Systems and methods for automation of surgical correction of spinal deformities

Abstract

A system and method for automating tasks in surgery to correct spinal deformity of a patient are disclosed. The system will obtain preoperative data characterizing a defined patient and individual vertebral position and orientation from the patient's images to generate operative instructions for an automated machine to operate a correction apparatus that corrects vertebral deviations and realigne spinal deformity of the defined patient. The system, using a machine learning model, will receive intraoperative and postoperative feedback data from a plurality of patients to optimize the creation of optimized future preoperative correction plan.

Description

SYSTEMS AND METHODS FOR AUTOMATION OF SURGICAL CORRECTION OF SPINAL DEFORMITIESBackground

[0001] Evolution of spinal instrumentation continued till the nineties when all pedicle screws constructs became the benchmark operative strategy, providing stable fixation and allowing for safer and more effective manipulation of the vertebral bodies. Since then, and despite the major technological advancements in imaging, preoperative planning, medical device industry, and robotics, the fundamental concepts of spinal deformity correction remained largely unchanged. Deviation of a vertebra from its normal position in any of the 6 degrees of freedom (rotation around or translation along the X, Y, Z axes) can cause significant clinical consequences and disturbed quality of life. Spinal deformities are mostly multiaxial, and restoration of three-dimensional (3D) alignment is critical; inability to achieve precise segmental 3D correction is a long- felt, not yet solved, problem.

[0002] Current techniques for surgical correction of spinal deformities imparts realigning the deviated vertebrae along a rigid rod. Solid rods are a major restricting factor to vertebral manipulation by limiting vertebral motion to along and around the rod’s axis leading to an incomplete, imprecise, and less predictable 3D correction of individual vertebral deviations. Optimizing solid rod bending to conform three-dimensionally with the normal anatomy of the spine produced many generations of “smart rods” including patient specific rods; however, robots in spine surgery remains at the lowest autonomy level (Robotic Assisted) without any capability to automate the execution of surgical tasks. Precision surgery and patient-specific realignment are evolving technologies with a promise of a better fit to patient anatomy and improved outcome. However, the latest surgical pathways lack the accuracy and predictability to match these developments. Digitization, automation, and segmental 3D correction willpotentially transform spinal deformity correction into a more precise, predictable procedure paving the way for higher levels of Robotic autonomy.

[0003] Segmental correction devices incorporating at least one joint mechanism aimed at correcting individual vertebral deviations can overcome the challenges associated with the current surgical techniques using solid rods. One example of these correction apparatuses is the segmental Multi-Axial Segmental Coupler presented by Elsebaie et al in publication no. WO 2024132079 Al. The Coupler incorporates at least one degree of freedom joint mechanism and uses numerical data (angles and distances) to correct vertebral deviations, thereby digitizing the correction process. The Coupler collectively is capable of segmentally correcting spinal deformities three dimensionally by correcting individual vertebral deviations in the 6 degrees of freedom. The Coupler uses the motion control coordinate system that calculates individual deviations within the spinal deformity in six degrees of freedom: three linear along the X, Y, and Z axes and three rotational around those axes.

[0004] Despite avoiding the problems associated with solid rods, segmental correction apparatuses do not solve the problem of the lack of automation of the execution of surgical tasks in spinal deformity surgery. Automation of surgical intervention is gaining popularity; it offers an accurate, repeatable, fully controlled and eventually safer (less potential human errors) workflow; moreover, it allows the incorporation of the latest technological advancement in the field. Automating surgical actions offers several advantages including enhanced precision, increased reliability, reduced risk of human errors, improved visualization, potential for personalized surgery. In spinal deformity correction surgery, performimg realtime calculations of vertebral deviations and their required corrections and repeating this process intraoperatively to adjust the corrections till the final target spinal realignementis reached is essential for an efficient and time saving workflow. Using correction apparatuses to automate the surgicalcorrection of spinal deformities will be faced with many challenges and difficulties. A major challenge is developing a system that arrange the harmonious execution of complex workflow and interactions between the imaging data, the patient data, the personalized normal alignement data, the automated machine data, the correction apparatus data, the spinal fixation system data as well as the introperative spine and imaging data once the patient is under general anesthesia and in the prone posisition. Another critical problem, the relation between the input degree of rotation of the screwhead of the joint mechanism of the apparatus and the final rotational and / or translation motion of the vertebra; this motion needs to be analyzed, processed and standardized for an effective and accurate outcome. This cascade of motions and elements will depend on the characteristics of the correction apparatus, the joint mechanisms within the apparatus, the spinal fixation system and will be affected by the complex variability in patient, disease, spinal deformity, spinal fixation systems as well as the normal anatomical alignment data for the spine and pelvis. Accordingly there is a need for a system to provide a clear methodSummary

[0005] Various embodiments of the present disclosure solve the above problems by providing a system and method for automation of surgical correction of spinal deformity. In spinal deformity surgery there has been significant improvements in surgical automation in the areas of acquisition and analysis of information; decision making and plan of action; as well as implant designing including personalized implants and automated rod bending. On the contrary, automation in execution of surgical action has not followed through; the first prerequisites for automation of surgical tasks in deformity correction are segmentation and digitization, whereby the deformity is divided into multiple motion segments and the information perceived from the preoperative images of the patient is converted to a set of numerical data (distances and angles) identifying the precise individual vertebral deviations within thedeformity. Using these digits to automatically create specific digital operative instructioins will automate the surgical correction of spinal deformities.

[0006] With the advent of newer imaging technologies, Computer Aided Diagnosis, 3D Modeling; the data acquired allow for a highly precise 3D preoperative planning. These advancements need to further translate to precision hardware that can deliver correction plans with a high degree of accuracy. Some embodiments of the present disclosure describe method and system for automated spinal deformity surgical correction system offering higher levels of robotic autonomy in execution of surgical tasks. The system receives and processes pre-operative imaging data, personal data, disease data, spinal deformity data characterizing a defined-patient to generate a set of digital commands or instructions as a preoperative plan for an automated machine to operate a correction apparatus. The methods and system include obtaining and processing intra-operative and / or post-operative feedback data regarding radiographic (imaging) and clinical outcome for a plurality of spinal deformity patients, and training a machine learning model based on the feedback data collected to generate more optimized preoperative plans for future patients sharing similar characteristics. With some embodiments of the present disclosure, The digital commands will be received and executed by an automated machine including an automated motorized screwdriver, robotic screwdriver, collaborative robot, or a robotic system.

[0007] In order to automate a segmental correction apparatus, the function of the device and the motion of the vertebra in relation to the correction apparatus input need to be analysed. According to some embodiments of the present disclosure the motion of vertebral correction as a result of operating the correction apparatus is divided into 2 types of motions. The First Motion (screwhead to rod / pedicle screw): is the motion of the end effector arm / rod attached to the vertebral anchor that results from rotating the screwhead of its corresponding joint mechanism. The first motion is based on the gear ratio of the Joint Mechanism used, this is a fixed ratio for each specific joint and canbe calculated by dividing the number of threads of the driven (output) gear by the number of threads of the driving (input) gear. Accordingly the First Motion will be automated by a process described as Automatic Automation, which is based on a system of rules and equations that can be done through mathematic calculations processed by a programmed software. On the contrary the Second Motion (rod / pedicle screw to vertebra) is the motion of the Vertebra as a result of the motion of the end effector arm / rod motion. This Second Motion is affected by many variables including size of the vertebra, tissue costraints and stiffness in different direction of vertebral motions, intervertebral and interpedicular screw distances, center of rotation of the end effector arm in relation to the center of rotation of the vertebra, distance between the correction apparatus and the vertebra, and the coupling phenomenon, which is defined as changing in one axis of motion due to changes made at a different axis, as well other variables related to the patient and the deformity. Therefore, the Second Motion will be automated by a process described as Automatic Automation will require algorithms and machine learning models to analyze these complex data, learn, and make prediction to create of a preoperative plan taking into account all the previous variables. Regarding the force and torque required there are also many variables affaecting the amount of force required to correct vertebral deviation including stiffness of the curve, level of the deformity, etiology of the deformity, associated surgical releases and osteotomies, congenital anomalies; these variables will need machine learning model to be taken into account when creating a preoperative instructions regarding the required input torque.Brief Description of the Drawings

