WO2013106430A1 - Method and system for automated dental implantation - Google Patents

Abstract

An automated dental implantation method and system using an image-guided robotic system in order to insure the accurate insertion of a dental implant within a patient's jaw bone. Patient-specific 3D models are reconstructed from preoperative Cone-beam CT images and implantation planning is performed with the virtual models. A two-step registration procedure is applied to transform the preoperative plan of the implant insertion into intra-operative operations of a robot. The two-step registration process includes registration between a virtual coordinate system and a reference coordinate system and registration between the reference coordinate system and an operation coordinate system in order to avoid direct contact between the robot and the patient during the setup stage, thus ensuring the safety of the patient.

Description

METHOD AND SYSTEM FOR AUTOMATED DENTAL IMPLANTATION

CROSS-REFERENCE TO RELATED APPLICATION

This Application claims priority to U.S. Provisional Patent Application No. 61/584,594, filed January 9, 2012, entitled METHOD AND SYSTEM FOR AUTOMATED DENTAL IMPLANTATION, the entirety of which is incorporated herein by reference.

FEDERALLY SPONSORED RESEARCH

N/A

FIELD OF THE INVENTION

The invention relates generally to the field of automated dental implantation methods.

BACKGROUND

Dental implantation is now recognized as the standard of care for tooth replacement for both the functional and aesthetic results that it provides. An important factor affecting the outcome of implantation is the successful insertion of the implant into the patient's jaw bone, which requires a high degree of anatomical accuracy. Computer and robotic techniques have been applied in many surgical applications including dental implantation. These techniques have succeeded in optimizing preoperative planning and have improved the quality of intra-operative performance. However, current computer and robotic implantation methods leave significant room for human error in the human drilling process. The manual controlling of space and heat parameters is hardly precise. Additionally, patients may have varying amounts of available bone in which to place the implant. For example, a minimum of 2 mm around the implant is required for support. These requirements for spacing suggest that sub-millimeter accuracy is necessary.

Computer-assisted technology has been applied in the field of dental implantation for both preoperative planning and intra-operative operation. An image-guided surgical system generates a preoperative surgical plan with the help of patient-specific medical data, i.e., for two- dimensional and three-dimensional models. For the intra-operative stage, two different principles are applied to transfer the virtually planned surgical scheme to actual surgical operation. These are navigation, or dynamic tracking, and guide. The navigation of dynamic process tracks the intra-operative position of the surgical tool with an optical or magnetic tracking system, which is superimposed onto the virtual model of the patient's anatomic structures and shown to the surgeon on a screen. The guide technique generates a drill guide (template) according to the virtually planned implantation position, which can be physically accessed by the surgeon, therefore navigation is not necessary. However, in both strategies, the surgical plan is executed manually by a surgeon. Thus the surgery outcome depends on the skill of the surgeon and potentially leading to unstable and unsafe results.

Therefore, there is a need for an effective robot-assisted dental implantation system, in which the execution of the surgical plan is done by a robot automatically but under the supervision of the surgeon.

SUMMARY OF THE INVENTION

The methods and devices described herein provide an image-guided robotic system for automated dental implantation. Patient-specific 3D models are reconstructed from preoperative cone-beam CT images and implantation planning is performed using these virtual models. A two-step registration procedure is applied to transform the preoperative plan of the implant insertion into intra-operative operations of the robot with the help of a Coordinate Measurement Machine (CMM). Experiments are carried out with a phantom which is generated from the patient-specific 3D model. Fiducial Registration Error (FRE) and Target Registration Error (TRE) values are calculated to evaluate the accuracy of the registration procedure.

In one aspect of the invention, a computer-assisted dental implantation method is provided. The method includes constructing a patient-specific model of a patient's jaw bone, developing an implementation plan utilizing the patient-specific model, performing a two-step registration process to transform the implementation plan into intra-operative operations, and operating a robot to perform the intra-operative operations on a patient according to the implantation plan In another aspect of the invention, a dental implantation system is provided. The dental implantation method includes a model generation module for reconstructing a patient-specific 3D model of a patient's jaw bone, a planning module for generating an implementation plan in accordance with the patient-specific 3D model, a computer-controlled robot for performing a dental implant procedure on a patient in accordance with the implementation plan, and a registration module for registering the implementation plan to the dental implant procedure.

