrees of freedom: three rotational and 3 translational).
  • “Translation/Rotation Error” is an accuracy defined in reference to the planned screw position in a screw-axis aligned coordinate system. After aligning the preoperative coordinate system with the postoperative volume, errors are computed as the translational and rotational offset between the preoperative planned screws and the postoperative Fitted Screws.
  • Safety Measurements and “Accuracy Measurements” according to the present disclosure may refer to one or more of the following: deviation of screw placement from the preoperative plan, the presence of medial pedicle breach, the maximal medial pedicle breach distance, the minimal distance to the pedicle cortical bone for non-breached screws, and the distance to extra-vertebral soft tissues.
  • Spinal fusion surgery is commonly used to treat degenerative disc disease.
  • Spinal fusion typically involves distracting and/or decompressing one or more intervertebral spaces, removing any associated facet joints or discs, and joining or fusing two or more adjacent vertebra together.
  • Fusion of vertebral bodies may involve fixation of two or more adjacent vertebrae, which may be accomplished through introduction of rods or plates, and screws (e.g., pedicle screws) or other devices into a vertebral joint to join various portions of a vertebra to a corresponding portion on an adjacent vertebra. Fusion may occur in the lumbar, thoracic or cervical spine region of a patient.
  • FIG. 6 illustrates a block diagram of a system 600 used to assess the accuracy of vertebral pedicle screw placement in accordance with an exemplary embodiment of the subject disclosure
  • the system 600 may be referred to as an orthopedic surgical system.
  • the system 600 includes a processor 604 or other logic control, a memory 606, an input device 602, and a display 608 or other output.
  • the processor 604 can include a memory, such as memory 606 or can be operatively connected to the memory 606.
  • the memory 606 can be volatile memory, persistent memory, and the like.
  • Memory in the processor 604 can be allocated dynamically according to variables, variable states, static objects, and permissions associated with objects and variables in the system. Such memory allocation can be based on instructions stored in the memory 606. Memory resources can be conserved relative to other systems that do not associate variables and other objects with permission data for the specific device.
  • the processor 604 can generate an output based on an input. For example, the processor 604 can receive an electronic and/or digital signal. The processor 604 can read the signal and perform one or more tasks with the signal, such as performing various functions with data in response to input received by the processing device.
  • the processor 604 can read information from the memory 606 that is needed to perform one or more functions. For example, the processor 604 can update a variable from static to dynamic based on a received input and a rule stored as data on the memory 606. The processor 604 can send an output signal to the memory 606, and the memory 606 can store data according to the signal output by the processor 604.
  • the processor 604 is configured to receive preoperative 3D image data of a planned pedicle screw placement in a patient’s vertebra.
  • the processor 604 is configured to obtain intraoperative or postoperative 3D image data of a pedicle screw placement in the patient’s vertebra.
  • the processor 604 is configured to determine an accuracy measurement of the pedicle screw placement in the patient’s vertebra based on the preoperative 3D image data and one or more of the intraoperative or postoperative 3D image data.
  • the accuracy measurement of the pedicle screw placement in the patient’s vertebra can be based on comparing a planned pedicle screw tip position and at least one of an intraoperative 3D image data and/or a postoperative 3D image data of a pedicle screw tip.
  • the accuracy measurement of the pedicle screw placement in the patient’s vertebra can be based on comparing a planned pedicle screw midportion position and at least one of an intraoperative 3D image data and/or a postoperative 3D image data of a pedicle screw midportion.
  • the accuracy measurement of the pedicle screw placement in the patient’s vertebra can be based on comparing a planned pedicle screw tail position and at least one of an intraoperative 3D image data and/or a postoperative 3D image data of a pedicle screw tail.
  • the accuracy measurement of the pedicle screw placement in the patient’s vertebra can be based on comparing a planned pedicle screw position of any one or more portions of the pedicle screw and at least one of an intraoperative 3D image data and/or a postoperative 3D image data of similar portion(s) of a pedicle screw.
  • the processor 604 can be configured to determine if a pedicle screw breach is present within a patient.
  • a pedicle screw breach can occur when a pedicle screw is malpositioned.
  • a pedicle screw breach can weaken the stability of the spine of the patient and damage nerves of the patient.