[0008] FIG. 1 is an illustration of the overall workflow for the process of automation of spinal deformity correction;

[0009] FIG. 2 is a perspective view of a segmental correction apparatus of prior art publication no. WO 2024132079 Al;

[0010] FIG. 3 s an exploded, perspective view of the correction apparatus of FIG.2;

[0011] FIG. 4 is a schematic illustration of an example of segmentation process of a dorsolumbar deformity spanning from the 2ndDorsal Vertebra to 4thLumbar Vertebra;

[0012] FIG. 5 is an illustration of a correction apparatus attached to 2 pedicle screws anchored to the corresponding adjacent vertebrae;

[0013] FIG. 6 is a flow diagram illustrating a process for generating the required corrections in the 6 directions of freedom from spine deformity images;

[0014] FIG. 7 is an exemplary table of the analysis of the actual deviations, target positions, and required corrections for one Lumbar and one Thoracic motion segments;

[0015] FIG. 8 is a flow diagram illustrating the process of generating the operative instructions to correct a single Axis deviation for one vertebral motion segment;

[0016] FIG. 9 are schematic illustrations and drawings for the automation process for each of the 6 joints of the correction apparatus controlling the 6 degrees of freedom;

[0017] FIG. 10 is an exemplary software application view of the set of the data required to generate automatic operative instruction for the automated machine or motorized correction apparatus;;

[0018] FIG. 11 is an illustration of the overall workflow for the process of automation of spinal deformity correction using a remote controller and motorized correction apparatus for a non-anesthetized patient at an outpatient setting.

[0019] FIG. 12 is a flowchart for generating preoperative plan or instructions for an automated machine controlling and operating a correction apparatus to correct spinal deformity based on processing pre-operative data through a machine learning model; and for using intra-operative, post-operative, follow data feedback data to train the machine learning model;

[0020] FIG. 13 is a flowchart for generating, using machine learning mode, an optimized preoperative spinal deformity correction plan for a defined patient using previously generated and stored operative plans of a plurality of patients sharing common characteristics with the defined patient.Detailed Description

[0021] The embodiments of the invention describe a system for automation of surgical tasks. The embodiments consist of generating an automated surgical plan based on radiographic and clinical data, then applying the automated surgical plan to patients during surgery.

[0022] Some aspects of the present disclosure Eire directed to automation of spine correction for the treatment of spinal deformities. The disclosure further provides methods for using vertebral deviations acquired from spine images to generate and deliver automated operative instructions to an automated machine in order to steer a correction apparatus (attached to the vertebrae) to manipulate the vertebrae to correct the spinal deformity. The correction apparatus in the current disclosure is meant to be any correction device incorporating at least one joint mechanism that, when attached to the spine, is capable of correcting vertebral deviation in at least one degree of freedom. The correction coupler described in publication WO 2024132079 Al is just one example of such a correction apparatus. The methods of the present disclosure may be used to treat spinal deformities including but not limited to scoliosis, kyphosis, kyphoscoliosis, degenerative scoliosis, spondylolisthesis, etc. The correction andfixation apparatuses are configured to correct the individual vertebral deviations thereby correcting the motion segments and subsequently the whole spinal deformity. The automated generated operative instructions and collectively the computer implemented methods can, for example, operate the correction apparatus to correct axial rotation, coronal Angulation, sagittal angulation, compression-distraction, medio-lateral and antero-posterior translations; therefore, it can effectively allow an automated correction of any vertebral deviation in all 6 axes of motion. The methods of the present disclosure are, in some non-limiting embodiments, designed in such a way that adapts the latest evolving technologies by using only numbers (angles in rotation and distances in linear motions) to analyze and describe all types of vertebral deviations; in addition, some embodiments of the present disclosure include a correction apparatus that incorporates uniaxial / uniplanar (robotic) joints to allow for precise, fully controlled, calculated and reproducible motions and corrections; these joints can be controlled and operated by readily available automated machines including robotic screwdrivers and collaborative robots.

[0023] The invention will now be described more fully with reference to the accompanying drawings, in which examples of embodiments of inventive concepts are shown. Embodiments of the present disclosure are directed to improving spinal surgery through use of digitization, automation and utilizing machine learning algorithms. These capabilities have the potential to benefit both surgeon, and patient during spinal deformity and other spinal surgeries.

[0024] Automation of surgical intervention has been classified regarding the degree of automation into broad categories including A. Decision-Making and Planning B. Execution of Surgical Action. In the current surgical spinal deformity correction pathways, there is great discrepancy between the 2 categories regarding the degree of automation. The advancements in Decision Making are not matched by their SurgicalExecution counterpart. Preoperative planning is currently precise, digitized, largely automated with the capability of a true three-dimensional analysis of spinal deformity and segmental vertebral deviations precise measurements. On the contrary, the Execution of Surgical Action is rather imprecise, less predictable, non-digitized, and non-automated; in addition, these pathways cannot achieve a true segmental 3D correction of vertebral deviations. These long felt not yet solved problems are in part due to the current adoption of manual correction techniques using solid rods. These problems can be solved with segmental correction devices capable of executing a digitized correction of individual vertebral deviations and allowing for automation of the Execution of Surgical Action. Digitization of correction of individual vertebral deviation requires segmental multi-axial devices capable of precisely (in degrees for rotation and millimeters for translation) correcting vertebral deviations in the 6 degrees of freedom. Developing system and method capable of preparing and delivering precise and patient specific digital operative instructions and commands (preoperative plan) capable of operating these devices is mandatory to their efficient function.

[0025] The following embodiments are described in the context of a spinal correction methods used to treat spinal deformities. Some descriptions of the present disclosure and figures utilize common or accepted motion terminology and the familiar human anatomical and spine surgery terminologies. Some examples in accordance with principles of the present disclosure will now be described, by way of example only, in detail with reference to the accompanying figures; not all features of the various applications and implementations of the present disclosure are described in the following figures.

[0026] The overall workflow for the automation of spinal deformity correction is depicted in FIG. 1 Method 102 receives data 107 related to patient 105 (having spinal deformity) including personal data, disease data and spinal deformity data in addition to correction apparatus 104 characteristics data 108 as well as imaging data 109 relatedto preoperative and / or intraoperative spine and pelvis imaging 101 of the spine and pelvis. The method will also receive the the normal anatomical alignment data in addition to the patient’s vertebral deviation imaging data to calculate the required correction data. The method, using machine learning models, processes the received data to generate operative instruction 110 to an automated machine 103 (Automated Motorized Screwdriver, Robotic screwdriver, Collaborative Robot, or Robotic Device Motor) to connect and operate the correction apparatus 104 attached to the vertebrae in order to manipulate the vertebrae in precise precalculated motions effectively delivering correction of the vertebral deviations and subsequently realigning the vertebrae and correcting the spinal deformity. As such the method transforms the digital data acquired from the patient’s images 101 to electric signals to the Robot 103 that drive a correction apparatus 104 to correct the deformity.