Although methods and devices similar or equivalent to those described herein can be used in the practice or testing of the present invention, suitable methods and devices are described below. All publications, patent applications, and patents mentioned herein are incorporated by reference in their entirety. In the case of conflict, the present specification, including definitions, will control. The particular embodiments discussed below are illustrative only and not intended to be limiting. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram of system constructed in accordance with the principles of the present invention;

FIG. 2 is a diagram illustrating the relationship among coordinate systems in accordance with the principles of the present invention;

FIG. 3 is a flowchart illustrating an exemplary process performed by the automated dental implantation system of the present invention;

FIG. 4 illustrates a patient's jaw model with fiducials and fixed registration points according to the principles of the present invention;

FIG. 5 is a table showing the measured coordinates of the fiducials and fixed registration points in accordance with the principles of the present invention; and

FIG. 6 is a table showing the calculated coordinates in the two-step registration process of the present invention.

FIG. 7 is a table showing a comparison of registration results before and after reference coordinate fixation, and coordinate system orientation pre-alignment; and

FIG. 8 is a table showing the orientation error after registration.

DETAILED DESCRIPTION

The present invention is directed towards an automated dental implantation method and system using an image-guided robotic system in order to insure the accurate insertion of a dental implant within a patient's jaw bone. Patient-specific 3D models are reconstructed from preoperative Cone-beam CT images and implantation planning is performed with the virtual models. A two-step registration procedure is applied to transform the preoperative plan of the implant insertion into intra-operative operations of a robot with the help of a Coordinate Measurement Machine (CMM). The two-step registration process includes registration between a virtual coordinate system and a reference coordinate system and registration between the reference coordinate system and an operation coordinate system. Thus, instead of registering the virtual coordinate system directly with the operation coordinate system, the two-step registration procedure of the present invention is designed to transform the preoperative plan of the implant insertion into parameters for intra-operative operation. This is done in order to avoid direct contact between the robot and the patient during the setup stage, thus ensuring the safety of the patient.

The present invention is a robot-assisted dental implantation system and method, in which the execution of a surgical plan is done by a robot automatically but under the supervision of a surgeon. The method and system of the present invention can be used with any dental implant including the dental implant described in the provisional application entitled "Method and Device for Natural Root Form Dental Implants", under attorney docket number 7967-60, being filed concurrently with the present application. The present invention improves upon the "guide"-type of image-guided dental implantation of the prior art by providing a laboratory robot that automatically drills a dental implant site directly in the patient's jaw bone. A virtual plan can be created and transferred to the patient site thus eliminating errors that may occur while transferring the virtual plan first to the drill guide and then to the patient. Errors caused by the surgeon such as misalignment of the drill guide, tremor, or identification error can also be avoided. Therefore, a more accurate and reliable execution of the surgical plan can be expected.

Referring now to the drawing figures in which like reference designators refer to like elements, there is shown in FIG. 1 a diagram of an exemplary embodiment of system 10 of the present invention. System 10 includes two phases, a pre-operative phase 12 and an intraoperative phase 14. In the preoperative phase 12, a model generation module 16 uses images of a patient to reconstruct a patient-specific three-dimensional geometrical model.

Patient images can be created using known imaging techniques such as cone-beam computer tomography ("CT"), electroencephalography ("EEG"), magnetoencephalography ("MEG"), magnetic resonance imaging ("MRI") and electrocardiography ("EKG"). In a non- limiting example, CT is used throughout this application with the understanding that any imaging technique may be used. Fiducials can be attached to the patient's jaw bone when the cone-beam CT scan is taken and used as points of reference. The resulting 3D geometrical model is a jaw model of the patient, with fiducials attached.

