Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

The ultimate aim of research related to surgery is to reduce human dependency during operational procedures. The current chapter focuses on such automation efforts undertaken in the field of bone machining. State of the art in the field has been presented. The need for such efforts arises from the fact that involvement of human being in the surgery as an operator requires a lot of experience to develop the precise skills and it always involves a risk of human error. Another important aspect is the constraint on geometry which can be machined in manual operation. With this background, the efforts related to automation and robotics in the field of bone machining are described.

7.1 Computer Aided Planning

There have been many in vitro efforts in case of conventional machining and explorations have begun for non-conventional techniques as well. Some of the efforts help plan the surgery by automated generation of details about bone to be operated on [1–14]. The collected information is then fed to an algorithm which in turn helps to plan the surgery. Extensive information about computer aided planning of orthopedic surgery can be found in a book by Haaker and Werner [15], here only a brief discussion is presented. The main tasks to be executed during planning are listed below [14].

  1. 1.

    Image data acquisition and processing—This consists of gathering the data based on various methods and process this data to generate tomographic information about the bone for tool guidance.

  2. 2.

    Matching—It involves registration of patient’s anatomy with the medical image. The features on actual bone are matched with the image. External feature markers can be introduced before gathering the image data.

  3. 3.

    Referencing—The features identified during matching also serve as reference markers during surgery. Dynamic actions keep the bones constantly under vibrations and the reference points help in tracing exact location of interest.

  4. 4.

    Tool tracking—In order to visualize the real time operation within the medical images, it is necessary to track the location and orientation of the tools. Tracking can be done with the help of infrared LED and with the help of tool features such as tip in case of a drill or a cutting edge in case of saw and chisel.

  5. 5.

    Tool visualization—It is the static or dynamic image formed with the help of gathered data from bone and instrument trackers. Examples for this kind of presentation are sections through interesting regions of the data volume, maximal intensity projections (X-ray simulations), or rendered 3D scenes.

Fig. 7.1
figure 1

Example of CT scan aided surgery planning : The femoral tunnel is positioned so that the cutting miller or drill emerges at the desired point and forms the correct angle of inclination with the condyle (reprinted from Petermann et al. [16] with permission. © Elsevier)

Full size image

As explained above, the data gathering system and successive planning are required. Normally there are three methods of planning as listed below.

  • Computer tomography (CT) scan based imaging—Even though this method is commonly used in neuro and spinal surgeries, it has also been employed in case of orthopedic surgeries [16]. CT scans conducted on the human body are processed to obtain a 3 D image. Matching is done with the patient’s anatomy by fixing the tracking markers visible to the computer. This data is then utilized to operate a pointing device visible to the computer. The surgeon has access to 3D template with the markers during the operation and determines the precise locations on the bone where the tool should be at. An example of CT scan aided planing is shown in Fig. 7.1.

  • Intraoperative fluoroscopy based imaging— This method is commonly employed in the trauma surgery. Modified fluoroscopy with the help of computer software helps to create an anatomical map. Basic principle behind this technique is based on X-ray imaging of the bones. Some of the X-rays are pass through where as some are scattered. The directly passed rays form an image whereas the scattered X-rays with charged direction and relatively low energy are traditionally considered for radiation exposures. The multiple fluoroscopic images can be acquired and processed by computer software to provide real time multi-planar images without the need for extensive fluoroscopy exposure. The software superimposes the position of surgical instruments and the path of an implant onto real time imaging, allowing the surgeon to modify implant trajectory without the need for further fluoroscopy . 3-D fluoroscopy is an evolution of this technique and consists of a mobile C-arm unit, modified to incorporate a motorized rotational movement that is linked to a computer to provide multi-planar 3-D images of bony structures. An example of intraoperative fluoroscopy is shown in Fig. 7.2 where the bone and tool are seen in two different views [17].

  • Image free Computer software has an anatomical model, built up from a database of stored CT scans, for the procedure to be performed. The computer model is then augmented by ‘surface registration’ whereby a pointing device held by the surgeon , and visible to the computer, marks out predetermined areas of the patient’s anatomy (Fig. 7.3). This system avoids radiation exposure to the patient and surgical team. The technique is refereed to as bone morphing [18]. The data is collected as surgeon makes a movement. Data is gathered from predetermined number of random points. The bone morphing algorithm then utilizes this data to construct a image based on the radius of the bone. Finally, following three steps are required to obtain this 3D statistical shape similar to the one shown in Fig. 7.3:

    1. (a)

      acquisition of training shapes

    2. (b)

      definition of a point-to-point correspondence between training shapes

    3. (c)

      statistical analysis.

