Considerations on the Design, Printability and Usability of Customized 3D-Printed Upper Limb Orthoses

Featured Application

Practical information is provided, aimed at reducing the manufacturing time for customized 3D-printed orthoses by analyzing factors such as build orientation and process parameters. This includes comparing flat and thermoformed orthoses with those customized based on 3D scans. Printability issues are addressed for these thin prints that include many open pockets and another one of their dimensions around two-thirds of the forearm length.

Abstract

This paper investigated the feasibility of using 3D printing processes, specifically material extrusion (MEX) and vat photopolymerization (DLP—Digital Light Processing), to produce customized wrist–hand orthoses. Design, printability, and usability aspects were addressed. It was found that minimizing printing time for orthoses with intricate shapes, ventilation pockets, and minimal thickness is difficult. The influence of build orientation and process parameters, such as infill density, pattern, layer thickness, and wall thickness, on printing time for ten parameter configurations of orthoses in both ready-to-use and flat thermoformed shapes was examined. The findings revealed that the optimized orientations suggested by Meshmixer and Cura (Auto-orient option) did not reliably yield reduced printing times for each analyzed orthoses. The shortest printing time was achieved with a horizontal orientation (for orthoses manufactured in their ready-to-use form, starting from 3D scanning upper limb data) at the expense of surface quality in contact with the hand. For tall and thin orthoses, 100% infill density is recommended to ensure mechanical stability and layer fill, with caution required when reducing the support volume. Flat and thermoformed orthoses had the shortest printing times and could be produced with lower infill densities without defects. For the same design, the shortest printing time for an orthosis 3D-printed in its ready-to-use form was 8 h and 24 min at 60% infill, while the same orthosis produced as flat took 4 h and 37 min for the MEX process and half of this time for DLP. Usability criteria, including perceived immobilization strength, aesthetics, comfort, and weight, were evaluated for seven orthoses. Two healthy users, with previous experience with traditional plaster splints, tested the orthoses and expressed satisfaction with the 3D-printed designs. While the Voronoi design of DLP orthoses was visually more appealing, it was perceived as less stiff compared to those produced by MEX.

1. Introduction

Wrist–hand orthoses or splints are required for immobilizing patients’ upper limbs during the healing process by limiting joint motion, supporting proper joint alignment, or maintaining the position of weak muscles or joints during daily activities as part of conservative management [1,2]. These orthotic devices are frequently encountered owing to a large incidence of wrist-related injuries [3], and customization to the patient’s hand, essential for their effectiveness, makes Additive Manufacturing (AM) technology (colloquially known also as 3D Printing—3DP) particularly advantageous in this context. Additionally, it offers the necessary design flexibility to enhance the comfort of these devices and promote patient adherence to wearing them.

The proper fit of an orthosis to the limb is essential, as an ill-fitting one can cause discomfort, pain, and skin issues, leading the user to discontinue using the orthosis prematurely, even before completing the rehabilitation program. Therefore, static 3D-printed wrist–hand orthoses (3DP-WHOs) are a modern alternative to the traditional immobilization methods that involve the use of plaster of Paris [4], fiberglass, or thermoplastic splints [5].

The significance of this research is emphasized by the high occurrence of hand trauma, a condition requiring medical attention despite its non-fatal nature. As an example, from January 2009 to December 2018, around 2.6 million cases were reported annually by USA emergency departments [6]. The most common wrist–hand injuries include sprains, strains, carpal tunnel syndrome, fractures, tendonitis, and ligament injuries. It is worth noting that even minor hand injuries can negatively impact daily life by making the ability to perform tasks a challenge. Moreover, the substantial costs associated with the disability resulting from painful or unstable wrist joints are a notable concern in hand injury care [7]. Consequently, real-world scenarios encompass a broader range of potential cases that may require the use of immobilization methods, with prescribed durations of 7–45 days, depending on the specific pathology. This approach remains fundamental for addressing various wrist and hand conditions, with the choice of material and method influenced by multiple patient-related and financial considerations.