The 3D geometrical model of the patient's jaw is then loaded into a planning module 18, which could be a software program that generates a dental implantation plan 20 for the patient, which can be viewed on a computer screen. The dental implantation plan provides the user, i.e., physician or the manufacturer of the dental implant, with a drill guide template. The template includes virtual implantation positions that enables a computer-guided robot to insert the dental implant within the patient's jaw during the operational phase. Using the visualized

implementation plan 20, a dental surgeon can virtually insert a dental implant into model 16 and the parameters of the implementation plan 20 can be saved. Plan 20 may include such parameters as implant size, the target coordinate of the insertion and the orientation of the insertion.

A fiducial is an object used in the field of view of an imaging system and that appears in the image produced. It is typically used as a point of reference. Fiducials can be attached to the jaw bone or teeth of the patient during an initial visit and prior to a CT image being taken. Thus, any fiducials used on the patient and that appear in model 16 are identified and their coordinates are saved in a virtual image coordinate system for the registration process.

After the registration process (to be discussed in greater detail below), which determines the relationship between the preoperative image data and the intra-operative physical body of the patient, coordinates of a set of points are measured in both a reference coordinate system and an operation (robot) coordinate system. This occurs during the intra-operative phase 14. Thus, as shown in FIG. 1, the model generation module 16 and the planning module 18 are used to design the dental implant insertion by generating the implementation plan 20. A reference coordinate system, such as the coordinate system of Coordinate Measurement Machine ("CMM") 22 is used to register the implementation plan 20 to the intra-operative operation of a computer-controlled robot 24. This process obviates the need for the robot 24 to directly touch a patient 26 before actual surgery. Finally, the automated robot 24 conducts the drilling operation for the dental implant insertion according to the implementation plan 20.

As discussed above, the registration process determines the relationship between the preoperative image data (i.e. implementation plan 20) and the intra-operative physical body of the patient 26. The registration process is conducted to obtain transformation matrices between two different coordinate systems, i.e., a virtual coordinate system, and an operation coordinate system. The registration process aligns different sets of data obtained from different objects, i.e., the patient's anatomy, medical images, and operation tools and/or at different times.

The present invention utilizes a two-step registration process in order to register data resulting from the preoperative phase 12 to the intra-operative phase 14. FIG. 2 illustrates an exemplary registration process 28 used by an embodiment of the present invention. Three different coordinate systems are involved: a virtual coordinate system 30, a reference coordinate system 32, and an operation coordinate system 34. In one embodiment, the reference coordinate system 32 is that of the Coordinate Measurement Machine ("CMM") 22. A CMM 22 is a maneuverable robotic arm having a tip for measuring the position, coordinates or orientation of a specific location. Other measuring devices such as an optical or magnetic tracker can be used in place of the CMM 22.

The use of the reference coordinate system 32 results in a two-step registration process in order to avoid directly touching the patient with the robot 24 prior to surgery. The two-step registration process of the present invention is not limited to a particular registration technique. For example, paired-point registration techniques may be used. Paired-point registration aligns a set of N points (N>3) in the first coordinate system with another set of corresponding points in the second coordinate system, using a one-to-one mapping. Paired-point registration may be used when artificial fiducials or a tracking probe is used to physically contact each of these points.

Referring once again to FIG. 2, the virtual coordinate system 30 refers to the coordinate system of preoperative patient-specific medical data 36, such as a 3D geometric model reconstructed from 2D images. The operation coordinate system 34 refers to the coordinate system of the surgical tool, which can be a dental drill-bit 38 attached to the robot 24. The reference coordinate system 32 refers to coordinate system of the CMM 22. The present invention is not limited to a particular type of reference coordinate system 32 or robot 24. For example, in one non-limiting embodiment, the reference system is a Gold Faro CMM 22 made from FARO Technologies Inc., and the robot 24 is a commercial robot from Mitsubishi®, having six degrees of freedom ("DOF") with a position repeatability of ±0.02 mm.

Reference coordinate system 32 acts as a bridge which connects the virtual coordinate system 30 with the operational coordinate system 34. In the operation coordinate system 34, a dental drill-bit 38 is attached to the end-effector of the operational arm of robot 24. The relative position between the tip of the drill-bit and the end-effecter of the robot is calibrated. Therefore, when the robot 24 moves, the coordinate of the drill-bit tip changes accordingly.

