Planning of Surgical Treatment with the Use of 3D Visualization and Finite Element Method



Fig. 1
Left: The marked points on the hip joint border and the approximated plane are shown. Right: Resulting position of the artificial hip joint in correspondence to the mirrored, healthy hip part [16]



Another programme which is used for preoperative planning of surgical procedures is SQ PELVIS. It enables virtual planning of operations of pelvis injuries on the models created on the basis of DICOM images [5]. The segmentation of tissues on the grounds of the Hounsfield scale plays an essential role in the planning process. Having generated a satisfactory model one may position implants and match them to the individual needs of the examined patient (Fig. 2).



A329170_1_En_9_Fig2_HTML.gif


Fig. 2
Virtual planning of pelvis stabilization with the use of SQ PELVIS system. Left: Virtual reduction and fixation of the fractured bone. Right: The direction and length of the screws [5]

Another approach aims to support the surgeon by providing them with templates which facilitate technical aspects of carrying out the operation, for instance, a navigation system (Fig. 3) which was used in the work of Gras et al. [13] to plan the position of the stabilizing screws in pelvic ring injuries.



A329170_1_En_9_Fig3_HTML.gif


Fig. 3
Sterile touch screen of the navigation system (Vector Vision, Brainlab) displaying standard images (lateral view, inlet, outlet) and an auto-pilot view. Red bar: virtually planned SI-screw; yellow line: prospective path of the navigated guide wire (trajectory), green bull’s-eye: reflecting the exact positioning of navigated instruments to achieve the planned screw position [13]

Another example can be provided by operative planning in orthognathic surgery [10]. The standard planning is done on the basis of CT scanning (Fig. 4). However, there are also special programmes, such as Mimics and 3-matic software (Materialise) [23] for planning the corrections of the facial skeleton. Similar procedures supporting treatment in orthognathic surgery were developed, among others, by: Cutting [8, 9], Yasuda [30] and Altobelli [1].



A329170_1_En_9_Fig4_HTML.jpg


Fig. 4
Example of computer-aided surgery (CAS) of a patient with Crouzon syndrome. Simulation and result of Le Fort II distraction before surgery and after CT planning [10]

In more advanced research new devices are being developed with the purpose of supporting the doctor during the surgical procedure, for example: a neck jig device presented in the work of Raaijmaakers et al.[25]. The Surface Replacement Arthroplasty jig was designed as a slightly more-than-hemispherical cage to fit the anterior part of the femoral head. The cage is connected to an anterior neck support. Four knifes are attached on the central arch of the cage. A drill guide cylinder is attached to the cage, thus allowing guide wire positioning as pre-operatively planned (Fig. 5).



A329170_1_En_9_Fig5_HTML.jpg


Fig. 5
Neck jig designed to drill a guide wire in a pre-determined position and direction, seen from medioposterior (left) and anterolateral (right) [25]

Apart from planning a procedure for an individual patient, new methods of engineering support make it possible to choose optimal parameters for the operation. An example can be provided by the application of a method of finite elements in the biomechanical analysis of the system after simulated virtual treatment. In the research of Jiang et al. [19] planning of corrective incisions (scaphocephaly) was done as well as biomechanical analysis of the obtained models was performed (Fig. 6). Thanks to that, it is possible to choose the most favourable variant of the operation. In addition to that, the research of Szarek et al. [27] analysed the level of stress in the hip joint endoprosthesis resulting from variable loads during human motor activity.



A329170_1_En_9_Fig6_HTML.gif


Fig. 6
Distribution of stress (top) and displacements (bottom) in the skull vault before the surgery and in five variants of corrective incisions [19]

Analysing the influence of preoperative planning and 3D virtual visualization of the examined cases on the quality of treatment, it can be stated that the engineering support provides assistance for the vast majority of doctors in the scope of complex assessment of the phenomenon and preparation for a real-life procedure. The conducted research has proven that [18] both the planning time and labour intensity are reduced by around 30 % if 3D models are available. In addition to that, the precision (accuracy) of predicting the size of the resection area (e.g. in the case of tumours) increases by about 20 % (Fig. 7). Moreover, according to subjective feelings of the examined doctors their confidence in the established diagnosis has risen by around 20 % in the case of 3D planning.