[0027] Referring to FIG. 2 and 3, they are figures from a prior art publication No. WO 2024132079 Al, they described a correction apparatus or correction and fixation coupler for digitized surgical execution of segmental correction of spinal deformity. The device is capable of segmentally correcting spinal deformities in three dimensions by correcting individual vertebral deviations in the 6 degrees of freedom. A motion control coordinate system calculates individual deviations within the spinal deformity in six degrees of freedom: three linear along the X, Y, and Z axes and three rotational around those axes. The systems or device can correct any of twelve directions six rotations (three clockwise and three counterclockwise) and six linear (three positive and three negative). The system of Patent Application No. WO 2024132079 Al include at least a Multiaxial Coupler Unit with a jointed body incorporating multiple self-locking uniaxial “Robotic” joints (revolute and linear), and 2 end effector arms / rods. The first and second arms attach to anchors (pedicle screws) secured to vertebrae; the Coupler’s joints are designed to be selectively mobilized in calculated kinematic solutions to mobilize the corresponding end effector arms. The motion ofthe joint(s) enables the first and second arms, and hence the attached vertebrae, to move relative to one another from their actual deviated position to a target position to correct the vertebral deviation and realign the spinal deformity. FIG. 2 illustrates a perspective view of a correction-fixation coupler unit publication No. WO 2024132079 Al and FIG. 3 illustrates a exploded, perspective view of the correctionfixation coupler unit showing three joint mechanisms, a first and a second end-effector arms or rods as well as the enclosures of the body of the coupler. These figures are shown and described here only as explanatory example of a segmental correction apparatus aimed at correcting vertebral deviation to realign spinal deformity.

[0028] The current invention can be used without limitation with any correction device that uses digital data for correction of vertebral deviation as well as any uniaxial, multiaxial, ball and socket jointed device aiming at segmental correction of spinal deformities.

[0029] The joint mechanism within the apparatus controls motion in at least one single axis either in rotation or translation. The joint mechanism will be operated and controlled by an actuator or screwhead and it will deliver its motion to an end effector arm / rod attached to the vertebra via an anchor or pedicle screw. The correction motion of the vertebra using the apparatus can be fully controlled and digitized, therefore automated, a significant advantage not possible using any of the current surgical pathways. To achieve this digitized correction, each joint mechanism will be operated based on precise calculated predetermined instructions. These digital instructions or command will be received by an automated machine that connects and control the motion of the apparatus, allowing the apparatus or coupler to deliver the precise targeted vertebral deviation correction for the corresponding degree of freedom. Operating, driving, or rotating the actuator or screwhead of each joint can be done using one of the following ways: A. Automated Motorized Screwdriver (non-robotic) held controlled andoperated by a user / surgeon. B. Robotic Screwdriver controlled and guided by a user / surgeon. C. Collaborative Robot (COBOT) working hand in hand with the surgeon D. Autonomous Robotic System taking over control to fully execute the task of correction of vertebral deviation supervised by surgeon. E. Remote controller held by a user / surgeon to remotely operate a motorized correction apparatus incorporating an internal motor. These devices controlling and operating the segmental coupler will receive a precisely calculated prepared operative instruction or operative plan prepared and delivered by a computer system using machine learning model. This workflow will allow higher degrees of Robotic autonomy in spinal surgery.

[0030] Referrring to images 101 in FIG. 1 showing Anteroposterior, Lateral Radiographs, and 3D reconstruction of the whole spine and pelvis 6, all the 3 figures are showing a spinal deformity including an upper thoracic and lower lumbar curve. The 3D reconstruction of the vertebral column can be generated based on low dose standing Radiographs using the EOS biplanar imaging technology which is currently available and increasingly used in clinical practice. 3D reconstruction images can also be acquired using CT scans, MRI, or any other current or future imaging modality. As such preoperative imaging is usually taken in the supine position for MRI and CT scan; in the standing / erect position EOS imaging; in Supine or Standing in Plain Radiographs. Intraoperatively with a posterior surgical approach, images are taken in the prone position on a radiolucent operative table 106 (FIG. 1). Spinal deformity magnitude as well as individual vertebral deviations may change with the change of patient’s position, therefore preoperative prone imaging simulating the intraoperative patient position will be acquired before surgery to represent the most accurate vertebral deviations for the patient during the intraoperative correction procedure. In addition, during our automated correction of individual vertebral deviations motion coupling (defined as changing in one axis of motion due to changes made at a different axis) may occur at the same motion segment and / or at the adjacent levels; this may result inchanges of the initially measured vertebral deviations. Therefore, additional images during different stages of corrections and / or real time tracking sensors or cameras may be needed in order to accommodate for these possible position changes. Augmented reality and machine vision technology may help in this context.

[0031] After receiving the digital images, a segmentation process, as shown in FIG. 4, will divide the spinal deformity into multiple motion segments; a typical motion segment includes 2 adjacent vertebrae with an intervertebral disc in-between. Individual vertebral deviations will be measured as deviations of a First Vertebra VI relative to the Second Vertebra VO; therefore, the correction apparatus will manipulate VI in a precalculated corrections relative to VO to align the motion segment. Segmentation can be done between adjacent and non-adjacent vertebrae (skipping one or more vertebrae) according to the site of placement of the correction apparatus. Based on these deviations the correction required at each motion segment will be identified and the correction apparatus that will be used to correct these deviations will be chosen. The computer will receive the data related to individual joint mechanisms of the correction apparatus 104 that will be used to correct the deviations for the specific segment including the input / output ratio of the apparatus for this specific motion. Using a computer implemented method the computer will generate operative instructions to an automated machine (robot) 103 which will engage with the interface of the specific joint of the correction apparatus 104. Once engaged the robot will drive the screwhead of the joint mechanism to operate the apparatus to manipulate the vertebra in a certain precalculated motion into a known direction, magnitude, and force to correct the corresponding vertebral deviation. Repeating this process in all the motion segments included in the deformity will collectively correct the whole spinal deformity.

[0032] FIG. 4 depicts a diagram showing a vertebral column with dorsolumbar spine deformity. The spine in the diagram starts from the second Dorsal (D2) to the fourth Lumbar (L4) with the deformity starting from D4 to L2. D4 is the first deviated vertebra at the cephalic (top) end of the curve and L2 is the first deviated vertebra at the caudal (bottom) end of the curve. In one of the sequences of correction of spinal deformity. In one of the possible sequences of correction, the correction process will start from the top 701 and bottom 702 of the curve working till the apex 703 of the deformity in the middle area. D3 is a neutrally oriented vertebra VO, adjacent to D4 which is the first deviated vertebra at the top VI; the process starts by correcting VI relative to VO. L3 is a neutrally oriented vertebra V-0 adjacent to L2 the first deviated vertebra at the top VI; the process starts by correcting V-l relative to V-0. The last vertebra to be corrected in this specific sequence will be the apical vertebra VA which is the most deviated vertebra around the middle of the curve. The sequence of the correction of the vertebral motion segments will be planned according to the type of the curve. The type, number, magnitude, direction, force of corrections will be calculated for each motion segment and executed by the corresponding correction apparatus. This process can be repeated sequentially from both ends till the apex to collectively correct the whole spinal deformity. Another possible correction sequence is to start with the motion segments with the least resistance and go back and forth between different motion segments giving time for the tissue yielding to decrease the force required for correction. A torque limiter can be used with the motorized or robotic screwdriver to make sure that the force applied to the spine stays within safe limits.

[0033] FIG. 5 depicts a correction apparatus 104 which manipulates the vertebral body VI 511 relative to VO 512 to correct the deformity and realign the motion segment. The first arm 501 extending from the first end of the apparatus 104, and the second arm 502 extending from the second end of the apparatus 104, each of the arms attach to thecorresponding pedicle screws 503 and 504 anchored in vertebral body VO and VI respectively. The interface of the correction apparatus 104 include 6 screwheads or drivers 505, 506, 507, 508, 509, 510; each controlling the mobilization of one of 2 the arms in one of 6 directions of freedom. Rotating the screw head in a certain direction at a specific angle of rotation will deliver a precalculated motion in a specific direction with a specific magnitude for 1 of the 6 degrees of freedom. Similarly Rotating the screw head at a certain torque will deliver a precalculated force transmitted to mobilize the corresponding vertebra. Each screw head will be numbered from 1 to 6 and color- coded for ease of use; the 6 colors could be Red, Green, Yellow, Brown, Blue, and Orange and each screw head will be calibrated to identify the amount of rotation with each marking for the 360 degrees (single full rotation). A correction apparatus will contain a single or multiple joint mechanisms / screw heads according to the number of deviations requiring correction in the corresponding deformed motion segments. The correction apparatus 104 in FIG 14 includes 6 screwheads for 6 joint mechanisms for the 6 directions of motion. Screwhead 505 control X Axis Rotation, Screwhead 506 control X Axis Translation, Screwhead 507 control Y Axis Rotation, Screwhead 508 control Y Axis Translation, Screwhead 509 control Z Axis Rotation, and Screwhead 510 control Z Axis Rotation. Regarding the correction apparatus, FIG. 2 and 3 showed one example of vertebral deviation correction apparatus of prior art publication No. WO 2024132079 Al; however, a correction apparatus in the current disclosure include any correction device with at least one joint mechanism either uniaxial, multiaxial or ball and socket joints intended to correct vertebral deviation in at least one degree of freedom.