Recent literature discusses 3DP-WHOs’ outcomes from clinical practice for different types of pathologies by assessing their medical efficacy and users’ feedback [8,9,10]. From an engineering perspective, the studies primarily focus on topics related to the acquisition of upper limb patient data [11], design considerations [12], and mechanical performance [13]. The main objectives are reducing the manufacturing time and automating the design process while ensuring that the orthoses meet their functional requirements. Achieving these objectives would make 3DP-WHOs more accessible in medical facilities (3D Printing Point-of-Care) [14]. The first objective can be accomplished by printing the orthoses flat and subsequently thermoforming them to fit the patient’s limb [15], thus saving time and material. Both objectives can be achieved by generating flat orthoses using a predefined template and input from a hand therapist measuring the patient’s upper limb dimensions [16]. Other solutions for semi-automating the design process of 3DP-WHOs in their functional or ready-to-use form have also been proposed in the literature [17]. However, only a limited number of papers have discussed strategies to reduce the printing time and costs [18,19], although these are important factors related to the printability of customized WHOs, defined here as the capability to successfully 3D print orthoses that meet production efficiency criteria while also being functional and aesthetically pleasing.

The design process of a customized WHO typically begins with a 3D scan of the patient’s hand and forearm, followed by scan data processing and orthosis 3D modeling using specific software [11]. Thus, the virtual orthosis model is generated directly in its ready-to-use form corresponding to the anatomy of the patient’s upper limb and then sent to the 3D printer for manufacturing [12]. 3DP-WHOs often include open pockets for breathability and improved hygiene, enabling skin observation and reducing the weight of the orthosis. A critical design consideration, therefore, is the geometry of these ventilation pockets, which often, during the 3DP process, require support structures to maintain their overhangs. These supports negatively impact the printing time, cost, and surface quality. Moreover, since 3DP-WHOs are frequently reported to have thicknesses up to 4 mm [18,20,21], orienting them in a vertical position means printing thin and tall walls. These thin walls may lead to intra-layer filling defects or inter-layer adhesion defects, ultimately impacting their mechanical performance [13]. Conversely, building the orthoses horizontally can save time [18], but it affects the quality of the surfaces (resulting in a staircase effect) in contact with the patient’s hand. Optimizing the build orientation in accordance with the process parameter values is, therefore, important. It typically involves dedicated slicers that virtually slice the orthoses in different orientations and assess the support structures’ volume and placement, as well as printing time, through an iterative process.

Unlike previous studies that primarily focus on orthoses general design and material selection, the current study investigated how different print orientations and slicer software settings influence the printing time associated with orthoses manufacturing, the first of which compares parameter settings across commonly used slicers.

Summarizing, this paper introduces several novelties in the study of customized 3DP-WHOs:

-

Investigation of printability aspects related to the manufacturing process of customized 3D-printed orthoses of different designs by considering optimal orientations generated in several applications for decreasing the printing time, which currently is a limitation to a wider spread of these medical devices;

-

Analysis of process parameters’ impact on printing time for complex designs, a subject previously only discussed by Górski et al. [18];

-

Comparison of users’ experiences with orthoses designed based on 3D scans and those created by thermoforming to the user’s hand is an aspect that has not been addressed in the literature so far.

Thus, this paper’s main objective is to explore these novelties through the lens of material extrusion (MEX) and vat photopolymerization (DLP—Digital Light Processing) processes, assessing the design and printability aspects of ten different 3DP-WHO configurations. Furthermore, seven of the 3D-printed orthoses were assessed by two healthy users using a questionnaire following a functional test.