  • a preoperative 3D CT (or MRI) image of a patient may be obtained and visualization/planning techniques can be used to plan pedicle screw placement, for example, to create a preoperative plan for pedicle screw placement.
  • Screw position may be planned to avoid medial, superior and inferior breaches. The planned 3D screw position can be transferred into a robotic planning environment.
  • a pedicle screw can be inserted in the patient during surgery according to the preoperative plan.
  • Registration of an image can be performed.
  • a registration can include rotating and/or translating one 3D image until the image matches and/or aligns (e.g., best matches and/or aligns) with another image.
  • a registration can be performed that aligns one or more preoperative images, intraoperative images, and/or postoperative images.
  • a scan can be moved relative to another scan until the anatomy matches.
  • the preoperative scan can be considered an anchor or target position
  • the postoperative (or intraoperative) scan can be moved until the bony anatomy lines up on top of the corresponding preoperative anatomy.
  • a pedicle screw breach may be calculated using 3D image data of the outer circumference of the screw implant and the pedicle walls.
  • a distance from the pedicle screw to the cortical bone of the patient’s vertebra may be determined.
  • the position of the pedicle screw and the position of the cortical bone can be identified via intraoperative imaging and/or pre-operative images. Determining a distance from the pedicle screw to a cortical bone of the patient’s vertebra may be used to identify whether a pedicle screw breach is present.
  • the distance from the screw to the medial-inferior pedicle wall may be calculated using the 3D outer circumference of the detected screw (e.g., approximated as a cylinder) and the 3D envelope of the pedicle wall as detected by the segmentation step. Pre-Op CT may be used by the segmentation step.
  • the anatomic location of the relevant section of the pedicle wall may be determined. The anatomic location of the relevant section of the pedicle wall may include the most-breached point (for breached screws) or the point closest to being breached (for non-breached screws).
  • the accuracy measurement may be a distance from a pedicle screw to an extra-vertebral anatomic structure of the patient.
  • a preoperative 3D CT (or MRI) image may be obtained including at least a vertebra that will receive a pedicle screw and at least one extra-vertebral anatomic structure (e.g., a nerve, an artery, a vein, or a visceral organ such as a spinal cord, a phrenic nerve, a vagus nerve, an aorta, a vena cava, an esophagus, a lung, a trachea, and a bronchus).
  • a nerve, an artery, a vein, or a visceral organ such as a spinal cord, a phrenic nerve, a vagus nerve, an aorta, a vena cava, an esophagus, a lung, a trachea, and a bronchus.
  • Screw position is planned to avoid medial, superior and inferior breach.
  • the planned 3D screw position is transferred into a robotic planning environment.
  • a pedicle screw is inserted in the patient during surgery according to the preoperative plan.
  • An intraoperative 3D scan e.g., a CT scan, an MRI scan, a 3D cone-beam CT, flat-panel X-ray detectors, OEC 3D (GE Healthcare), 3D fluoroscopy (O-Arm, FE, Ziehm, etc.)
  • O-Arm FE, Ziehm, etc.
  • Segmentation techniques are applied to the preoperative image to determine the envelope of each vertebra having an inserted screw.
  • the segmentation techniques may be based on neural networks.
  • segmentation techniques based on neural networks may use convolutional layers.
  • the segmentation techniques may be trained on large datasets of spine scans that may be manually labeled and/or verified. Training techniques may include providing one or more algorithms a desired label map, such as each voxel being labeled with the correct value.
  • the training techniques may consist of adjusting model weights until predictions match the ground truth.
  • a segmentation can be performed by providing a new image as input to the model. In such embodiment, predicted labels can be produced as an output of the model.
  • the accuracy measurement of a pedicle screw position may be based on geometric offsets of the pedicle screw based on data obtained from the preoperative 3D image data and the intraoperative or postoperative 3D image data.
  • a preoperative 3D imaging technology suitable for the present disclosure include a CT scan, and an MRI scan.
  • Non-limiting examples of an intraoperative/postoperative 3D imaging technology suitable for the present disclosure include a CT scan, an MRI scan, a 3D cone-beam CT, flat-panel X-ray detectors, OEC 3D (GE Healthcare), and 3D fluoroscopy (O- Arm, FE, Ziehm, etc.).
  • Accuracy can be defined in reference to the planned screw position, in a screw-axis aligned coordinate system. After aligning the coordinate space of the postoperative image with the coordinate space of the preoperative image, accuracy or errors can be computed as an amount of translational and/or rotational offset between the preoperative planned screw trajectory and the surgically placed screw position.