[0034] FIG. 6 depicts a flow diagram illustrating a process for generating the required corrections in the 6 directions of freedom from Spine Deformity Images. The computer receives the digital images of the spine and pelvis including the spinal deformity as acquired preoperatively or intraoperatively 601. 3D reconstruction images for thewhole spine and pelvis are created 602, then the spinal deformity is divided into multiple motion segments, each motion segment include 2 vertebrae and a disc inbetween 603. For each motion segment, identification of direction and magnitude of vertebral deviation of a first vertebra relative to the 2ndvertebra in the 6 degrees of freedom is done 604. Receive the normal anatomical alignment measurements and patient specific spino-pelvic target alignment measurement for the patient 605. The optimal position of the motion segment will be determined based on many variables related to the patient, the disease, the deformity 606. The optimum position of the first vertebra relative to the second vertebra will be determined in the 6 degrees of freedom; in 4 out of 6 the value of the deviation should be zero namely coronal Angulation, axial rotation, anteroposterior displacement, and mediolateral displacement. For sagittal angulation the direction will be positive in the dorsal spine (Kyphosis) and negative in the lumbar spine (Lordosis) and the magnitude will depend upon the level of the motion segment. As for craniocaudal translation the optimum position will depend upon the ideal normal disc height, which will again depend on the level of the motion segment. With the technology of artificial intelligence and machine learning these values in addition to the torque required to execute these corrections will be determined and personalized for each patient for the specific optimum patient specific alignment. Based on the actual deviation and the optimum position, the required corrections will be determined both in directions and magnitude 607. Step 608 represents the final correction data that the correction apparatus needs to deliver in order to correct the corresponding vertebral deviations. This process will be repeated for all the motion segments included in the spinal deformity which will collectively correct the whole spinal deformity 609 and 610.

[0035] FIG. 7 shows an exemplary table 1 including the detailed data representing the process of correction calculations for vertebral deviations in 2 separate motion segments. For each motion segment the rows represent the Actual Deviation, Target OptimumPosition, and Required Correction for each of 6 degrees of freedom. As an example, for motion Segment L2 -L3 in the Lumbar spine the deviations are as follows: 1° Kyphosis, 2 mm Translation to the Left, 4° Axial Rotation to the Right, Disc Height 7 mm, Coronal Angulation, 7° to the Right, and 2 mm Anterior Translation. The optimum position for this level for this patient (and for a normal spine) is Zero in Coronal Angulation, Zero Axial Rotation, Zero Anteroposterior Translation, and Zero Mediolateral Translation. The “Normal” target Sagittal angulation for L2-3 in this patient is 5° Lordosis and the normal Disc Height is 10 mm. Based on the Actual Deviation, and the Target Optimum Position, the Required Correction for the 6 degrees of freedom were calculated as follows: 6° Lordosis, 2 mm Translation to the Right, 4° Axial Rotation to the Left, 3 mm Distraction, Coronal Angulation, 7° to the Left, and 2 mm Posterior Translation. Table 1 also shows another example of the detailed data for D8-D9 Dorsal Motion Segment.

[0036] In some embodiment of the current invention, the method and system of the current patent application are described to precisely deliver digitized operative instructions to an automated motorized screwdriver or a robotic system including at least one motor configured to operate and mobilize at least one joint mechanism by driving an actuator or rotating a screwhead controlling the at least joint mechanism. There are 3 digital data entries required to operate each joint mechanism: a. direction of rotation of the screwhead b. amount (magnitude) of rotation of screwhead and c. torque or force applied to rotate the screwhead. These 3 digital entries or instructions or commands will control the motion of the end effector arm of the device allowing it to deliver the targeted correction of the corresponding attached deviated vertebra. These instructions need to be precisely optimized in order to have a safe and effective outcome for the surgical spinal deformity procedure. For a safe, effective, precise, and reproducible workflow, the system and methods will use machine learning algorithms or similar strategies to maximally optimize these commands. The final aim of the prepared planis to deliver a plan resulting in the best possible patient reported outcome with the least possible adverse effect. Using Artificial Intelligence and Machine Learning, the system and method will use multiple feedback loops to automatically find the best operative plan fitting for each defined patient and each specific spinal deformity from previously collected, aggregated, and stored operative plans. Training of the machine learning model with feedback loops will allow it to create a preoperative plan (instructions) requiring the fewest number of subsequent changes. The more data it analyzes, the better it becomes at making accurate predictions regarding the most suitable preoperative correction plan for each defined patient.

[0037] For automation of execution of surgical action in correcting spinal deformities, the segmental Device or Coupler used to correct the deformity comprises a number of joint mechanisms corresponding to the number of vertebral deviations the device is intended to correct. Each joint mechanism controls and correct one degree of freedom. To correct a deviation in one degree of freedom, the actuator or screwhead controlling the corresponding joint mechanism is rotated in predetermined calculated a. specific direction with b. a specific magnitude / degree using c. a specific torque / force to execute the precise target vertebral correction safely and effectively. These 3 data entries or digital instructions constitute the main operative instructions required to operate the device and correct the deviation of the targeted vertebra. The surgeon using an automated motorized screwdriver, or the robot receiving these digitized operative instructions will drive or operate the actuator or screwhead controlling the joint mechanism. These instructions include: A. The direction of rotation of the screwhead. B. The amount / magnitude / degree of rotation of the screwhead. C. The amount of torque / force applied to the screwhead to deliver enough force, within safe limits, to correct the targeted vertebral deviations.

[0038] Having 3 defined digital data entries as operative instruction or preoperative plan to correct spinal deformity is a great advantage compared to all complex plans described in other robotic and robotic assisted surgery in different surgical specialties. The complex data in the operative instructions or operative plans required to operate Robots is a major challenge in developing robots with higher levels of robotic autonomy. Having only 3 digital entries (numbers) as the operative plan to perform a complex type of surgery like correction spinal deformity is an achievement that can potentially pave the way for Robots higher levels of autonomy.

[0039] The process of “Execution of Surgical Action” using a correction apparatus entails rotating the screwhead of the joint mechanism of the correction apparatus to mobilize its corresponding end effector arm attached to a pedicle screw anchored to a targeted deviated vertebra to manipulate the vertebra and correct vertebral deviation. The motion delivered by the correction apparatus is delivered first to the end effector arm which will move and subsequently mobilized its attached vertebra. The delivered motion can be analyzed to First Motion transmitted to the arm and Second Motion transmitted to the vertebra. FIG. 5 shows the 2 types of delivered motion of the correction apparatus. The first Motion 513 is the result of an internal process, occurring within the body of the apparatus, in which the screwhead (Input) rotation results in a corresponding motion of the end effector arm (output). The second Motion 514 is an external process, occurring outside of the correction apparatus, in which the motion of the end effector arm (output) results in a corresponding motion of the targeted attached vertebra (Target). Motion I (Input to Output): “Input” is the screwhead rotation and “output” is the end effector arm / rod delivered motion (rotation or linear). The motion in stage 1 is transmitted via the motion of the gears of the joint mechanism. Motion II (Output to Target): “Output” is the end effector arm motion and “Target” is the vertebra delivered motion. The motion in stage 2 is directly transmitted form the end effector arm to the deviated target vertebra.