2. Materials and Methods

The methodology applied in this research included the following main stages:

• Design wrist–hand orthoses based on 3D scanning users’ hands and forearms, as well as by using a dedicated application that generates a flat-shaped orthosis based on the user’s hand size and shape. For this step of the process, CATIA V5 and Meshmixer software, as well as the web app described in [19], were used. The 3D models were then processed to generate Voronoi patterns and diverse open pocket models. All design data were exported in an STL file format.
• Have the orthoses 3D-printed using MEX and DLP after investigating diverse build orientations and process parameter settings using Cura and Chitubox slicers corresponding to analyzed processes. Printing time and cost, as well as surface quality, were the criteria used to analyze the development approaches (flat vs. 3D scanned models) and the orthoses designs.
• Conduct usability tests. A satisfaction questionnaire was administered using a Likert scale to evaluate various aspects of the orthoses and gather user preferences.
• Discuss the results.

Figure 1 is a detailed presentation of the activities conducted in each step of the research workflow.

2.1. 3DP-WHO Design Process

For this study, two AM processes, MEX—material extrusion (Ender 3 Pro 3D printer (Shenzhen Creality 3D Technology Co., Ltd., Shenzhen, China) and polylactic acid—PLA) and vat photopolymerization (AnyCubic DLP 3D printer (ANYCUBIC Technology Co., Ltd., Shenzhen, China) and 405 nm basic resin) were used to produce WHO samples (

Table 1. 3DP-WHOs used in the research.

Sample	Design	3D Printing Process	Approach	Material
1	Tear-like open pockets	MEX	Flat + thermoformed	PLA
2	Tear-like open pockets	MEX	Ready-to-use	PLA
3	Elliptical open pockets	MEX	Ready-to-use	PLA
4	Elliptical open pockets	MEX	Flat + thermoformed	PLA
5	Elliptical open pockets	DLP	Ready-to-use	Basic resin
6	Hexagonal open pockets	MEX	Flat + thermoformed	PLA
7	Hexagonal open pockets	MEX	Ready-to-use	PLA
8	Voronoi 2.7 mm thickness	DLP	Ready-to-use	Basic resin
9	Voronoi 3.7 mm thickness	DLP	Ready-to-use	Basic resin
10	Voronoi 3.7 mm thickness	MEX	Ready-to-use	PLA

).

Orthoses were designed based on the 3D scanning data of two healthy users acquired with an Artec Eva 3D scanner (Artec Europe S.à r.l, Senningerberg, Luxembourg). The scan’s 3D data were processed and exported as STL files using Artec Studio (Artec Europe S.à r.l, Artec 3D, Senningerberg, Luxembourg). The 3DP-WHOs included Voronoi patterns generated using Meshmixer software (Autodesk Inc., San Francisco, CA, USA) as well as elliptical, tear-shaped, and hexagonal open pockets modeled using CATIA V5 (Dassault Systemes, Vélizy-Villacoublay, France) with the goal of addressing a wider range of design and printability related aspects. The lengths of the wrist–hand orthoses were 164 mm and 175 mm, respectively, with a thickness of 2.7 mm for samples 1–8 and 3.7 mm for samples 9–10. Three of the orthosis models were also generated using the flat and thermoformed approach described in [13] for enabling a comparison of their printing time and surface quality with those of orthoses produced directly in their ready-to-use form (based on 3D scans). Thermoforming on the user’s hand was performed using the method investigated by Alexandru, et al. [22], which involved immersing the flat orthoses in hot water at 60 °C to exceed the glass transition temperature of PLA and to render the material malleable. After shaking off the hot water droplets to prevent burns to the user’s forearm, the orthosis was molded to fit the shape of the hand. When cooled down, the orthoses retained their molded shape. Three Velcro straps were subsequently positioned on the palm, wrist, and forearm areas to secure the orthosis on the hand.

Figure 2 shows the digital versions of the orthoses processed in Meshmixer for Voronoi pattern generation (Figure 2a) and those modeled in CATIA V5 starting from 3D scans (Figure 2b). The Voronoi pattern was created by reducing the mesh size of the orthosis to control polygon dimensions using the ‘Make Pattern’ option (Mesh + Delaunay Dual Edges) for pattern generation and setting the size and spacing between elements. To ensure that the orthoses are robust for daily use, the density of the patterns can be adjusted in specific areas using Meshmixer’s ‘Sculpt → Brushes → Reduce’ and ‘Sculpt → Brushes → Refine’ options.