In the two-stage registration process shown in FIG. 2, a set of preoperative fiducials 31 is attached to the patient during the preoperative phase 12. In another embodiment, the patient's teeth may be used as anatomical registration points. In order to insure the safety of the patient 26, the coordinates of the fiducials 31 attached to the patient are measured by the CMM 22 and not by the robot 24. In the intra-operation phase 14, after the patient's head is fixed on a surgical table or chair, the fiducials, now considered to be intra-operative fiducials 33 are measured by the CMM 22. In this way, the relationship between the virtual coordinate system 30 and the reference coordinate system 32 is initialized.

In order to register the preoperative phase 12 to the intra-operative phase 14, a second registration step is required to register the reference coordinate system 32 to the operation coordinate system 34. Because the robot 24 is designed to avoid contact with the fiducials 33, a second set of registration points 35 fixed in the operation (robot) coordinate system 34 is used to conduct this registration. For example, if the coordinate of a target point is designated as tv in the virtual coordinate system 30, the two-step registration process can be described as follows: Step 1 : registration between the virtual coordinate system 30 and the reference coordinate system 32 is performed based on one set of points (fiducials) 31, which are attached to the patient 26 when cone beam CT images are taken and remain in the same position until the CMM 22 records their coordinates. The transformation from the virtual coordinate system 30 to the reference coordinate system 32 is determined (denoted as TV2R) and the coordinate of the target point in reference coordinate system 32 can be calculated as tR = TV2R(tv ). Step 2: registration between the reference coordinate system 32 and the operation coordinate system 34 is done based another set of points (fixed registration points) 35 and the transformation from reference coordinate system 32 to operation coordinate system 34 is determined (denoted as tR20). The coordinate of the target point in the operation coordinate system 34 can then be calculated as tO = TR20(tR) = TR20(TV2R(tv)).

FIG. 3 is a flowchart illustrating the steps performed by an embodiment of the present invention. In the pre-operative phase 12, a patient-specific model is reconstructed, at step S40. Any fiducials 31 attached to the patient's jaw bone are attached at this time, at step S42. Using planning software, a pre-operative implementation plan is generated, at step S44. Fiducials 31 attached to the patient in order to reconstruct the patient-specific model are identified, at step S46. Data generated during the pre-operative phase 12 is then registered to the intra-operative phase 14 at step S48 and the robot 24 is directed to drill the implant site hole in the size and shape of the dental implant, at step S50, according to the implantation plan. The dentist or surgeon can then emplace the implant into the robotically drilled hole.

The results of the point-based registration process can be quantitatively evaluated by a number of indicators. For example, an indicator such as Fiducial Localization Error ("FLE") can be used to determine the error in locating the fiducial points. It cannot be calculated directly because the actual position of a fiducial 31 is unknown. Fiducial Registration Error ("FRE") indicates the distance between fiducial points in one coordinate system and their corresponding coordinates in another coordinate system, after registration. It is usually measured by the Root Mean Square ("RMS") value of the points set. Target Registration Error ("TRE") is this distance for a target point after registration. The "target" refers to the point whose position needs to be determined for a certain medical operation. As with FLE, indirect methods are used to estimate the value because the actual coordinates of the target is unknown. In one embodiment, a "predetermined optimal value" is chosen for the target coordinate and the TRE calculated accordingly.

In one embodiment, when actual fiducials 31 are not attached to the patient, a plurality of fiducials 52 can be virtually generated and attached to the patient's jaw model. As shown in FIG. 4, five virtually generated fiducials 52 are attached to a patient's jaw model to mimic the bone fiducials 31. Small semi-spheres, for example having a diameter of around 1 mm, can be used as the attached virtual fiducials 52, although any shape may be used. The semi-sphere shape is not easily broken during the printing process. The fiducial coordinates are measured in the virtual coordinate system 30 by identifying the apex of each semi-sphere. In the reference coordinate system 32, the fiducial coordinates are measured by touching the apex of each semi- sphere with the tip of the pointer. In order to minimize localization error, multiple measures are made for each fiducial position, and their mean value is recorded as the final coordinate for a given fiducial 52.