Thus it is clear that, more the number of times the algorithm analyses the data, more will be the accuracy. This data is then used for planning of the surgery assisting in navigation of the tools and alignment of the bone (Fig. 7.4 [12]).

Fig. 7.2
figure 2

In vitro image-interactive navigation . In two previously acquired C-arm images, the current location of one pedicle instrument is displayed. Left right oblique view with the tool’s axis shown as a black circle in the center of a cross-hair. Right lateral view. The instrument is represented as a black line with a cross-hair identifying the tool’s tip. A dotted line indicates the trajectory that the instrument is about to follow (reprinted from Nolte et al. [17] with permission. © Springer)

Full size image
Fig. 7.3
figure 3

Digitization of a cloud of random points on the bone surfaces of the patient (reprinted from Stindel et al. [18] with permission. © 2002 Informa UK ltd)

Full size image
Fig. 7.4
figure 4

Three-dimensional view of pedicle screw fixation-using navigation (reprinted from Rambani and Varghese[12] with permission. © Elsevier)

Full size image

It can be concluded that the technology of surgery planning can result in accurate positioning and better alignment assessment of the deformity, improved gap balancing, and a decreased incidence of fat embolism due to the avoidance of intramedullary instrumentation [12, 19–21]. On the other hand, it has some demerits at this point of time such as the learning curve associated with embracing a new technology, prolonged surgical time, setup, and maintenance costs and the lack of long-term evidence on clinical outcome [22]. The key characteristics of computer aided surgery planning that are currently adopted in the clinical environment are summarized in Table 7.1. This points towards more advancements in the field of surgery planning in order to make this process efficient and timely.

Table 7.1 Key features of various methods of computer aided surgery planning
Full size table

7.2 Robot Assisted Surgery

Another aspect of automation is employment of robots for actual surgical operations [23–37]. This becomes feasible as a result of bone not undergoing large amount of deformation during machining. On a broader scale, the robotic systems are classified as passive, active, positioning , and cutting /milling aids. In passive systems, the robot remains under the control of a surgeon . Active systems on the other hand have much autonomy to perform the operations. The reference points are tracked and machining can be done by robot independently. Positioning systems are similar to methods described in previous section. The positioning of tools is done via robots based on algorithms which are then operated by surgeons . Last type that is cutting/milling aids combine capability of navigation and passive systems. The initial efforts of robotic surgery date back to 1987. Since then the technique is constantly evolving. A brief chronological summary of evolution of surgical robotics are schematically shown in Fig. 7.5 [38] which gives the extent of rapid progress in this field. The general steps followed during robot assisted surgery are summarized (Fig. 7.6 [39]). The first step consists of image acquisition from which the data is extracted for the purpose of feeding it to the planning algorithm . After these steps, the actual execution of surgical step by a robotic unit takes place followed by post surgery monitoring and evaluation.

Fig. 7.5
figure 5

Evolution of surgical robotics (reprinted from Gomes [38] with permission. © Elsevier)

Full size image
Fig. 7.6
figure 6

Steps during robot assisted surgery (reprinted from Korb et al. [39] with permission. © Elsevier)

Full size image

There are certain advantages of robot assisted surgery over conventional surgery as pointed by Adili and listed below [37].

  • Improved accuracy and precision in the preparation of bone surfaces.

  • More reliable and reproducible outcomes

  • Greater spatial accuracy.

  • Ability to make repetitive motions tirelessly.

  • Ability to accurately and predictably position and orientate equipment to a re-programmable point.

  • Ability to move to a location and then hold tools there for long periods accurately, rigidly, and without tremor.

  • Ability to actively constrain tools to a particular path or location to achieve accurate cuts and positioning of equipment.

  • Ability to make precise micromotion adjustments.

Because of these advantages, robotic techniques have been employed in many surgical fields such as neurology , maxifocal , cardiac , opthalmo , and orthopedic. However, current discussion will be focused on robotics employed in orthopedic surgery and few case studies related to conventional and non conventional methods will be covered.

Robots render suitable for orthopedic surgery because of the structure of the bone. The rigidity of bone allows the robotic devices to secure on them. Being radio opaque, the skeletal structures can be easily imaged and can feed data to planning algorithms used in robot navigation . Easily definable reference points or the implantation of dynamic reference bodies (fiducials) greatly facilitate the registration process in case of bones that all robotic systems require. Bones can be relatively easily manipulated and fixed in known positions and then robotic devices can generate high forces needed to create accurate cuts. As the robot system can be constrained and bounce off hard surfaces; the possibility of avoiding/minimizing cutting of the vulnerable surrounding soft tissues is very high.