Figure 3 illustrates the development process for the ready-to-use orthoses, which includes STL hand model importation, mesh cleaning and smoothing, orthosis surface generation, 3D modelling of the orthosis by including open pockets, and 3D printing (by DLP, in this example).

2.2. 3D Printing Process

Before 3D printing any object, including WHOs, the operators typically analyze different build orientations and process parameter settings using slicing software. In this research, the Cura Ultimaker slicer for the MEX process and the Chitubox Basic slicer for the DLP process were employed. This step helps in evaluating aspects like volume and position of support structures, filament consumption, printing time, and surface quality (Figure 4). Given the complexity of 3DP-WHO designs, this study explored not only the common vertical (upstanding) and horizontal orientations (Figure 4a), but also optimized orientations recommended by Meshmixer software (Figure 4b), the Auto-Orientation Cura 5.2.2 add-on (Figure 4c), and build orientation for the part to fit the 3D printer workspace (Figure 4d). The printing speed and temperatures for the extrusion and platform were set based on the type of material and past experience to produce good quality prints.

The mechanical resistance of 3DP-WHOs in response to wrist joint movements (flexion/extension, ulnar/radial) depends not only on the values of process parameters but also on the chosen build orientation [13]. At the same time, process parameters such as infill density, layer thickness, and the number of perimeters also influence the production efficiency aspects mentioned earlier.

Table 2. Three-dimensional printing process parameters.

Factors	Levels
Process parameters (orthoses produced by a MEX process)	Varied parameters	Fixed parameters
	Build orientation: vertical, horizontal, optimized Infill density: 100%; 80%; 60% Infill pattern: lines; zig-zag; grid; triangles Layer thickness: 0.1 mm; 0.2 mm	Printing speed: 55 mm/s Printing temperature: 210 °C Build plate temperature: 60 °C Line width: 0.4 mm Wall line count: 2 Top/bottom layers: 2 Build plate adhesion: Brim Support density: 5% Support pattern: zig-zag Material: PLA Support overhang angle: 63°
Process parameters for orthoses produced by a DLP process)	Layer height: 0.1 mm; 0.2 mm Wall thickness: 1.2 mm; 1 mm	Exposure time: 3 s Lift distance: 5 mm Lift speed: 2000 mm/min Bottom exposure time: 40 s Retract speed: 2500 mm/min Infill density: 100% Infill structure: Grid3D Material: Basic resin

presents the factors examined for each process along with the printing parameters used to assess their influence on both printing time and surface quality. Initially, these investigations were conducted based on the information provided by slicers. Subsequently, seven orthosis configurations were chosen for manufacturing, and their usability was evaluated (samples 3–9), with the findings discussed in Section 3.

Figure 5 presents two of the samples during the manufacturing stage: one created through a DLP process (sample 5), and the other 3D-printed in a flat shape (sample 1) intended for subsequent thermoforming on the patient’s hand. Several other manufactured orthoses are shown in Figure 6.

2.3. 3DP-WHOs Usability and Satisfaction Questionnaire

A usability assessment was undertaken to evaluate the effectiveness and satisfaction levels provided by the 3D-printed orthoses during daily activities. Samples 3, 4, 5, and 8 were tested by one user, while samples 6, 7, and 9 were tested by the other user. The orthoses differed in weight, open pocket design, and manufacturing process, with each tested orthosis customized to fit the user wearing it. The healthy users wore the orthosis for several minutes while performing different tasks, including pressing keys (Figure 7a,b) and lifting objects (Figure 7c,d). The tasks performed did not involve any health risks that could physically or psychologically affect the volunteers.