[0040] The required correction of vertebral deviation is calculated from the difference between the position of deviated vertebra and the target anatomical patient specific vertebral alignment. Once the correction parameters are calculated, the surgeon will need to operate the device to deliver the targeted precalculated correction. For each of the 2 Phases (Internal and External) included in this process the 3 elements of motion are: a. Direction b. magnitude c. Force. Each of the 3 elements will be discussed in each of the 2 Phases of correction in order to analyze the optimum way of its calculation.

[0041] Motion 1 (Internal): A— Direction: The Direction of rotation of the screwhead is a binary choice either Clockwise or Anticlockwise which can be represented in the operative instruction by binary digits (0 and 1). Each screwhead (input) rotation either Clockwise CW or Counterclockwise CCW will result in a specific direction of its endeffector arm; subsequently the arm will either rotate CW or CCW for revolute joint or the arm will move in and out of the coupler in a linear motion (positive and negative) for translational joints. The relation of direction of screwhead rotation relative to the direction end effector arm motion is known and directly determined based on the design of the joint mechanism; not requiring calculations or software. B— Magnitude (degree of rotation): Each uniaxial joint within the Coupler has its unique “input / output ratio”. In a self-locking worm gear for example , the input (worm) typically rotates 360 degrees for every single rotation of the output gear (worm wheel), resulting in a 1:1 (or "one to one") input to output degree of rotation ratio, where the " 1 " represents the output rotation and the "1" represents the input rotation; however, the actual gear ratio depends on the number of teeth on the worm wheel and the number of "starts" on the worm. The calculation for gear ratio is simple: divide the number of teeth on the driven gear (or output gear) by the number of teeth on the driving gear (or input gear). This can be represented by the gear ratio formula: Gear Ratio (GR) = Number of Teeth on Driven Gear (T2) / Number of Teeth on Driving Gear (Tl). One full turn or part ofa turn of the screwhead (Input / Driving Gear) will result in a corresponding predetermined degree of motion delivered at its arm (Output / Driven Gear). The precise magnitude and direction of screwhead rotation required to operate a joint mechanism to execute a desired specific vertebral motion is predetermined from calculations of the input out ratio of the joint mechanism. These digital instructions (in degrees) can be generated manually using simple ratios and equations; however, a software was developed to automatically generate these operative instructions instantaneously with less potential for human errors. C— Torque: Similar to the magnitude, each uniaxial joint has its unique “input / output force ratio”. In a worm gear, the input-output force ratio is directly related to the gear ratio, meaning that for every unit of force applied to the input (worm), the output (worm wheel) will experience a significantly higher force due to the high reduction ratio achieved by the worm gear design; essentially, a small input force can produce a large output force, but at the cost of a much slower output speed. Worm gears typically have a very high gear ratio, calculated by dividing the number of teeth on the worm wheel by the number of threads on the worm, allowing for large force amplification. Formula for calculating the force ratio: Gear ratio = (Number of teeth on worm wheel) / (Number of threads on worm). If a worm gear has a worm with 1 thread and a worm wheel with 40 teeth, the gear ratio is 40: 1. This means that for every unit of force applied to the worm, the worm wheel will produce 40 units of force. A torque applied to the screwhead will generate a corresponding force delivered at the end effector arm that will be used to manipulate the vertebra. The precise torque of screwhead rotation required to operate a joint mechanism to deliver a desired specific force is predetermined from calculations of the input out force ratio of the joint mechanism according to the force required to for correction intended by the joint. A software FIG 10 was developed to automatically generate these operative instructions.

[0042] All the instructions or digital entries for a precise operation of stage I are based on ratios and calculations without the need for complex algorithms, ML, or feedback loops. Simple mathematical equations will be enough to determine the direction, magnitude and torque required at the screw head to deliver the targeted direction, magnitude, and force delivered at the end effector arm. Motion II (External): A.B. The direction and magnitude of motion of end-effector arm is either in rotation (clockwise and anticlockwise) for revolute joint or linear displacement to move in and out of the coupler (positive and negative) for translational joints. This same direction of motion of the arm will be transmitted directly to the anchor (pedicle screw) and accordingly to the targeted vertebra. However, the combination of direction and magnitude of motion delivered to the end effector arm of the device will not produce the same exact direction and magnitude of motion of the targeted vertebra. There are many factors and variables resulting in this difference, some of main reasons for this discrepancy include but are not limited to: 1. The axis of rotation of the end effector arm is different than the axis of rotation of the target vertebra. The arm is attached to the vertebra via a posterior pedicle screw inserted in the pedicle, so the axis of rotation of the arm is posterior and on one side (Right or Left) compared with the axis of rotation of the vertebral body which is more anterior and central. Rotating around different axes of rotation will change the direction of motion. 2. The shape and size of the target vertebra vary with age, gender, disease, and anatomical level; all these variables can affect the position center of rotation of the vertebra as well as the distance between the arm of the coupler and the centroid of the vertebra. 3. The soft tissue and bony constraints of the vertebral motion segment / s will greatly vary with many variables including age, gender, disease, level, etiology of the deformity, and type of surgery including the presence of osteotomies. 4. Coupling Phenomenon: a motion that occurs along or around an axis of the cartesian coordinate system often produces another motion along or around another axis. Correction of deviation in one axis can be simultaneously associated with another motion around a second axis in another plane.

[0043] For clarification on how the change of axis of rotation can significantly change the direction and magnitude of motion delivered to the vertebra we will mention 2 examples. An axial rotation of the end effector arm when it is delivered to the vertebra via a unilateral pedicle screw the motion will result in a certain degree of axial rotation with an accompanying degree of translation as the axis of rotation of the arm is eccentric in the vertebra. Another example, a linear displacement of the arm if it is done bilaterally with 2 couplers and 2 pedicle screws it will result in a linear displacement of the vertebral bodies however if it is done on one side with unilateral screw it will result in a linear displacement on the ipsilateral side as well as rotation of the whole vertebra in the coronal plane. C. The torque or force applied to the screw head or actuator of the joint mechanism to manipulate the vertebral body should be high enough to produce the required motion while staying within the safe limits of the implant and bone mechanical strength. Otherwise, less than optimum correction occurs, or complications arise due to failures occurring at the implant or bone-implant interface. There are many variables which can affect the magnitude of force required for effective deformity correction as well as other variables affecting the threshold for mechanical failure 1. Correction Force: To effectively manipulate a vertebra with a certain magnitude of motion, the torque required to execute this motion will vary significantly between different patients depending on age, gender, level, severity of deformity, stiffness of the curve, etiology of the deformity, Body Mass Index. 2. Threshold for mechanical failure: The safe limit for implant and bone mechanical failures varies depending on many factors including the bone density, the pull-out strength of the anchor, the diameter, length, type of threads of the screw as well as the position of the screw within the pedicle.

[0044] Accordingly, in Phase II the amount of rotation and torque required to be delivered to the vertebra to achieve a safe and effective correction cannot be determined based on a fixed equation or formula. There are many variables related to the patient, disease,type of deformity, type of spinal implants that can affect these requirements for each defined patient. This will require Machine Learning Model that can take all these variables into account as well as feedback loops for training the Machine Learning model and optimizing the preoperative instructions for each defined patient.