A329170_1_En_9_Fig7_HTML.gif


Fig. 7
Comparison between viewing 2D CT images and 3D displays of thoracic cavities in determining the resectability of lung cancer. Left: Planning time. Right: Accuracy of predicted resectability [18]



2 Engineering Support Procedure for Preoperative Planning


Surgical treatment within the skeletal system is always the last resort in the case when other preventive methods have failed. For instance, when the application of orthopaedic equipment has not brought the desirable effects. On the basis of several-year tests carried out in co-operation with surgeons a general scheme of engineering support procedure has been developed for pre-operative planning of surgical operations (Fig. 8).



A329170_1_En_9_Fig8_HTML.gif


Fig. 8
Developed procedure of engineering support procedure for preoperative planning

In the first phase the attending physician gives a diagnosis of the disease. Usually, within the framework of a regular diagnosis a CT or MRI examination is done, thanks to which 2D images of individual cross sections are obtained. On the grounds of the Hounsfield scale, in the programme Mimics® Materialise [31] it is possible to segment the tissues of interest (e.g. bones, cartilages) and then generate a 3D model. In the next stage, on the basis of the constructed geometrical model it is possible to carry out detailed morphological measurements in order to determine the type of the defect and degree of the disease progression.

On these grounds the patient is qualified for the surgical procedure by the doctor. Additionally, the programme 3-matic® Materialise [32] makes it possible to do the analysis of bone thickness, which is very helpful in the selection of surgical tools for the operation. Mimics programme enables all sorts of modifications of the obtained model as well as simulations of the planned operation. In consultation with the doctor, bone incisions and displacements are simulated with the purpose of obtaining the desirable treatment effects. After the correction has been planned, it is advisable to conduct the morphometric analysis once again in order to check the values of indexes which were used in the preoperative evaluation.

Next, the model is prepared to be introduced into computing environment. Discretization of the model, i.e. the creation of the volumetric mesh and its optimization is done in the 3-matic programme. Then, the model is exported to Ansys Workbench® environment [33] in order to carry out biomechanical analyses. The primary objective of FEM analysis is to check whether during bone modelling or implanting no fracture or damage to the structure occurs. It is particularly important while planning endoscopic surgical procedures due to the fact that any unforeseen fracture of the bones makes it necessary to stop the microinvasive surgical treatment and complete the operation with the use of classic methods. A numerical simulation provides thus an individual risk assessment of the surgical procedure and may be a decisive factor while selecting a variant of the operation.

Finally, by comparing the results of the performed analyses it is possible to make the most advantageous and the safest choice of the operative variant. It must be emphasized that preoperative planning is an absorbing and time-consuming process, and therefore, not suitable for all kinds of operations. Its application is justified and brings many notable benefits in the case of particularly complicated surgical procedures.

In the further part this paper presents examples of procedures of engineering support for preoperative planning in the cases of surgical corrections of head shape in infants, corrections of chest deformities as well as spine stabilization.


2.1 Application of Engineering Support in Preoperative Planning of Head Shape Correction in Infants with Craniosynostosis


Craniosynostosis is a condition in which one or more of the fibrous sutures in an infant skull prematurely fuse by turning into bone (ossification), thereby changing the growth pattern of the skull [2, 20, 21, 29]. One of the most common cases of craniosynostosis is trigonocephaly, i.e. premature fusion of the metopic suture leading to a deformity of a triangular shaped forehead. Planning of the endoscopic correction of trigonocephaly in a two-month boy was done. CT images of the head were imported to Mimics environment in order to generate a 3D model of the skull. The primary task was to decide on the basis of a morphological analysis whether to qualify the patient for the classic surgical procedure or microinvasive one. The bone incisions and displacements were planned as well as possible variants of correction were developed.

In the first place, the analysis of bone thickness was performed in order to initially determine a type of the operation [22]. Maximum and minimum thickness was measured in the sites of fusion of the metopic suture subject to resection. Thickness in those points was respectively: max 7.01 mm, 4.46 mm and min 2.02 mm (Fig. 9). Thickness points of the whole skull were also determined. Average thickness of bones equalled 2.0 mm with a standard deviation of 1.2 mm.



A329170_1_En_9_Fig9_HTML.gif


Fig. 9
Bone-thickness analysis performed in 3-matic software

Subsequently, points necessary for setting indexes determining an incorrect shape of the head were marked on the model [28]. These are the following points: euryon, metopion, sphenion c, lateralis orbitae, medialis orbitae, nasion (Fig. 10).