In FIG. 4, an additional set of virtual points can be defined by marks on the patient's jaw model. In FIG. 4, eight virtual registration points 54 are shown for the second registration step (from the reference coordinate system 32 to the operation coordinate system 34). The coordinates of these registration points 54 can be measured in both the reference coordinate system 32 and the operation coordinate system 34. In the operation coordinate system 34, the coordinates are measured by commanding the robot 24 to move along the X, Y and Z axis, until the tip of the drill-bit 38 attached to it reaches the apex of the semi-sphere. It should be noted that choice of five virtual fiducials 52 and eight virtual registration points 54 can vary and it is within the spirit of the invention to use any number of virtually generated fiducials 52 and fixed registration points 54 in order to provide optimal results.

The present invention is further illustrated by the following specific examples. The examples are provided for illustration only and should not be construed as limiting the scope of the invention in any way.

Data was recorded for the two-step registration procedure described above using the fiducials 52 and fixed registration points 54 shown in FIG. 4. The FRE for registrations between the virtual coordinate system 30 and the reference coordinate system 32, and between the reference coordinate system 32 and the operation coordinate system 34 was calculated. The FRE is the root-mean-square in fiducial alignment between the virtual coordinate system 30 and the reference coordinate system 32. To evaluate the target registration error, one of the set of fiducials 52 was designated as the "target" and the registration process was conducted using the remaining four fiducials 52. After registration, the coordinate for each target point in the operation coordinate system 34 was calculated. Accordingly, the robot 24 was commanded to move the tip of the drill-bit 38 to the target coordinate and hold its position there.

One of the five fiducials is then designated as the "target" fiducial. This target fiducial represents the location in the patient's mouth where the actual drilling will take place. Because one of the five fiducials (the five semi-spheres) 52 is designated as the target, the designated position of the target in the operation coordinate system 34 can be measured directly by the robot 24. Therefore, with respect to the TRE estimation described above, this measured coordinate in the operation coordinate system 34 is defined as the "predetermined optimal value" while the current coordinate of the drill-bit tip 38 was recorded as a "registered value". After the two-step registration, the TRE value was calculated as the difference between the predetermined optimal value and the registered value.

Similarly, TREs for each pair of two-coordinate system registration were also recorded. The coordinates of the fiducials 52 and the fixed registration points 54 were measured (in millimeters, 10 measures per point) and are listed in the table shown in FIG. 5. The table of FIG. 5 contains fiducial coordinates (measured by the robot 24 for experimentation purposes but would not normally be taken in an actual operation procedure.

FIG. 6 shows the standard deviation ("STD") values of ten measures for each target point. For the reference coordinate system 32 and the operation coordinate system 34, the standard deviation of 10 measures has a mean value of 0.04 mm and 0.15 mm (N=13), respectively. The results indicate that the CMM 22 and the robot 24 provide high localization accuracy. Further, the measurement error of the CMM 22 and the robot 24 were tested by comparing their readings with known distances. The results shown a mean error of 0.03 mm (N=26) and 0.07 mm (N=l 1) for the CMM 22 and the robot 24, respectively.

The localization error in the image space is relatively large when compared with the error in the physical space. It is shown that the standard deviation of measurement in the virtual coordinate system 30 ranges from 0.21-0.35 mm, with a mean value of 0.26 mm (N=5). A reason for this may be the design of the fiducials 52. In one embodiment, the fiducials 52 used were five semi-spheres each having a diameter of 1 mm, and the identifying point for each of the fiducials 52 is the top point or apex of the semi-sphere.

Other actions can be taken to minimize TRE values including fixation of the CMM 22 to the worktable of the robot 24 to avoid disturbance from unexpected movement and vibration. Other options are increasing the number of points for registration and utilizing an optimized calibration block for obtaining the transformation between the reference coordinate system 32 and the operation coordinate system 34. By combining some or all of these efforts, a sub- millimeter accuracy can be achieved.

The table shown in FIG. 6 provides the resulting coordinates in the two-step registration procedure of the present invention. The TRE value for each target point was calculated by computing the distance between the calculated and the measured coordinates of the target point in the same coordinate system. For each step of paired-point registration, the Root Mean Square (RMS) value of the difference between measured coordinates and their corresponding coordinates after registration was calculated as the FRE. In the first step of registration, the four fiducials 52 other than the fiducial designated as the target were used to conduct the paired-point registration between the virtual coordinate system 30 and the reference coordinate system 32. This resulted in five FRE values for step 1 of the two-step registration process, any of which corresponds to one target.