On a 1–5 Likert scale (1—totally disagree, 5—totally agree), the following questions were asked:

• How satisfied are you with the weight of the orthosis?
• How satisfied are you with the design of the orthosis?
• How secure is the orthosis on your forearm?
• How satisfied are you with immobilization strength provided by the orthosis?
• How satisfied are you with the comfort provided by the orthosis?
• How satisfied are you with the orthosis while engaging in the activities?

Additionally, the users were asked to rank the orthoses in order of preference and provide a justification for their choices.

3. Results

3.1. 3D Printed Orthoses Manufacturing Time Results

No significant differences were recorded in the printing time for other infill patterns (zig-zag, grid, triangles), especially for orthoses 3D-printed in their ready-to-use shape, where the surface available between the contours/perimeters on each layer is very thin. For the flat orthoses, such as sample 4 (flat elliptical pockets design), the printing time for 100% infill density lines was 6 h and 6 min. For the other patterns, the difference was only 1–4 min, representing around a 1% difference. Therefore, only the results for one type of pattern are presented further.

Table 3. Three-dimensional printing time for MEX-produced WHOs.

Sample	3D Printing Time
	Layer Thickness, Infill Density, Line Pattern	Flat	Vertical Orientation	Horizontal Orientation	Optimized Meshmixer	Auto-Orientation Cura 5.2.2
1	0.2 mm, 100% 0.2 mm, 80% 0.2 mm, 60%	4 h 40 min 4 h 19 min 3 h 55 min	-	-	-	-
2	0.2 mm, 100% 0.2 mm, 80% 0.2 mm, 60%	-	17 h 19 min 17 h 41 min 17 h 4 min	Similar with Auto-orientation solution	16 h 10 min 16 h 43 min 16 h 7 min	16 h 28 min 16 h 24 min 15 h 51 min
3	0.2 mm, 100% 0.2 mm, 80% 0.2 mm, 60%	-	10 h 55 min 11 h 16 min 10 h 56 min	8 h 49 min 8 h 45 min 8 h 23 min	17 h 23 min 16 h 47 min 16 h 34 min	12 h 2 min 12 h 2 min 11 h 40 min
4	0.2 mm, 100% 0.2 mm, 80% 0.2 mm, 60%	6 h 6 min 4 h 54 min 4 h 37 min	-	-	-	-
6	0.2 mm, 100% 0.2 mm, 80% 0.2 mm, 60%	4 h 14 min 3 h 31 min 3 h 22 min	-	-	-	-
7	0.2 mm, 100% 0.2 mm, 80% 0.2 mm, 60%	-	17 h 15 min 16 h 58 min 16 h 45 min	17 h 37 min 17 h 14 min 17 h 4 min	10 h 28 min 10 h 33 min 10 h 23 min	11 h 23 min 11 h 34 min 11 h 25 min
10	0.2 mm, 100% 0.2 mm, 80% 0.2 mm, 60%	-	12 h 30 min 12 h 26 min 11 h 53 min	11 h 17 min 11 h 6 min 10 h 34 min	20 h 29 min 18 h 33 min 17 h 47 min	15 h 42 min 15 h 46 min 15 h 52 min

displays the findings related to 3D printing times for MEX-produced orthoses, examining four distinct build orientations and the line infill pattern.

Figure 8 illustrates the effect of build orientation on the quality of the orthosis surface in contact with the hand for two layer thicknesses: 0.2 mm (Figure 8a) and 0.1 mm (Figure 8b).

The infill density influences the strength of the printing, which is even more relevant when discussing thin wall features, as is the case for WHOs. Figure 9 presents, as an example, an orthosis with elliptical pockets vertically oriented for slicing (Figure 9a). Two infill densities (100%—Figure 9b, and 60%—Figure 9c,d) are set here showing the source of defects caused by the orthosis design and sparse infill deposition within a layer when using the concentric pattern (Figure 9d).