  • the accuracy measurement is based on a position of a pedicle wall of the patient’s vertebra obtained from the preoperative 3D image data and a position of the pedicle screw relative to the position of the pedicle wall obtained from the intraoperative or postoperative 3D image data.
  • the accuracy measurement is based on a position of a medial wall the vertebral pedicle obtained from the preoperative 3D image data and a position of a medial aspect of a pedicle screw obtained from the intraoperative or postoperative 3D image data.
  • the processor 604 may be configured to detect one or more implants (e.g., other implants) in the patient. For example, when inserting a screw posteriorly, the processor 604 is configured to detect screws and/or other implants that may have been implanted within the patient via a different approach, such as via an anterior approach or from a past surgery. In such embodiments the processor 604 is configured to determine the accuracy measurement of the pedicle screw placement based on other implants detected within the patient via preoperative images.
  • implants e.g., other implants
  • the processor 604 may be configured to detect one or more anatomical characteristics of the patient, such as the spinal cord of the patient.
  • the processor 604 may detect the anatomical characteristics of the patient based on received 3D image data.
  • the processor 604 may be configured to determine placement accuracy of the pedicle screws relative to the spinal cord of the patient.
  • the processor 604 may be configured to determine placement of pedicle screws based on a combination of two or more of preoperative 3D image data of the patient, intraoperative 3D image data of the patient, and/or postoperative 3D image data of the patient.
  • the processor 604 may be configured to determine placement of pedicle screws based other devices detected within the patient, anatomical characteristics of the patient, and the like.
  • FIGS. 2, 3, and 4 illustrate methods of determining an accuracy measurement of a pedicle screw placement in a patient’s vertebra according to exemplary embodiments.
  • the steps shown in FIGS. 2, 3, and 4 may be shown in the order illustrated in the respective figures, or any other order useful for determining an accuracy measurement of a pedicle screw placement.
  • FIG. 2 shows a registration method 200 according to an exemplary embodiment of the present invention.
  • Registration method 200 includes a segmentation step 204.
  • Segmentation step 204 includes an input of a Pre-Op 3D Imaging 202 and an output of Binary Masks 206.
  • the Segmentation step 204 can be a discrete, modular element.
  • One or more segmentation techniques can be used with the Segmentation step 204.
  • the Segmentation step 204 can be treated as a black box that can receive raw CT as input and provide voxel-wise labels of one or more (e.g., all) of the vertebrae.
  • Pre-Op 3D imaging 202 can include volumetric image data containing at least all levels of the spine of the patient involved in the surgery.
  • Binary mask 206 can include voxel-wise segmentation masks that can be generated for one or more (e.g., each) bones for which screws are planned. The mask can correspond to the anatomical envelope of the bone of the patient and/or can overlap (e.g., largely overlap) with the rigid structure of the relevant bone of the patient.
  • a binary mask and vertebral segmentation can be used interchangeably herein.
  • One or more (e.g., each) vertebra can be labeled with a binary mask.
  • Voxels corresponding to the bone can be designated with a value (e.g., 1 ) and all other voxels can be designated with a different value (e.g., 0).
  • the binary mask can result in a 3D shape of the patient’s vertebra being labeled.
  • the binary mask i.e., vertebral segmentation can be performed via one or more (e.g., two) steps.
  • Step one includes a masking during registration (e.g., alignment) between preoperative and postoperative images.
  • the masking performed in step one allows the algorithm to consider (e.g., only consider) the bone voxels when optimizing alignment.
  • Step two includes computing a breach distance. In step two, the distance from the detected surgical screw position to the envelope of the segmentation can be computed as an accuracy measurement and/or safety measurement.
  • Registration method 200 can include Screw Detection step 220.
  • Screw Detection step 220 can include input Post-Op CT 218 and Pre-Op Plan 208 as well as output Fitted Screws 222.
  • Post-Op CT 218 can include Volumetric image data containing at least all levels of the spine of the patient involved in the surgery.
  • PreOp Plan 208 can include planned screw trajectories e.g., screw trajectories inputted by a user such as a spine surgeon. Plans can include the dimensions (shaft length, shaft diameter, etc.) and intended position and insertion trajectory of each screw.
  • one or more positions of screws implanted within a patient can be determined via an intraoperative 3D image and/or a postoperative 3D image.