There is only one FRE value for step 2 of the two-step registration process because the registration between the reference coordinate system 32 and the operation coordinate system 34 is done by using the set of fixed registration points 54, with none of the targets included.

However, TRE values calculated by setting one of the fiducals 52 as the target point are also listed in the table shown in FIG. 5. The mean and standard deviation of TRE values after the first step are 1.08 mm and 0.72 mm, respectively. The TRE values after the second step of the registration process is the final TRE of the registration process. These TRE values for the five targets range from 0.74 mm to 2.29 mm. The mean value of TREs equal 1.42 mm and the standard deviation is 0. 70 mm.

The first step of the two-step registration process was conducted with one of the five fiducials 52 set as the target. The result was five registration errors (defined as FRE) recorded in the first step of the registration process. Its value was 0.30±0.08 (Mean±SD) mm, ranging between 0.23 and 0.43 mm. In the second step, the registration was done only once and the registration error was 0.19 mm. As a result of the two-step registration disclosed herein, the tip of the dental drill-bit 38 can be commanded to move to the designated position automatically with a target registration error of TRE = 1.42±0.70 mm (N=5).

From the table in FIG. 6, the TRE values have an expected distribution for both the results from the first step of the registration process and the results after the second step of the registration process where the smallest TRE tends to be the one for fiducial #3 and the TRE becomes bigger as the distance from the current target point to the center of the set of fiducials 52 is increased.

In FIG. 7, a table showing a comparison of registration results before and after the fixation of the reference coordinate system 32, and coordinate system orientation pre-alignment, is shown. One of the problems affecting the registration error might be the relative movement between the reference coordinate system 32 and the operation coordinate system 34. Therefore, a Faro arm or some other compatible device, is affixed onto the same location of the robot 24, so that the spatial relationship of the reference and the operation coordinate system 34 remains rigidly unchanged during the whole procedure. From FIG. 7, one can tell that the final TREs decreased from 1.42±0.70 mm to 1.14±0.61 mm after the fixation of the Faro Arm.

Also, the original orientation of the X-, Y-, Z-axes relative to the phantom model matters to the registration accuracy. Before, we did not reorient the respective CSs prior to performing registration. However, it is important to realign axes of the "from" CS to align with the "to" CS, in order to get a better registration result FIG. 8 is a table showing the orientation error after registration. FIG. 8 lists the results when the coordinate system orientation is pre-aligned. TREs after stepl decreased from

1.09±0.69 mm to 0.24±0.13 mm, and final TREs decreased from 1.14±0.61 mm to 0.38±0.16 mm. Thus, it provided a significant improvement in the registration accuracy.

The present invention provides an image-guided robotic system for dental implantation.

Instead of registering the virtual coordinate system 30 directly with the operation

coordinate system 34, a two-step registration procedure is designed to transform the preoperative phase 12 of the implant insertion into parameters for intra-operative operation. A Coordinate Measurement Machine (CMM) 22 is introduced to act as the reference coordinate system 32 in order to avoid direct contact between the robot 24 and the patient 26 during the setup stage, thus ensuring the safety of the patient. Phantom experiments show the system design is feasible and provide a TRE of 0.38±0.16 mm with an orientation error of 1.99±1.27°.

Any improvement may be made in part or all of the devices, system and method steps. All references, including publications, patent applications, and patents, cited herein are hereby incorporated by reference. The use of any and all examples, or exemplary language (e.g., "such as") provided herein, is intended to illuminate the invention and does not pose a limitation on the scope of the invention unless otherwise claimed. Any statement herein as to the nature or benefits of the invention or of the preferred embodiments is not intended to be limiting, and the appended claims should not be deemed to be limited by such statements. More generally, no language in the specification should be construed as indicating any non-claimed element as being essential to the practice of the invention. This invention includes all modifications and equivalents of the subject matter recited in the claims appended hereto as permitted by applicable law. Moreover, any combination of the above-described elements in all possible variations thereof is encompassed by the invention unless otherwise indicated herein or otherwise clearly contraindicated by context.