One of the earliest studied system for orthopedic application was ROBODOC\(^{TM}\) which consists of a planning station, a robot for machining, and a controlling station [38, 40]. The robotic arm can be carefully designed so as to have flexible and controllable movement (Fig. 7.7 [40]). Various tools can be attached to the arm to perform precise machining . Locating pins need to be mounted on the bone in order to provide navigational guide for the robotic system. The CT scans obtained here after are transfered to the planning station. The surgeon can then plan the machining parameters , for example by checking different positions, and assess the impact on anteversion, neck length, and stress loading in case of hip replacement with an implant [32]. The cutting times in case of hip replacement are 25–30 min during which the surgeon monitors the process from the control station. The process can be stopped in case of unusual events. After completion of milling, the remaining operations are done in conventional manner. The schematic of ROBODOC\(^{TM}\) system is shown in Fig. 7.8. It has also been reported that, even though the cuts are made in reproducible and tireless manner, the inserted navigational pins play a critical role [32]. It has been reported that, after clinical trials, a severe pain may occur in the area where the navigational pins were inserted [41]. Careful choice of location on the other hand has been reported to reduce the pain to much lower level [42]. ROBODOC\(^{TM}\) has been successfully employed in clinical trials, for example it was used in 1000 surgeries in Frankfurt, Germany [43]. Apart from ROBODOC\(^{TM}\), the other systems employed are PAKY/RCM\(^{TM}\) [44], kawasaki\(^{TM}\), BRIGIT\(^{TM}\), and Acrobot\(^{TM}\) [32] and have been reported to work on similar principle.

Fig. 7.7
figure 7

Design of robotic arm illustrating degrees of freedom with the arrows (reprinted from Camarillo et al. [40] with permission. © Elsevier)

Full size image
Fig. 7.8
figure 8

Schmatic of ROBODOC\(^{TM}\) with various subsystems

Full size image
Fig. 7.9
figure 9

Work flow of laser robotic surgery

Full size image
Fig. 7.10
figure 10

Setup for robot assisted laser machining of bones [45]

Full size image

Even though robot assisted surgery wields a great potential, the following few key issues still remain associated with it.

  • Significant forces are required to cut/penetrate the rigid bone which might compromise accuracy.

  • Increased operative exposure and disruption of soft tissue structures to allow access for current robotic end effector designs and bone motion sensors can compromise clinical outcomes.

  • Significant capital cost.

  • Steep learning curve for operating personnel.

  • Additional technicians are required in the operating room.

  • Most importantly, additional invasive procedures are required to implant the dynamic reference bodies (fiducials), leading to increased operating time for the procedure.

Considering these issues, it is clear that surgical robotics has a long way to go to become feasible in routine surgeries at grass root level hospitals. A collective effort of medical professionals and engineers is needed to design the system with least issues. Nonetheless, the robots are being explored in surgeries, especially for non-conventional bone machining techniques such as laser machining [45].

Lasers due to their inherent monochromatic and inherent nature posses inherent precession machining characteristics as discussed in previous chapters. It has been proposed that if the lasers are combined with surgical robotics, the control and precession would be further better [24, 28, 46]. Another advantage of laser is some types can easily be transmitted through glass fibers allowing main laser unit to be located remotely away from the surgery theater to improve radiation safety of patients and operators alike. The basic work flow adopted for laser based surgery is similar to one opted for conventional methods (Fig. 7.9). The process starts with collection of data about the bone geometry in order to choose the ablation pattern and figure out execution of the chosen pattern [47]. Simulation follows the collection of data, which can virtually determine the ablation characteristics of various input parameters and determine the most optimum condition on a computer. These optimum parameters are then fed to the laser robot setup (Fig. 7.10 [45]), and the most optimum ablation region is figured out based on lens characteristics. The head movement is decided by the algorithm and the ablation operation is thus performed by the laser held in robotic arm/head to machine the desired portion of the bone.

Thus robot based orthopedic surgery, irrespective of what machining technique (mechanical or heat based) is used, involves several major components as depicted in Fig. 7.9. These major components are first level of essential gradients of a semi or fully automated robotic surgery. However, each of these components also comprise of many subcomponents to execute their role in synchronized manner during an integrated robotic surgery. Due to need of integration of these components in synchronized manner at multiple levels for desired outcome, design and fabrication of a complete robotic orthopedic surgery system is a complex and long term endeavor. Such enduring efforts would require multidisciplinary collaborative efforts among all engineers (design, mechanical, computer, electrical, and biomedical), material scientists, physicists, and orthopedic surgeons . As iterated in Chap. 1, the orthopedic field is growing in leaps and bounds around the globe. Hence, this is likely to drive the development of integrated robotic orthopedic surgery system for several reasons including a need for accurate and faster surgery, cost effective surgery and rapid post surgery recovery in the near future.