  • the positions of the screws may include the x, y, and z coordinates of the screws.
  • a headless screw can be identified as a single cylinder, although other types of screws (e.g., polyaxial screws) can be identified as two or more cylinders (e.g., one cylinder for the shaft of the screw and another cylinder for the head of the screw) that can rotate relative to one other about a fixed pivot point.
  • the position of the geometric model may be optimized until the position aligns with the CT data of the screw, which may be provided via an intraoperative 3D image and/or a postoperative 3D image.
  • the optimization may reposition the model to maximize the high-intensity (e.g., metal) voxels inside the model, which can result in a Fitted Screw model.
  • the position of the Fitted Screw model can be used as a detected screw position.
  • Screw positions may be stored as X-Y-Z coordinates of screw tip and tail in the local coordinate system (e.g., rotation about the screw axis may not be detectable or relevant).
  • Screw detection step 220 may be robust to imaging noise and extraneous metal implants and may include one or more of the following steps.
  • the preoperative volume may be converted to a point cloud and threshold to keep only high-intensity voxels, corresponding to metal implants.
  • the preoperative planned screw trajectories may be aligned (e.g., rigidly aligned) to the thresholded point cloud using standard Iterative Closest Point (ICP) techniques. Randomly initialized alignments may account for sensitivity to initialization, keeping the best solution.
  • the output of this step can include a single affine transformation mapping the preoperative plan to postoperative image space.
  • a parametric, articulated form of ICP may be performed to improve the alignment between preoperative plan and thresholded point cloud.
  • the spine may be modeled as a series of coordinate systems corresponding to the segmented vertebrae of the preoperative spine. Techniques may allow neighboring vertebrae to move relative to each other to account for discrepancies in spine position between pre- and postoperative scans, but may penalize large deviations to enforce a prior that the spine should have approximately the same size and shape.
  • the output of this step is an affine transformation for each vertebra, mapping from the preoperative plan to postoperative image space.
  • screw positions may be optimized.
  • the dimensions and the approximate position from the matched cluster can be known.
  • Screws may be modeled as a cylinder for the shaft and (if relevant) a cylinder for the head (for polyaxial screws these cylinders can rotate relative to each other).
  • the goal of the optimization step is to align the screw model (cylinders) with the data (point cloud).
  • the screw heads that had previously been masked out are included in the point cloud.
  • a RANSAC search can be used to fit an initial line for the screw shaft, then an open-source nonlinear solver is employed to optimize the screw position to maximize the number of points in the point cloud that are inside the envelope of the screw model.
  • Initialization step 210 can include inputs of Pre-Op Plan 208 and Fitted Screws 222.
  • the Initialization step 210 can include an output of initial alignment 212.
  • a goal of the initialization step 210 can include obtaining an initial estimate of the alignment between preoperative and intraoperative 3D imaging (e.g., CT) volumes for each bone having a screw in it. Voxel intensities corresponding to cortical bone from pre- and postoperative volumes are converted to point clouds for alignment.
  • preoperative and intraoperative 3D imaging e.g., CT
  • Each vertebra is allowed to rotate and translate relative to its neighbors in order to align the high-intensity voxels from the Pre-Op volume with the high- intensity voxels of the intraoperative volume by minimizing a cost function with three terms: data term, rotational regularization, and translational regularization.
  • Translational regularization can be defined herein as differences in distances between neighboring vertebrae.
  • the intervertebral distances in the registration should match intervertebral distances in the preop segmentations.
  • Data term can be used herein as a distance between aligned pre- and intraoperative point cloud.
  • Rotational regularization can be used herein as a difference in rotations of neighboring vertebrae.
  • the output of the Initialization step 210 is a rigid-body homogeneous transformation matrix for each vertebra of the patient.
  • the matrix maps the preoperative vertebrae onto the intraoperative vertebrae.
  • the regularization terms reduce the effect of outliers by enforcing consistency in the relative positions of neighboring vertebrae in the preoperative and intraoperative scans, while still allowing for some articulated motion.
  • An Optimization step 214 can be performed. Inputs to the Optimization step 214 can include Pre-Op CT 202, Post-Op CT 218 (and/or an Intra-Op CT), Binary Mask 206, andjnitial Alignment 212.
  • the Optimization step 214 can include an output Final Alignment 216.
