ADVANCED SURGICAL TECHNOLOGIES FOR PLASTIC AND RECONSTRUCTIVE SURGERY - 08/09/11
Résumé |
Nasal reconstruction using a forehead flap was one of the most complex procedures done during the early history of surgery. This was done in India, using a leaf as a template for the nose to be constructed (Figure 1). Subsequently, surgeons used modeling clay as a template for nasal reconstruction. In its most developed form complex templates from a number of materials were used in Russia during the first half of the twentieth century in the work of Limberg.21 He used paper templates to study the geometry of flaps and to analyze the three-dimensional distortion caused by tissue rearrangement. Limberg attempted to introduce basic mathematical principles into surgical practice. His use of paper and subsequently cloth models was designed to allow the surgeon to begin to grasp the consequences of geometric manipulation on living tissue. However, he found that these materials were limited in their ability to match the properties of skin and underlying tissues.
The introduction of the computer midway through the twentieth century provided a tool that had the potential to model the complexity of real tissues (see Figure 1. However, the original computers did not have the power to solve large enough sets of equations to represent skin and soft tissue. In addition, mathematical models of the mechanical behavior of materials had to be modified to more accurately represent human tissues. Most recently, computer animation techniques have yielded simulations that behave like human skin, muscle, and bones. An interactive graphic interface to this computer model would allow a surgeon to plan and simulate the outcome of surgical procedures of these tissues. The most advanced interface is virtual reality using a specialized helmet and electronic gloves to allow the surgeon to see and feel the reconstructed nose (see Figure 1.
The computer has been used in the past to provide an expert system to help the surgeon decide the best operation for a specific problem. An expert system is a flow-chart summarizing an expert's approach to a certain problem. The computer functions through the use of an algorithm written into the software, similar to how chess playing computers determine their moves in specific cases. The expert system presents predetermined case histories and images, as well as a menu of possible actions from which the user can choose. The rhinoplasty simulator developed by Constantian is a very elegant example of the use of an expert system.5 The system then will display a prediction of outcome based on the procedures chosen. All possible outcomes are prestored in the simulation system.
The computer can use two-dimensional and three-dimensional data to display the results of surgery. These programs are currently used to design rhinoplasties, midface advancements, and mandibular osteotomies.22 The computer is used to display two-dimensional images that have been digitally retouched or painted by the surgeon or operator. Because graphic representations have no information about the physical properties of tissue, they rely solely on the surgeon to predict the outcome of his surgical plan and then on his ability at photo-retouching to create the final image.
The three-dimensional patient-specific data obtained from radiology studies such as CT and MR imaging scans can be formatted by computer graphic rendering techniques into visual three-dimensional objects. Surgical simulation systems have been based on segmentation (cutting) and rearrangement of the volume data. This technique is very useful in bone surgery, such as craniofacial surgery,1, 6 as the modeled tissue is rigid.
To overcome the limitations of two-dimensional and three-dimensional geometric static models, the computers require a model of the physical properties of the tissue to analyze relevant mechanical consequences of a proposed surgical procedure. The ultimate goal of the patient model is that it accurately accounts for the patient's tissue at each location of the body for computer simulation. Three-dimensional patient data currently obtained in medicine (e.g., CT scans, MR imaging, positron emission tomography [PET] scans) are encoded volumetrically, that is, each point in space is defined by an absolute reference frame independent of the patient, and the material is encoded at each of these points. These data are not immediately amenable to modeling because there is no information as to how each piece of material connects to other pieces of the material.
A solution to the above problem is to create a finite element mesh of the human tissue to approximate the true properties of the skin, muscle, or bone. A finite element mesh (FEM) divides a material with complex geometry into regions (elements) that, taken together, approximate the behavior of the entire material. Each region (element) is defined by the boundaries it shares with other elements. A matrix with the material properties of the elements will predict each element's distortion, given the restriction that it must still share the same borders with the other elements. There are several mathematic algorithms for the construction of finite element meshes based on data in volume data sets.3
Although there is much interest in the head and neck, the limbs provide a simpler test bed for creating finite element meshes. The biomechanics of the joints of the limb are better understood than the facial muscles or temporomandibular joint. One of the earliest surgical simulators used a computer simulation of the lower extremity musculotendonous system to analyze tendon transfer operations (Figure 2).8 Using a computer-generated model of the hip joint with muscle and tendon actuators, the outcome of hip arthroplasties (total hip joint replacements) were predicted.9 By simulating 41 muscle-tendon complexes, the maximum force generated by leg abduction, adduction, flexion, and extension in relation to the position of the hip joint itself was found.3 Using this model, one can predict the effect of hip prosthesis position on individual muscle groups. Expanding the model to include major muscles of the entire lower extremity, Delp and Zajac made several interesting observations concerning the affect of tendon lengthening and transfers on muscle strength.10 These procedures often are performed on patients with gait or posture abnormalities due to stroke or cerebral palsy. The significance of accurately predicting outcomes of reconstructive surgery is great.