Figure 10 shows the results of increasing the parameter called Overhand Angle for reducing the volume of the support structure, thus influencing the printing time. However, decreasing the support volume also has an impact on the stability of the orthosis during the printing process. Since the orthoses are thin and tall, the vibrations generated by the movements of the extruder nozzle may lead to detachment from the build platform, as illustrated in Figure 11 for sample 7, manufactured vertically, along with delamination caused also by the build orientation for this type of print (Figure 11c).

Table 4. Three-dimensional printing time for DLP-manufactured wrist–hand orthoses.

Sample	Levels	3D Printing Time
5	Layer height: 0.1 mm/Wall thickness: 1.2 mm Layer height: 0.2 mm/Wall thickness: 1.2 mm Layer height: 0.1 mm/Wall thickness: 1.2 mm Layer height: 0.1 mm/Wall thickness: 1 mm	3 h 44 min 1 h 52 min 3 h 43 min 1 h 52 min
8	Layer height: 0.1 mm/Wall thickness: 1.2 mm Layer height: 0.2 mm/Wall thickness: 1.2 mm Layer height: 0.1 mm/Wall thickness: 1.2 mm Layer height: 0.1 mm/Wall thickness: 1 mm	3 h 44 min 1 h 52 min 3 h 44 min 1 h 52 min
9	Layer height: 0.1 mm/Wall thickness: 1.2 mm Layer height: 0.2 mm/Wall thickness: 1.2 mm Layer height: 0.1 mm/Wall thickness: 1.2 mm Layer height: 0.1 mm/Wall thickness: 1 mm	3 h 44 min 1 h 53 min 3 h 44 min 1 h 53 min
10	Layer height: 0.1 mm/Wall thickness: 1.2 mm Layer height: 0.2 mm/Wall thickness: 1.2 mm Layer height: 0.1 mm/Wall thickness: 1.2 mm Layer height: 0.1 mm/Wall thickness: 1 mm	4 h 18 min 2 h 13 min 4 h 18 min 2 h 13 min

displays the printing time data for samples 5, 8, and 9 produced using the process parameters listed in Table 2 using the AnyCubic DLP 3D printer. In the context of this printer’s limited workspace size, optimizing the part’s orientation to reduce printing time was not investigated because the part had to be positioned within the available working space to ensure proper fit (Figure 12).

Considering the Voronoi design of samples 5, 8, 9, and 10, as well as their small thickness (2.7 mm and 3.7 mm), 100% infill density was used for DLP-produced orthoses.

3.2. 3DP-WHO Cost Analysis

Three-dimensional printed orthosis costs are also an important factor to analyze. These costs include not only production costs related to the quantity of materials used, energy costs associated with printing time, and the 3D printer amortization (or depreciation) costs, but also the design time. By far, the most significant cost component is the design time, especially for the first approach based on 3D scanning the patient’s upper limb, which requires specific engineering skills.

For example, the flat orthoses were generated using a freely available web app. The costs of orthoses 1, 4, and 6 were calculated based on material/filament consumption (52% of the total cost), energy consumption (24% of the total cost), 3D printer amortization (22% of the total cost), and post-processing (2% of the total cost). For orthosis 1, the material consumption was 42 g (13.92 m of filament), and the printing time was 4 h and 19 min at 80% infill density (Table 3), resulting in a cost of EUR 8.45. In contrast, the production costs for a ready-to-use orthosis, such as orthosis 3, are double due to a printing time of 8 h and 49 min at 100% density, a 0.2 mm layer thickness (Table 3), and 26.32 m of filament.

For DLP, orthosis 5, the cost was EUR 5.23 for 3 h and 43 min of printing time with a layer height of 0.1 mm and wall thickness of 1.2 mm.

Thus, it can be seen that the production costs are low, which is another benefit of 3D-printed orthoses. This opinion is shared in other studies, such as [23,24], which addressed 3D-printed orthosis (full cast) costs and concluded that they are half the price of conventional orthoses. The following mean printing time and costs were reported in [23]: 129 min and EUR 187 for 3D-printed orthoses, compared to 269 min and EUR 398 for conventional orthoses. A note should be made here that, in our study, orthoses are not full casts fully encircling the arm, but splint-like.