  • Volumetric registration is a known problem in medical imaging. With initialization, such as the Initiation step 210, alignment can be reliably performed with a multi-scale rigid body optimization routine. An open source implementation that maximizes Mutual Information between the two volumes being registered can be used. Such techniques can also match between heterogeneous image contrast (e.g., CT to MRI or MRI-T1 to MRI-T2).
  • image contrast e.g., CT to MRI or MRI-T1 to MRI-T2
  • FIG. 3 shows an accuracy method 300 according to an exemplary embodiment of the present invention.
  • the accuracy method 300 may include inputs Pre-Op Plan 302, Fitted Screws 304, and Final Alignment 308, which may be substantially the same as Pre-Op Plan 208, Fitted Screws 222, and Final Alignment 216 described herein.
  • Accuracy method 300 may include output Translation/Rotation Error 310.
  • the Final Alignment 308 may be used to bring the Pre-Op Plan 302 and the Fitted Screws 304 into a shared coordinate system. Translation and rotation errors 310 may be based on geometric offsets.
  • Translation/Rotational Error 310 in a screw-axis aligned coordinate system, accuracy may be defined in reference to the planned screw position. After aligning the postoperative Fitted Screws 304 with the preoperative plan 302, errors can be computed as the translational and rotational offset.
  • the Template matching 404 may include an input Binary Mask 402 and an output Labeled Surface 406. Safety measurements and/or accuracy measurements may require an understanding of anatomical structure, since measurements can be made in reference to specific anatomical regions or landmarks.
  • the volumetric voxelized Binary Mask 402 can be converted to a triangulated mesh via one or more techniques (e.g., the Marching Cubes technique). The conversion to the triangulated mesh can result in a manifold, watertight surface in 3D space representing the vertebral envelope.
  • Automated labeling of surface regions may be performed by finding pointwise correspondence with a template surface model. For each vertebral level, a generic surface model can be prepared with anatomical regions manually annotated. Upon correspondence with the template, the regions can be mapped onto the surface mesh under analysis.
  • Correspondence between the segmentation and the template can be found using an open source implementation of the Functional Maps framework.
  • a goal can include minimizing geodesic distortion between the template surface and the corresponding points on the target surface. For example, points that are close together on the template can remain close together in the target.
  • This method can be applied to surfaces as well as other domains (e.g. gridded volumes, tetrahedral meshes, etc.).
  • Functional Maps can be used for the As Rigid As Possible alignment problem that is invariant to rotation/translation of the two surfaces to be aligned.
  • the Breach Detection step 410 can include inputs such as the Labeled Surface 406, the Fitted Screws 408, and the Final Alignment 414.
  • the Breach Detection step 410 can include the output Safety Measurement 412 (i.e., accuracy measurement).
  • the breach distance is computed as the maximal distance from any point on the screw that protrudes into the spinal canal to any coplanar point on the intersected pedicle surface.
  • Coplanar points are points that are in the same coronal plane. For example, breach distances are always perpendicular to the screw axis and reported as positive values.
  • the approach (e.g., automated approach) of the present disclosure can be used with one or more (e.g., any) 3D data sets including 3D Fluoroscopy (O-Arm, GE, Ziehm, etc.), CT scan, and/or MRL
  • 3D Fluoroscopy O-Arm, GE, Ziehm, etc.
  • CT scan CT scan
  • MRL MRL
  • the method and system of the present disclosure can be used in any surgical intervention that involves a preoperatively planned instrumentation in which accuracy can be quantified by registration of preoperative and intraoperative scans.
  • Segmentation e.g., automated segmentation
  • computer vision techniques can be employed to align each preoperative vertebra with its intraoperative counterpart. Positions along one or more (e.g., all three) axes may be compared.
  • detection e.g., automated detection
  • Distance from the screw to the medial-inferior pedicle wall can be calculated, for example, by using the 3D outer circumference of the detected screw (approximated as a cylinder) and the 3D envelope of the preoperative pedicle wall as detected by a segmentation technique. The anatomic location of the relevant section of the pedicle wall can be determined and/or reported.
  • the most-breached point (for breached screws) and/or the point closest to being breached (for internal screws) can be determined and/or reported.
  • the minimum pedicle diameter can be calculated as the minimum diameter of the pedicle measured perpendicular to the planned screw axis (e.g., the largest cylinder that could geometrically fit down the pedicle without intersecting the pedicle walls in any direction).