Chen and Zeltzer2 present a method that combines realistic computer animation and valid biomechanical simulation of muscle. Taking human animation beyond simulating surface geometry of skin, Chen details the modeling of individual muscles. Using reconstructed three-dimensional images from CT, MR imaging data, and a three-dimensional modeling program, a polyhedral model of the human calf muscle (gastrocnemius) was constructed. The biomechanical model was synthesized using the finite element method. By developing a model with the capacity to simulate actual muscle force and visualize the dynamics of muscle contraction, Chen has created an animated character that changes shape accurately and realistically.
McKenna has developed a system to simulate “complex human kinematics.”23 His model of the human figure contains more than 90 degrees of freedom, with 28 degrees of freedom in each foot, incorporated anatomic diagrams, a three-dimensional digitized skeleton, and clinical and cadaver studies of the biomechanics of limbs and joints. Simulated actions include rising to standing from knees, reaching, rising on toes, and walking and falling under gravity (Figure 3). The complexity of this foot model could be used to model the underlying joints in the head and neck, including the temporomandibular joint and neck.
Satava has created a “virtual abdomen” to teach medical students specific anatomic details of abdominal organs, and to instruct surgical residents in surgical techniques and operative procedures.35 This computer model allows the viewer to see the anatomy from outside of the organs, as in a traditional open laparotomy, but also from the inside in a “fly-through” mode, as in an endoscopic procedure. There also are laparoscopic tools in the model to perform simulated minimally invasive surgery. These types of tools could be used in a simulator for endoscopic sinus surgery to help to train residents in how to learn and perform removal of tumors. Data fusion could be used to prevent the surgeon from inadvertently injuring the eye or brain while performing the resection of adjacent sinus tumors (see later discussion).
Similar to the earlier work of Limberg with paper and cloth models, Larabee compared a two-dimensional FEM simulation of human skin to pigskin to analyze flap advancements.18 Kawabata used a two-dimensional FEM to analyze the effect of various Z-plasty parameters.17 Motoyoshi used an FEM model of facial soft tissue to predict the outcome of orthognathic surgery.24 Lee and others have used an FEM with overlying detail similar to computer-aided plastic surgery (CAPS) to generate computer-synthesized facial expression.20
One of the most difficult areas of the human body to simulate is the face. Through it, one communicates verbally and nonverbally expresses emotions and thoughts, and feeds one's body. The fluidity, individuality, and complex musculature of the face complicates attempts to accurately model the subtlety and complexity of facial expression. Although early efforts at key frame animation of the face proved satisfactory for two-dimensional modeling, the time required to specify the large number of key frames required for three-dimensional simulation is impractical.27 In the early 1980s, Platt and Badler simulated the human face using a three-layered model including skin, muscle, and bone.29 Skin is represented by a set of points with three-dimensional coordinates. Bone is represented as a rigid surface below the skin. The muscle is a group of muscle-fiber points that are connected by elastic arcs to the overlying skin and underlying bone. Points on the skin are connected also to neighboring points through arcs. By integrating the network of points, one can demonstrate how the application of force or tension on one section of the model will effect more distant areas of the same surface.35
The author's program, CAPS, allows the planning, analysis and visualization of plastic surgery on the soft tissue of the face. Pieper had developed a detailed model of the face using reconstructed images from CT and MR imaging scans of the patient.28 It is a prototype computer program that a surgeon could use as a sketch pad to predict and compare the outcome of facial plastic procedures on a patient-specific physical model. One can select incision placement, move tissue, and suture. The CAPS program uses an FEM to simulate plastic surgery by removing certain elements and then redefining the remaining elements as sharing their (formerly separate) borders, just as a surgeon excises tissue and defines new shared edges with sutures. When the computer calculates the distorting forces and applies this to the patient specific model, one can visualize the consequences of the surgery.
Although one would like to create an exact model of human facial tissues, this still remains to be done. The present model is of a homogeneous layer of uniform thickness, a simplification of the actual human face with varying thickness of facial soft tissue. This is analogous to predicting the outcome of facial surgery by using a detailed facial model made of uniformly thick foam rubber.
Le texte complet de cet article est disponible en PDF.Plan
| Address reprint requests to Joseph M. Rosen, MD, Division of Plastic and Reconstructive Surgery, Dartmouth-Hitchcock Medical Center, One Medical Center Drive, Lebanon, NH 03756–0001 |
Vol 31 - N° 2
P. 357-368 - avril 1998 Retour au numéro
Article précédent
- USING ADVANCED SIMULATION TECHNOLOGY FOR CRANIAL BASE TUMOR EVALUATION
- Gregory J. Wiet, Donald L. Stredney, Roni Yagel, Dennis J. Sessanna
- VIRTUAL ENVIRONMENTS : Surgical Simulation in Otolaryngology
- Charles V. Edmond, Gregory J. Wiet, LTC Bill Bolger
Bienvenue sur EM-consulte, la référence des professionnels de santé.
L’accès au texte intégral de cet article nécessite un abonnement.
Déjà abonné à cette revue ?