[0045] Phase I and Phase II executions shows clearly the dichotomy between automatic and autonomous behaviors in surgical robots and automation of execution of surgical actions. Preparing Phase I instructions require an Automatic Process: these are completely predictable, as they follow well-established deterministic theories. There are variations of behaviors for an automatic system, and these can be programmed by a simple software to facilitate the mathematical calculations. Phase II require an Autonomous Systems: these by contrast, are able to make large adaptations to a change in external conditions by planning its tasks. The planning function requires a wider domain knowledge and the use of cognitive tools, such as ontologies or logical rules that do not exist within an automatic system. A Machine Learning Model is one of the commonly used tools for such autonomous functions. The categorization of Phase I and Phase II is described for the purpose of explanation and analysis of different motions and functions, however all these details will be integrated in a single computer-based method where the required corrections is received by the Computer System and the operative plan is delivered automatically to the robot or automated motorized screwdriver.

[0046] FIG. 8 is an exemplary flow chart for a method of determining the operative instructions for the First Motion based on vertebral deviations and the corresponding joint mechanism’s characteristics. For a specific motion segment, deviation will be calculated for Vertebra 1 (a more deviated vertebra) to Vertebra Zero (a normally positioned or less deviated vertebra) 801. The process of correcting the deviations around and along the X Axis 802 will be demonstrated. For the rotation around the XAxis representing sagittal angulation 803, the correction required 805 will be calculated based on calculating the target position relative to the actual deviated position 804. The required correction and its corresponding joint mechanism’s input out ratio 806 will be received by a computer, and a computer implemented method will determine the final operative instruction 808 that will be fed to the automated machine controlling the correction apparatus or to the motorized correction apparatus directly. The process of correction of translation along the X Axis is also shown in FIG. 8.

[0047] FIG. 9 is a schematic illustrations and drawings showing the method of automated correcting of spinal deformity for a single motion segment in the 6 degrees of deviation. For each degree of freedom, the Computer will receive direction and magnitude of required correction as well as the input out ratio of its corresponding joint mechanism. A method will generate the direction and magnitude of rotation that the corresponding screwhead of the correction apparatus will need in order to align the 2 vertebrae and correct the deformity in this specific degree of freedom. Motion 901, 902, 903 are rotational motions while 904, 905, and 906 are translational. After the patient is anesthetized and intraoperative images are acquired and analyzed, for each motion segment there will be possible deviations in 6 degrees of freedom. An average spinal deformity including 10 segments, there will be 60 possible deviations and therefore 60 required corrections and for each correction there will be a corresponding joint mechanism’s input output ratio. All these complex calculations cannot be done manually by a human without a computer implemented methods that generate the operative instructions automatically and instantaneously. If we put into consideration the continuous changing of vertebral deviations with ongoing corrections; in addition to the real time sensors that help in identifying the position of the vertebra after each correction adjustment this process can be repeated many times during surgery which makes it impossible to implement and execute the process in a timely and precise waywithout the help of the system and method and the software developed for this purpose of Automatic Automation.

[0048] FIG. 10 shows an example of an actual view of the software application that generates the operative instructions for phase 1 motion by automatic automation. The function of this process is automatically generated to generate the input of the screwdriver, another view of the final stage of the computer implemented automation process for the 6 degrees of freedom as the data entered and processed in the Computer where each correction angle and direction is entered as well as the ratio of the joint input output; both entered data will automatically generate the operative instruction in the form of screwhead rotation and direction of the corresponding joint mechanism in the correction apparatus. In addition the same process is done for the force measurements

[0049] FIG. 11 shows the workflow of the remotely operated automation of surgical correction of spinal deformity where the correction apparatus is motorized with an internal automatic motor that receives signal from a handheld remote control. This type of wireless signals will have an advantage of correction of the deformity while the patient is awake at an outpatient setting. The Correction apparatus will be surgically implanted at the specific level / s with deformities requiring correction. Post operatively t the operator will use a remote controller to control and operate the correction apparatus to correct the deformity either at one or multiple sessions taking the advantage of the phenomenon of tissue yielding which can facilitate correction and reduce the stresses on the apparatus and fixation.

[0050] For a safe and effective automated surgical procedure, a precise preoperative plan is mandatory. Avoiding errors and risks associated with non-optimized preoperative plan is a priority especially with the potentially catastrophic complications associated with the surgical treatment of spinal deformities. The uncertainty associated with the multiple variables related to the patient’s anatomy,disease status, and characteristics of the deformity makes delivering an effective precise operative plan not possible using the conventional methods in Phase II motion which requires autonomous automation. The use of a Machine Learning Model and algorithms will create preoperative instructions based on the specific inputs for each defined patients and the plan can be optimized based on the previously collected plans taking into account and adjusting for these variabilities. Training the MLM using intraoperative and postoperative feedback from all the patients subjected to this procedure will help in the creation of the optimal Correction Apparatus operative instructions for a defined patient. In automated execution of surgical action, including automated correction of spinal deformities; the aim is to find the best plan fitting for each defined specific patient and specific procedure to be implemented based on the previous experience with similar procedure. This will allow for higher degrees of autonomy of the function of surgical robots in spinal deformity.

[0051] The novelty and simplicity of the current invention is the result of sing a digitized device or apparatus to execute the correction of spinal deformity, the operative plan will be a known digitized set of commands prepared to be received by the automated screwdriver, robotic screwdriver, or robotic system. For the correction of each degree of freedom the plan will include 1 binary digital command for the direction of rotation of the head of the screw of the joint mechanism. And 2 digital or numerical commands 1 for the amount or degree of rotation of the screw head (in degrees) and the other for the torque applied to the screwhead (in Newtonmeter). The main advantage of this system is that it does not require complex Surgical Robots or complex operative instructions and commands to achieve high degree of automation of execution of surgical action. An automated programmable screwdriver or a readily available robotic screwdriver or a Collaborative robot used in some industry can be optimized and certified according to the surgical tools and robotic regulations to perform these functions. Regarding the operativeinstructions or commands, a vertebra can be deviated in a maximum of 6 degrees of freedom (usually less); the operative instructions incorporate 3 digital entries (one binary 0 and 1) for each degree of freedom; therefore, the maximum number of instructions required to correct any motion segments controlled by a Coupler will be 18 digits (with 6 binary digits). This is a simple straightforward operative command that will be used to execute one of the most complex surgical procedures in spine surgery (spinal deformity correction).

[0052] Many of the Robotic Screwdrivers and Collaborative Robots (COBOT) readily available and used in the industry have many of the required capability to operate the Spinal Deformity Correction Couplers or devices. They have: - Robotic arm: A multi-jointed arm that positions and moves the screwdriver - Screwdriver attachment: A screwdriver that can be standard or custom-made for specific application. - Vision system: Cameras and sensors that track the screwdriver's position and alignment - User interface: A way to program parameters like torque settings and movement sequences - Safety measures: COBOTS are equipped with sensors and collision detection technology to operate safely near workers - Integrated machine learning software to analyze large amounts of data learn from their experiences experience to Improved efficiency and effectiveness and make predictions.

[0053] Referring to FIG 12, to generate the spinal surgery plan and train the machine learning model 1201, data is needed from multiple sources to establish a baseline machine learning model that is then trained over time to provide improved patent specific outcomes from the generated spinal surgery plans. FIG 12 illustrates an operational flowchart for gen-erating a spinal surgery plan based on processing pre-operative patient data through a machine learning spine model 1201, and for using intraoperative feedback data and / or post-operative feedback data to train the machine learning spine model 1201.

