**3.1 Time and material consumption**

As a result, an orthosis design methodology is proposed. Time consumption of individual steps of this process has been recorder to calculate the average length of orthosis production by modern technologies.

Overall duration of the scanning and data postprocessing of individual subjects is summarized in **Table 2**. Based on these results the average duration of the arm and forearm scanning is 2 minutes and 20 seconds, and the duration of data postprocessing is 5 minutes and 20 seconds. While scanning the area of interest no complications and errors have occurred. All collected data has satisfactory quality of the surface necessary for orthosis design.


#### **Table 2.**

*Subject scanning and data postprocessing duration.*

*Orthoses Development Using Modern Technologies DOI: http://dx.doi.org/10.5772/intechopen.95463*

No extra modifications of the scan 3D models were necessary during the orthosis design phase. Thanks to this fact, the design process of orthoses took approximately 3 minutes. Design duration for each orthosis is summarized in **Table 3**.

When positioning individual orthoses models on the virtual building platform of the 3D printer, the software automatically calculates the volume of the used model material, support material and the overall time of production. Average volume of model material used for the orthosis production is 65,94 cm3 , 57,40 cm3 of support material and the average time of production is 5 hours and 18 minutes. These data are summarized in **Table 4**.

#### **3.2 Verification results**

The results of comparing models generated from the 3D scanner software and digitally solidified 3D scan models shows that from the deviation in the range of ±0.050 mm the models are identical. Only differences are on the edges of the


#### **Table 3.**

*Orthoses design duration.*


#### **Table 4.**

*Orthoses printing parameters.*

models. An example is shown in **Figure 8**. Digitally solidified scan models were chosen for the actual to nominal comparison.

In **Table 5** the average deviation of solidified scan models to actual scanned models are summarized. To evaluate the differences between the actual scan model and the solidified model, the maximum deviations for 75%, 90% and 95% surface coverage were determined. These data indicate that for e.g. 95% of all values are the maximum deviation. By the deviation values of Orthosis 4 all comparisons have a deviation of less than 0.01 mm at 75% coverage, so there are only minimal changes compared to the actual model. At 90% this value is less than 0.03 mm and at 95% less than 0.08 mm. In these cases, the sets of deviations are already affected mainly by deviations caused by the closing of edges of the solidified model and possible defects.

When designing orthoses, the wall thickness was set to 2 mm. The actual thickness was measured in the "Wall thickness module" of VGStudio MAX. As a result of the analysis, the actual values range from 2.006 mm to 2.097 mm and the standard deviation is less than 0.24 mm (**Table 6**).

#### **Figure 8.**

*Original scan model and solidified scan model comparison with a detail on the edge of the model.*


#### **Table 5.**

*Average variation values of solidified scan models to actual scanned models.*


*Orthoses Development Using Modern Technologies DOI: http://dx.doi.org/10.5772/intechopen.95463*

#### **Table 6.**

*Actual wall thickness values.*

When the scan model is compared with the actual orthosis model, it is necessary to perform their mutual alignment before performing the analyzes, as the scanned orthosis has a different coordinate system than the designed model created in the Meshmixer software. Due to the shape of the orthosis (absence of planes and simple shapes such as cylinders, etc.), their mutual alignment is possible using the Best-fit and RPS methods (reference positioning system). When using the RPS method, the Best-fit of the objects is the first step, and second the subsequent transfer of points to the current model (orthosis scan), after which the alignment itself is performed. The **Figure 9** shows an example for Best-fit alignment and RPS alignment. The figure above compares the scan of the orthosis to the solid model using the Best-fit method and below using the RPS method.

**Figure 10** shows the deviations between the two methods of alignment. It can be seen from the histogram (**Figure 11**) that the deviations are almost symmetrical with respect to zero, i.e. the two alignments are rotated relative to each other, which can also be seen in the figure.

Significant differences in models (orthosis shape) do not allow the distribution points to be distributed on all orthoses in the same way. To eliminate the effect of point placement for RPS alignment, the orthosis scan model and actual model were aligned with each other using the Best-fit method.

**Table 7** shows the data for the average deviation of the original individual orthoses model with respect to the solidified model. To evaluate the differences between the solidified scan model and the original orthosis model, the average deviation value and the maximum deviations for 90% and 95% surface coverage were determined. The average value of the deviation is close to zero and thus the distribution of deviations has the character of a normal (Gaussian) distribution. The average value for 95% coverage is 0.419 mm and 95% coverage 0.576 mm.

To control the quality of production, the thickness of the orthosis over its entire surface was also evaluated. The results in **Table 8** show that the average thickness is 1.956 mm and the standard average deviation is 0.206 mm. Compared to the nominal solidified model, the average wall thickness of the actual manufactured orthosis is smaller by 0.0931 mm and the value of the standard deviation is greater by 0.0203 mm, which are negligible differences.

**Figure 9.** *Best-fit alignment (up) and RPS alignment (down).*

**Figure 10.** *Deviation between the best-fit and the RPS alignment.*

Duration of the inspection and verification process has not been recorded, since it is not a part of the design methodology. Only the results of this process are relevant.