Claims

CLAIMS What is claimed is:

1. A computer-assisted dental implantation method comprising:

constructing a patient-specific model of a patient's jaw bone;

developing an implementation plan utilizing the patient-specific model; performing a two-step registration process to transform the implementation plan into intra-operative operations; and

operating a robot to perform the intra-operative operations on a patient according to the implantation plan.

2. The dental implantation method of Claim 1, wherein constructing a patient-specific model of a patient's jaw includes capturing a CT image of the patient's jaw bone.

3. The dental implantation method of Claim 2, wherein the constructing and developing steps are performed during a pre-operative phase and the operating step is performed during an intra-operative phase.

4. The dental implantation method of Claim 3, wherein the two-step registration process comprises:

performing registration between coordinates of a virtual coordinate system and coordinates of a reference coordinate system; and

performing registration between the coordinates of the reference coordinate system and coordinates of an operation coordinate system.

5. The dental implantation method of Claim 4, further comprising attaching one or more fiducials to the patient during the pre-operative phase.

6. The dental implantation method of Claim 5, further comprising measuring, by the reference coordinate system, coordinates of the one or more fiducials during the pre-operative phase.

7. The dental implantation method of Claim 4, wherein the reference coordinate system is that of a coordinate measurement machine (CMM).

8. The dental implantation method of Claim 4, wherein performing registration between the reference coordinate system and the operation coordinate system includes using fixed registration points in the operation coordinate system.

9. The dental implantation method of Claim 4, wherein the reference coordinate system prevents the robot from contacting the patient during the pre-operative phase.

10. A dental implantation system comprising:

a model generation module for reconstructing a patient-specific 3D model of a patient's jaw bone;

a planning module for generating an implementation plan in accordance with the patient-specific 3D model;

a computer-controlled robot for performing a dental implant procedure on a patient in accordance with the implementation plan; and

a registration module for registering the implementation plan to the dental implant procedure.

11. The dental implantation system of Claim 10, wherein the patient-specific 3D model is reconstructed from cone beam CT images of the patient's jaw bone.

12. The dental implantation system of Claim 10, wherein the registration module performs a two-step registration process that registers coordinates of a virtual coordinate system with coordinates of a reference coordinate system and registers the coordinates of the reference coordinate system with coordinates of an operation coordinate system.

13. The dental implantation system of Claim 12, wherein reconstruction of the patient- specific 3D model and generation of the implementation plan occur during a pre-operative phase and the dental implant procedure is performed during an intra-operative phase.

14. The dental implantation system of Claim 13, wherein the model generation module measures coordinates of one or more fudicials that are attached to the patient during the preoperative phase.

15. The dental implantation system of Claim 14, wherein the reference coordinate system is that of a coordinate measurement machine (CMM).

16. The dental implantation system of Claim 15, wherein the CMM measures the coordinates of the one or more fiducials.

17. The dental implantation system of Claim 12, wherein the registration module registers the coordinates of the reference coordinate system with coordinates of the operation coordinate system by using fixed registration points in the operation coordinate system.

18. The dental implantation system of Claim 15, wherein the CMM prevents the robot from contacting the patient during the pre-operative phase.

19. A computer-assisted dental implantation method comprising:

constructing a patient-specific model of a patient's jaw bone;

developing an implementation plan utilizing the patient-specific model; performing a two-step registration process to transform the implementation plan into intra-operative operations, where the two-step registration process comprises:

performing registration between coordinates of a virtual coordinate system and coordinates of a reference coordinate system, wherein the reference coordinate system is a coordinate measurement machine (CMM); and

performing registration between the coordinates of the reference coordinate system and coordinates of an operation coordinate system; and

operating a robot to perform the intra-operative operations on a patient according to the implantation plan.

20. The computer-assisted dental implantation method of Claim 19, wherein performing registration between the coordinates of the reference coordinate system and the coordinates of the operation coordinate system includes using fixed registration points in the operation coordinate system.