A329170_1_En_9_Fig10_HTML.gif


Fig. 10
Three-dimensional model of skull with trigonocephaly with marked anatomic points

One set the values of indexes determining an incorrect shape of the skull in trigonocephaly. Those values were next compared to the standard values of children with a regular skull shape at the age of 0–2 months old in order to determine the way the correction should be performed. The results have been presented in Table 1. The measurements showed that the frontal angle was too acute but other indexes were within regular limits. No hypertelorism was detected, therefore the correction was going to be made only on the frontal bone without any interference in the orbital cavities. In this way it was determined that it was possible to carry out a microinvasive procedure. The main decisive factors at that stage of planning were as follows: the patient’s age, bone thickness (within 5 mm in the sites of potential incisions) as well as a correct distance between orbital cavities and a lack of deformities within facial skeleton. The doctor made a decision that the correction of the skull shape was going to consist in the cutting of the fused metopic suture and parting of the bones in order to obtain an optimum shape of the head.




Table 1
Craniometric measurements results in patient with trigonocephaly








































Index

Figure

Measured value

Normative value

Guidelines for surgery

Frontal bone angle

A329170_1_En_9_Figa_HTML.gif

121,8°

133,1º ± 5,6

increase

Naso-orbital angle

A329170_1_En_9_Figb_HTML.gif

103,5°

104,9º ± 5,9

regular

Index of the width of the inner orbits and the width of the skull

A329170_1_En_9_Figc_HTML.gif

0,14

0,16 ± 0,02

regular

Index of the width of the outer orbits and the width of the skull

A329170_1_En_9_Figd_HTML.gif

0,62

0,65 ± 0,03

regular

The virtual correction was performed in two stages. In the first one the frontal bone was separated from the rest of the skull alongside frontoparietal sutures. The lower limit was provided by the nasal bone and frontozygomatic sutures. The incisions of the frontal bone were planned in Mimics environment in the way guaranteeing an optimum forehead shape (Fig. 11). Dislocations, in fact rotations, of the fragments of the bones were done manually taking into consideration the doctor’s suggestions and actual conditions of the operation. Point nasion (n) on the nasal bone was defined to be a fixed point according to which the bones were parted to make the head shape round.



A329170_1_En_9_Fig11_HTML.jpg


Fig. 11
Virtual correction of forehead performed in Mimics

Having obtained an optimal visual effect of the correction, one measured displacement of several bone points which in the further planning phase were introduced into Ansys environment as boundary conditions. In the end, the average displacement of the bones from their initial position equalled 11 mm. The value of the frontal bone angle was checked once again in order to evaluate if it was now close to standard. After the procedure the angle was increased up to 132.7º, which produced a satisfactory effect of the correction (Fig. 12).



A329170_1_En_9_Fig12_HTML.jpg


Fig. 12
Measurements of skull after virtual correction. Left: Displacement necessary to correct the forehead shape. Right: Forehead angle before (121,8°) and after (132,7°) the surgery

Numerical simulations were carried out in Ansys environment. However, before that discreet models of individual bones had been prepared in 3-matic programme compatible with Mimics software.

For the examined models simplifications were adopted in three basic categories: geometry, material and boundary conditions. Skull geometry, which was generated on the basis of CT images, was imported from Mimics programme without skin, blood vessels and other structures. Also, one did not take into consideration joints between the anatomical elements of the skull, such as cartilages, sutures, etc. Preparatory proceedings were similar for all of the below-mentioned cases. The generated bone models with adequate incisions were digitalized. In order to optimize the model, Laplace’s method of approximating integrals was used several times (an inbuilt function of the software) with a coefficient equal to 0.4 ÷ 0.7. Next, they were divided into tetrahedral finite elements of Solid 72 type, whose maximum length of edges did not exceed 3 mm. Then, a volumetric mesh was generated in order to finally export the model to *.cdb format. Such actions caused the loss of geometrical details which were deemed irrelevant.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jun 14, 2017 | Posted by in GENERAL SURGERY | Comments Off on Planning of Surgical Treatment with the Use of 3D Visualization and Finite Element Method

Full access? Get Clinical Tree

Get Clinical Tree app for offline access