#### **Figure 11.**

*Histogram representing the deviations of the 2 alignment methods.*


#### **Table 7.**

*Average deviations of the original model to the solidified scan model.*


#### **Table 8.**

*Average thickness values of the orthoses surface.*

#### **4. Discussion**

When developing custom orthoses by modern technologies it is necessary to follow the steps of the method proposed in this study. First important step is the positive obtainment. Method of 3D scanning has shown to be very practical, fast, clean, precise and comfortable for the subject and the scanning staff. Working with a handheld 3D scanner is very intuitive and simple. Only 1 person and a laptop or a PC is required for the scanning process. Since the scanner is portable, it is not necessary that the subject must be in a special work environment. This fact is very important, if the subject is immobile or has movement difficulties and it's a great advantage when compared to the traditional plastering method. The positives obtainment with the postprocessing of the acquired data took less than 10 minutes average, which is much quicker than the traditional way.

Body segment positioning before scanning is important. It is necessary to stabilize the segment of interest in order to capture the desired shape. It depends on what physical health the subject is in. If the subjects have movement restrictions or have weak body strength, it is important to provide them some form of support or stabilize the scanned body segment with the help of an assistant. The assistant can help stabilize the subject, but must support the areas, which are not important for orthosis development. When scanning subjects with no movement or force limitations it is still helpful to give them some type of support, for example, in this study the subjects had their elbow joint resting on a desk, while the arm and forearm had been scanned.

Artec Eva 3D scanner, which has been used in this study is an expensive, professional scanner used mainly for scanning larger objects and structures in mechanical, design, architectural, automotive and similar industries. Nevertheless, it is also applicable in prosthetics and orthotics. In one of our studies [22], we concluded that a high-end 3D scanner is not necessary and that low-cost scanners can capture important body segments with sufficient precision for orthosis development.

Capturing the segment in its correct shape is mandatory when designing an orthotic device. If the scanned model has defects, deformities or other artifacts it cannot be used as a positive. The original model must be clean and precise so that there's no editing needed. If the 3D model of the scan is edited, it could end up in a difference between the surface of the orthosis and the body surface, which would lead to an incorrectly designed aid.

One of the objectives was to use an open-source 3D modeling software. We chose Autodesk Meshmixer, as it has functions suited to prosthesis and orthosis design. In a few easy steps it is possible to design simple orthoses suited for additive manufacturing. The design process of an arm and forearm orthosis in this software took less than 4 minutes. The interface is very clear and organized and the user does not need any special training. However, when designing medical devices, it is necessary that a skilled prosthetist operates the software. Main disadvantage of the software is that it does not have medical certification, so the models should be used only for educational and research purposes.

Choosing the correct material for the production is important from the point of manufacturing and application. Since the orthoses are meant to fix and stabilize the arm and forearm and skin contact is unavoidable, the used material needs to be strong and biocompatible. Also, the Fortus 450mc printer uses cartridges of materials suited only for this type of printer, so the range of materials is limited by the manufacturer. For these reasons the ABS-M30i has been chosen as the most suitable material. Other materials like PETG or PLA can be used for orthosis production [20], but since this 3D printer does not support these materials, they were not chosen for this study.

#### *Orthoses Development Using Modern Technologies DOI: http://dx.doi.org/10.5772/intechopen.95463*

All models have been positioned with the dorsal side facing the printer bed. The meaning of this was to avoid support generation on the inner shell of the models to minimize support material volume and reduce postprocessing difficulty. Support structures on the inner shell could also deform the surface during the production process. From the results of the nominal to actual comparison of some models it is clear that the parts of the models that arch above the inner shell have deformed during printing. For this reason, it is advised to put support structures even on the inner shell of the models to avoid deformation.

The maximum time length of an arm and forearm orthosis produced by high or low-temperature thermoforming set by the public insurance company in Slovakia is 5 and a half man-hours. The manufacturing of a single 3D-printed individual orthosis took less than 5 and a half hours. When we add the average time of actual labor, which was approximately only 10 to 15 minutes, the whole process took less than 6 hours, which is the approximate time of an orthosis production by conventional methods for a single patient. This means, that by using proposed innovative methods, the technician can save time and design other orthotic devices, while the previous ones are producing. This time length can vary depending on the technologies used in single steps, or the number of models being developed. Duration of conventional and proposed production of arm and forearm splints is summarized in **Table 9**.

From the results of the analysis we can see that the difference between the produced orthosis and the 3D model is negligible from the point of view of orthotic application. After postprocessing and application of straps and maybe lining, these orthoses are fully functional and ready to use.

Since the manufacturing technology used in this study is a high-end, professional 3D printer, it is possibl e for hospitals, or prosthetic workshops, to produce their orthoses externally. Price of 1 orthosis, considering the material and applied technology, is approximately 70 euros. The maximum cost of an arm and forearm orthosis produced by high or low-temperature thermoforming set by the public insurance company in Slovakia is 166 euros (including materials, technology and man-hours). This means that there is a 96-euro gap between the conventional orthosis price and the 3D-printed orthosis manufacturing cost. This gap can be used to compensate the scanning, designing process and labor payment. Cost of conventional and proposed production of arm and forearm splints is summarized in **Table 9**.

If these institutions can acquire their own low-cost 3D scanner and maybe a 3D modeling software, the development process is faster, simpler and more practical, which means that the amount of produced individual orthotic devices grows. This is a favorable state not only for these institutions, but mainly for the patients themselves.

The proposed methodology, which contains orthoses design and additive manufacturing, is an adequate method for orthoses production. This method could also be used for design and manufacturing of individual prosthetic sockets for lower or upper limb prosthesis, trunk orthoses, orthotic seating systems and disability aids.


**Table 9.**

*Production duration and cost of conventional and modern arm and forearm orthoses.*