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Surgery (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Medical Informatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Molecular Biology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Theoretical Computer Science (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Orthopedic Medicine & Surgery (AREA)
  • Physics & Mathematics (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Dentistry (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Radiology & Medical Imaging (AREA)
  • Quality & Reliability (AREA)
  • Artificial Intelligence (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

Un système chirurgical orthopédique comprend un processeur configuré pour recevoir des données d'image 3D préopératoires d'un placement de vis pédiculaire planifié dans une vertèbre d'un patient; obtenir des données d'image 3D peropératoires ou postopératoires d'un placement de vis pédiculaire dans la vertèbre du patient; et déterminer une mesure de précision du placement de vis pédiculaire dans la vertèbre du patient sur la base des données d'image 3D préopératoires et des données d'image 3D peropératoires ou postopératoires. La mesure de précision du placement de vis pédiculaire peut être basée sur une distance d'une vis pédiculaire à un os cortical d'une vertèbre d'un patient ou à une distance d'une vis pédiculaire à une région d'intérêt du patient.
PCT/US2025/020972 2024-03-21 2025-03-21 Procédé et système d'évaluation de précision de positionnement de vis pédiculaire Pending WO2025199473A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US202463568266P 2024-03-21 2024-03-21
US63/568,266 2024-03-21

Publications (1)