[0054] Referring again to FIG. 23, before starting a spinal deformity case, pre-operative data is provided and received by the system. The data include, without limitation, any one or more of Patient Data 1202 including: A. Personal Data 1203 date of birth, monarchal status; skeletal maturity status, height; weight / BMI; gender; race; bone Density. B. Disease Data 1204 including, associated muscular disease, connective tissue disease, neurological disease, skeletal disease, ambulation status. C. Spinal Deformity Data 1205 including etiology of the deformity, stiffness and flexibility index, presence of congenital malformation, associated intraspinal anomalies. D. Imaging data 1206 including I. Segmental Measurements: including A. Intervertebral Rotational Deviations, Intervertebral Axial Rotation (Z / cephalic-caudal axis) including Apical Axial Vertebral Rotation, Intervertebral Coronal Angulation (Y / anterior-posterior axis): Intervertebral Wedge Angle, Intervertebral Sagittal Angulation (X / right-left / medial-lateral axis): Lordosis & Kyphosis Angles. B. Intervertebral Linear Deviations including Disc Height (Z axis), Antero / Retro Listhesis (Y axis), Lateral Listhesis (X axis). II. Regional Spinal Measurements including A. Coronal: Main Coronal Curve / s COBB, Fractional Curve COBB, Secondary coronal curve / s COBB. B. Sagittal: Thoracic (T1-T12) & (T4-T12), Lumbar (Ll-Sl), Total Kyphosis. C. Axial: Superior Axial Intervertebral Rotation (Superior AIR), Inferior Axial Intervertebral Rotation (Inferior AIR), Torsion Index (TI). IILSpinopelvic and global measurement including Upper Instrumented Vertebra Pelvic Angle (UIV PA), Pelvic Incidence: PI, Pelvic Tilt: PT, Sacral Slope: SS, Pelvic Incidence Lumbar Lordosis Mismatch: PI-LL, Sagittal Vertical Axis: SVA, Tl Spinopelvic Inclination: Tl SPi, Tl Pelvic Angle: TP A, Coronal Balance: Coronal Balance, C2 Tilt, C2 Pelvic Angle (C2 PA), LI Pelvic Angle (LI PA), T4 Pelvic Angle (T4 PA).

[0055] The previous data is a exemplary and the list can include any data that can affect the surgical procedure and the automated correction process. All the data related to adefined patient will be received by the Machine Learning Model 1201 shown in FIG. 12 together with the data characterizing the Correction apparatus 1207 and the normal alignment data 1208 to generate the initial preoperative plan 1209 or operative instructions for the automated machine.

[0056] Various embodiments of the present disclosure solve the above problems by providing a surgical procedure planning system and method including multiple feedback loops for optimizing future surgical preoperative plans. Various embodiments of the present disclosure provide a system and method that utilizes the feedback of different stages of surgery either preoperative, intraoperative, postoperative and at follow up for development of future improved preoperative plans for individual patients. The system stores all generated operative plans to select the most optimal plan based on the patient outcomes that resulted from its use. Usually the best sets operative instructions or plans are those that needed the fewest revisions and adjustments during surgical procedure and that resulted in the best patient reported outcomes. In certain such embodiments, the system and method use suitable machine learning to optimize and determine the m o s t o p t i m u m preoperative planning for subsequent patients who share specific common features with a defined patient.

[0057] The present disclosure provides one example of a robotic or robotically assisted surgical procedure planning system and method with multiple feedback loops for automatically optimizing current and future automated, robotic, or robotically assisted surgical preoperative plans related to surgical correction of spinal deformity. FIG. 12 describe an example of the workflow for generating operative plans and different feedback loops for a defined patient having surgery to correct spinal deformity. The first 1213, second 1214, third 1215 feedback loops will improve the current surgical procedure and improve the operative plan or instructions intraoperatively for the defined patient as well as to train the MLM. The fourth 1216,fifth 1217, sixth 1218 feedbacks are intended to train the Machine Learning Model for optimizing future plans for patients sharing common characteristics with the defined patient.

[0058] Initial Preoperative Plan 1209: The system and method generate operative plan, such as intraoperative set of commands or instruction for the operation of the correction device, based on automated measurements and calculations, to achieve the targeted corrections for vertebral deviations and to align the vertebrae. The preoperative plan includes 3 main commands or instructions for each joint mechanism controlling one degree of freedom: a. the direction of rotation of screwhead b. magnitude of rotation of the screwhead c. the torque applied to rotate the screwhead in order to manipulate the deviated vertebra. In multi-axial vertebral deviations, and in multilevel spinal deformities the system can additionally generate some commands related to the sequence or order for operating multiple screwheads controlling multiple joint mechanisms incorporated in the device as well sequence of correction of multiple correction apparatuses. Based on the received patient data 1202 including personal data 1203, disease data 1204, deformity data 1205, and imaging data 1206 as well as correction apparatus characteristics data 1207 in addition to normal alignment data 1208 for the defined patient; he system and method of the present disclosure generates an electronic initial preoperative plan 1209 for a specific surgical surgical deformity correction and create the specific instructions planned for the automated machine that operates a correction apparatus prepared for a defined specific patient (Initial Plan 1209 ). The system and method then electronically provide or releases this selected Initial Plan 1209 to be available for the surgeon in order to check the plan before sending it to the automated machine.

[0059] First Feedback 2313 and Secondary Plan 1210: The surgeon will receive the initial plan 1209 and through a simulation software 1219 will virtually executethe plan to check its outcome before the surgical procedure. With more training, this step can also be automated without human input, and the initial plan is automatically checked and adjusted by a simulation software 1219. Based on the preoperative adjustment a first feedback 1213 is received by the machine learning model to generate a secondary plan 1210.

[0060] Second Feedback 1214 and Tertiary Plan 1211 (Intraoperative Pre-Correction): After initiating the surgical procedure for the defined patient and before the start of the actual automated or robotic execution of deformity correction; the system and or surgeon electronically collects certain individual patient related data found during the initial parts of the surgical procedure intraoperatively brfore executing any correction. These data may include intraoperative prone deformity radiographic measurements, the configuration of the device after attachment to the deviated vertebrae, position of the pedicle screw, quality of bone, Interpedicular Screws Distances (Second Feedback 1214). These additional data will be sent automatically to the system for further adjustments of the secondary plan 1210 and update; this may cause changes to the secondary preoperative plan 1210 creating a tertiary intraoperative plan 1211. This tertiary more accurate updated plan 1211 will be accordingly implemented during the rest of the actual surgical procedure on the defined patient in the operating room. The tertiary preoperative plan 1211 is also automatically sent for storage in the system to be used as part of a preoperative plan set to create or design more accurate or enhanced future initial preoperative plans for a subsequent patient sharing one or more specific common features with the defined current patient.

[0061] Third Feedback 1215 and Quaternary Plan 1212 (Intraoperative Post Initial Correction): When the tertiary plane is received by the automated screwdriver or robot, the correction is executed based on the sequence of the operative instructions and the digital commands for each of the joints. Once the correctionis executed the system the system and or surgeon electronically collects certain correction related data including post manipulation radiographic measurements, the configuration of the device after correction of vertebral deviation, any intraoperative difficulties or adverse effects during or after the correction maneuvers intraoperative post initial correction feedback. These additional data will be sent automatically to the system as a Third Feedback 1215. The system receiving this information and comparing it with the target correction parameters and accordingly, if needed, it will automatically generate an adjustment plans with a set of new adjustments on the current executed plan (Quaternary Plan 1212) this will be the last feedback that will benefit the current patient as well as train MLM for future plan creation.

[0062] Fourth Feedback 1216: After the final correction and before finishing the operation an Intraoperative post final correction feed back 1216 is generated and received by the Machine Learning Model.

[0063] Fifth Feedback 1217 (Postoperative): After the actual surgical procedure for the individual patient is completed, the system and method receive postoperative feedback (Fifth Feedback 1217) data from the patient regarding the outcome of the surgery and the actual correction of spinal deformity. This information includes wide range of different postoperative data such as postoperative local, segmental, regional, global spinal and spinopelvic realignment; Patient Reported Outcome Measures, and post operative adverse effect. The system and method of various embodiments of the present disclosure includes providing this further received patient result information in a third automated electronic feedback loop to enable the system and method to create more precise future initial for subsequent patients who share common features with the current defined patient.

[0064] Sixth Feedback 1218 (At follow up): Another feedback form of the patient data especially the radiographic measurements, maintenance of alignment, clinical outcome, and complications can be received by the system after a specific postoperative duration during follow up (6 months or 12 months for example). This information will detect late implant failure, loss of correction, junctional problems (Sixth Feedback 1218). The system and method of various embodiments of the present disclosure includes providing this further received patient result information in a third automated electronic feedback loop to enable the system and method to create more precise future initial for subsequent patients who share common features with the current defined patient.