Publication Number Publication Date
WO2025199473A1 true WO2025199473A1 (fr) 2025-09-25

Family

ID=97140333

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2025/020972 Pending WO2025199473A1 (fr) 2024-03-21 2025-03-21 Procédé et système d'évaluation de précision de positionnement de vis pédiculaire

Country Status (1)

Country Link
WO (1) WO2025199473A1 (fr)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200352651A1 (en) * 2017-11-22 2020-11-12 Mazor Robotics Ltd. A method for verifying hard tissue location using implant imaging
US20210064220A1 (en) * 2016-04-28 2021-03-04 Medtronic Navigation, Inc. Method and Apparatus for Image-Based Navigation
US20210212772A1 (en) * 2010-04-28 2021-07-15 Ryerson University System and methods for intraoperative guidance feedback
US20230038678A1 (en) * 2016-03-12 2023-02-09 Philipp K. Lang Augmented Reality Display Systems for Fitting, Sizing, Trialing and Balancing of Virtual Implant Components on the Physical Joint of the Patient

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210212772A1 (en) * 2010-04-28 2021-07-15 Ryerson University System and methods for intraoperative guidance feedback
US20230038678A1 (en) * 2016-03-12 2023-02-09 Philipp K. Lang Augmented Reality Display Systems for Fitting, Sizing, Trialing and Balancing of Virtual Implant Components on the Physical Joint of the Patient
US20210064220A1 (en) * 2016-04-28 2021-03-04 Medtronic Navigation, Inc. Method and Apparatus for Image-Based Navigation
US20200352651A1 (en) * 2017-11-22 2020-11-12 Mazor Robotics Ltd. A method for verifying hard tissue location using implant imaging

Similar Documents

Publication Publication Date Title
Esfandiari et al. A deep learning framework for segmentation and pose estimation of pedicle screw implants based on C-arm fluoroscopy
US20220398776A1 (en) Determining relative 3d positions and orientations between objects in 2d medical images
US6747646B2 (en) System and method for fusing three-dimensional shape data on distorted images without correcting for distortion
US8897514B2 (en) Imaging method for motion analysis
US8152816B2 (en) Computer-assisted planning method for correcting changes in the shape of joint bones
US12256996B2 (en) Systems and methods for generating a three-dimensional model of a joint from two-dimensional images
EP3486873A1 (fr) Détection automatique d'implant à partir des artefacts d'image
JP6426608B2 (ja) 解析モルフォミクス：高速医用画像自動解析法
US20220265352A1 (en) Cross-modality planning using feature detection
US11478207B2 (en) Method for visualizing a bone
Knez et al. Computer‐assisted pedicle screw trajectory planning using CT‐inferred bone density: A demonstration against surgical outcomes
US20230363822A1 (en) Technique For Visualizing A Planned Implant
Yaniv Registration for orthopaedic interventions
US9576353B2 (en) Method for verifying the relative position of bone structures
CN109155068B (zh) 组合式x射线/相机介入中的运动补偿
WO2025199473A1 (fr) Procédé et système d'évaluation de précision de positionnement de vis pédiculaire
Knez et al. Computer-assisted pedicle screw placement planning: Towards clinical practice
Popescu et al. A new method to compare planned and achieved position of an orthopaedic implant
US12257092B2 (en) Method to superimpose rendering over spine hardware implants on images produced by Cbct scanner system
Yao et al. Automated extraction of anatomic landmarks on vertebrae based on anatomic knowledge and geometrical constraints
CN119968170A (zh) 基于二维图像的自动手术计划方法以及系统
EP4014913B1 (fr) Détermination de courbe d'implantation basée sur l'intelligence artificielle
Tatum et al. A novel system for markerless intra-operative bone and bone fragment tracking
Zheng et al. Reality-augmented virtual fluoroscopy for computer-assisted diaphyseal long bone fracture osteosynthesis: a novel technique and feasibility study results
US20250186160A1 (en) Technique For Processing Medical Image Data Of A Patient's Body

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 25774204

Country of ref document: EP

Kind code of ref document: A1