[0065] FIG. 13 is a work flowchart illustrating the Machine Learning Model 1301 training and the use of the multiple previous operative plans in generating future operative plans. When a defined patient is scheduled for surgery the patient data 1302 is received by the MLM and these data are compared with previously generated operative plans 1303 for patients sharing common characteristics with the defined patient. The machine learning model will crate an optimized initial preoperative plan of the defined patient based on the most successful stored plans.

[0066] The disclosure of the embodiments of the invention is intended to be illustrative, but not limiting, of the scope of the invention. The correction apparatus means any device fixed to the vertebrae incorporating at least one joint mechanism intended to correct at least one axis of vertebral deviation. It will be understood that various modifications may be made to the embodiments disclosed herein. Those skilled in the art will envision other modifications within the scope and spirit of the claims appended hereto.

Claims

Claims:

1. A system for automation of spinal deformity correction comprising: receiving and processing a correction apparatus characteristics data, normal anatomical spinopelvic measurements and alignment data, and patient data including personal data, disease data, spinal deformity data, and imaging data; predicting, based on normal anatomical spinopelvic measurements and alignment data, relative to the patient data, the optimal target position and orientation of the vertebrae; generating, based on the patient imaging data relative to the predicted optimal vertebral position and orientation, the required corrections of vertebral deviations and intervertebral distances to realign the spine; receiving and processing the patient data, the correction apparatus characteristics data, and the required correction data; providing, using a machine learning model and algorithms, based on the received and processed data, operative instructions or preoperative plan to an automated machine that connects with and operates the correction apparatus attached to the spine to enable the automated correction of vertebral deviations of the targeted vertebrae and realign the spinal deformity2. The system of Claim 1, wherein the correction apparatus means a correction device incorporating at least one joint mechanism and intended to correct at least one axis of vertebral deviation.

3. The system of claim 1, wherein the correction apparatus characteristic data include joint mechanism type, locking type, configuration type and input output ratio of at least one joint mechanism incorporated within the correction apparatus.

4. The system of claim 1, wherein the imaging data include preoperative and or intraoperative images of the spine, pelvis and lower limbs including plain radiographs, CT scan, MRI, biplanar radiographs, fluoroscopy, machine vision, and augmented reality.

5. The system of claim 4, wherein the preoperative imaging data include patient’ s preoperative prone position images.

6. The system of claim 1, wherein the spinal deformity is divided into multiple motion segments, and further wherein measurement of the deformity is based on individual vertebral deviations and intervertebral distances.

7. The system of claim 1, wherein the generated required corrections include correction of at least one more deviated vertebra relative to a less deviated vertebra, further wherein the more deviated and the less deviated vertebrae being adjacent or non-adjacent.

8. The system of claim 1, wherein the operative instructions generated to operate the correction apparatus include direction of rotation of screwhead, magnitude or degree of rotation of screwhead, and torque applied to rotate the screwhead of at least one joint mechanism of the correction apparatus.

9. The system of claim 1 , wherein the automated machine includes an automated motor, automated motorized screwdriver, robotic screwdriver, robotic arm, collaborative robot, or autonomous robotic system.

10. The system of claim 1 , wherein the automated machine connects to and drives the screwhead or actuator of at least joint mechanism of the correction apparatus to mobilize at least one vertebra in a precalculated direction, magnitude, and force in order to correct the vertebral deviation and realign spinal deformity.

11. The system of claim 1 , wherein the correction apparatus can be motorized by incorporating an internal motor, further wherein the internal motor will be programmed to receive automated instructions from a remote controller held by an operator or surgeon to operate the correction apparatus while the patient is awake at an outpatient setting.

12. The system of claim 1, wherein the automation of surgical correction of spinal deformity can be partial automation, full supervised automation, or full unsupervised autonomous automation.

13. The system of claim 1, wherein vertebral deviations Eire measured inraoperatively using tracking sensors, gyroscope, accelerometer, or cameras to allow detection of any change in vertebral position and orientation during the surgical procedure to give feedback to the system to generate adjusted and updated intraoperative instructions to the automated machine operating the correction apparatus.

14. The system of claim 1 wherein the machine learning model and algorithms generate a preoperative plan or instructions to operate a correction apparatus to correct spinal deformity of a defined patient based on received pre-operative and intraoperative data of the defined patient, normal anatomical alignment data, and correction apparatus characteristics data.

15. The system of claim 1, wherein the system based on the received defined patient data, normal anatomical data, and at least one correction apparatus characteristic data and through machine learning model and algorithms generates a preoperative plan for the defined patient including direction of rotation of screwhead, magnitude or degree of rotation of screwhead, and torque applied to rotate the screwhead of at least one joint mechanism of the correction apparatus.

16. The system of claim 1, wherein the system based on the received defined patient data, normal anatomical data, and at least one correction apparatus characteristic data and through machine learning model and algorithms generates a preoperative plan for the defined patient including planned number and location of the correction apparatuses, number and direction of at least one joint mechanism incorporated in each correction apparatus, and sequence of correction of different joint mechanisms and different correction apparatuses.

17. The system of claim 14, wherein the system uses algorithms for analysis of data and training of a model to improve the preoperative plan using Machine Learning, Self-Learning, Artificial Intelligence, and Neural Networks.

18. A system of Claim 1 , wherein the machine learning model generates a future surgical preoperative plan for automated spinal deformity correction comprising: receiving prior to the surgical procedure, information regarding defined correction apparatus characteristics data, normal anatomical alignment data, and defined patient data including personal data, disease data, spinal deformity data, imaging data; providing an electronic digital initial preoperative plan or instructions for an automated machine to operate at least one correction apparatus; receiving first feedback information preoperatively from a surgeon after checking the correction outcome of the initial operative plan using a simulation software, executing algorithms of a machine learning model to create and provide the automated machine with a secondary intraoperative plan;receiving second feedback information regarding spinal deformity, spinal implants, correction apparatus during surgery after attaching the correction apparatus to the implant anchored to the spine and before executing correction of vertebral deviations; executing algorithms of a machine learning model to create and provide the automated machine with a tertiary intraoperative plan; receiving third feedback information regarding the spinal deformity, spinal implants, correction apparatus during surgery after executing the initial correction based on the tertiary intraoperative plan; executing algorithms of a machine learning model to create and provide the automated machine with a quaternary intraoperative plan; receiving fourth feedback information regarding the spinal deformity, spinal implants, correction apparatus during surgery and after executing the final correction based on the quaternary intraoperative plan; receiving fifth feedback information regarding the spinal deformity, spinal implants, correction apparatus and patient outcome after surgery; receiving sixth feedback information regarding the spinal deformity, spinal implants, correction apparatus and patient outcome at follow up; storing the patient data, operative plans, and surgical feedback information data for use in training the machine learning model to create optimized preoperative plan for a subsequent patient who shares common feature with the defined patient.

19. The system of claim 18, wherein the system further comprises: obtaining intra-operative feedback data and / or post-operative feedback data regarding spinal deformity surgery outcome for a plurality of patientstraining a machine learning model based on the intra- operative feedback data and / or the post-operative feedback data, obtaining pre-operative patient data of a defined patient and correction apparatus characteristics data; generating a spinal surgery plan for the defined patient based on processing the pre-operative patient data of the defined patient relative to the feedback data of the plurality of patients shaing common characteristics with the defined patient through the machine learning model, and providing pre-operative pain instructions for a correction apparatus to correct spinal deformity of the defined patient.

20. The system of claim 19, wherein the system receives data of a multitude of patient specific data including personalized normal anatomical vertebral realignment data to generate the operative instructions to correct spinal deformity using a correction apparatus to achieve patient specific alignment of the spine for a defined patient

Images