Cost-related aspects are also presented by Górski et al. [18], who discuss the influence of three printing strategies (layer, thickness, infill) and four types of filaments on mechanical resistance, printing time, and costs. The reported production costs were below a reference cost of USD 75 (conventional orthosis) with an acceptable printing time corresponding to two shifts (16 h).

3.3. 3D-Printed Wrist–Hand Orthosis Usability Results

Table 5. Users’ feedback on 3D-printed wrist–hand orthoses.

Criteria	User 1/ Samples 3, 4, 5, 8	User 2/ Samples 6, 7, 9
1 (Weight)	Sample 3: 4 Sample 4: 4 Sample 5: 5 Sample 8: 5	Sample 6: 4 Sample 7: 4 Sample 9: 3
2 (Design)	Sample 3: 5 Sample 4: 5 Sample 5: 5 Sample 8: 5	Sample 6: 4 Sample 7: 4 Sample 9: 5
3 (Secure)	Sample 3: 4 Sample 4: 5 Sample 5: 5 Sample 8: 3	Sample 6: 5 Sample 7: 5 Sample 9: 4
4 (Immobilization strength)	Sample 3: 4 Sample 4: 5 Sample 5: 4 Sample 8: 2	Sample 6: 5 Sample 7: 5 Sample 9: 3
5 (Comfort)	Sample 3: 4 Sample 4: 4 Sample 5: 5 Sample 8: 5	Sample 6: 4 Sample 7: 4 Sample 9: 5
6 (Functionality)	Sample 3: 5 Sample 4: 5 Sample 5: 4 Sample 8: 4	Sample 6: 5 Sample 7: 5 Sample 9: 4
Orthoses ranking	3, 4, 5, 8	6, 7, 9

shows the results of the usability assessment for the 3D-printed orthoses. As mentioned, the orthoses were designed for two users based on 3D scanning data and measurements of key dimensions of the hand (for the flat design), and the results are presented correspondingly.

As a general observation, all orthoses allowed the performance of tasks with no significant variation in the time required to complete them. Furthermore, both users, who had prior experience with traditional plaster casts, expressed a strong appreciation for the design of the 3D-printed orthoses, especially for the Voronoi design. Furthermore, both users considered that the surface quality of DLP-produce orthoses was superior to that of the orthoses produced by MEX.

4. Discussions

4.1. Process Settings’ Influence over 3D Printing Time

The results in Table 3 show that there were no significant changes in the printing time when switching between the infill patterns. For instance, in the case of sample 7, horizontally oriented, the printing time for the triangular infill was 16 h 35 min at 60% density, whereas for the zig-zag pattern, it was 16 h 28 min, and for the line pattern, it was 16 h 45 min. While it is acknowledged that the infill pattern does affect mechanical properties such as the compression or tensile strength of 3D prints [25,26], its influence on printing time remains minor, especially for the specific case of wrist–hand orthoses with thicknesses up to 4 mm. The data also prove that the printing time is quite large and justify the focus on finding solutions to reduce it. Additionally, the results from the Cura slicer indicated that decreasing the layer thickness from 0.2 mm to 0.1 mm (to improve surface quality) can result in a printing duration exceeding 24 h. Even if the 3DP-WHOs are not typically employed for emergency scenarios, a nearly day-long printing time is still not practical for this application.

Another interesting observation is that when using the Auto-orientation option in Cura (which operates by minimizing the print’s overhangs, thereby reducing the necessity for support structures) the orientation that also optimized the printing time was correct just for sample 2, while the orientation recommended by Meshmixer proved to be correct only for sample 7. This indicates that these optimization tools for build orientation are not consistently producing the expected outcome of reducing printing time.

The printing time for the flat orthoses was significantly less than the printing time of orthoses with similar designs but manufactured in the other orientations (for instance, samples 1 vs. 2, or samples 3 vs. 4, in Table 3). Furthermore, as demonstrated by [18], horizontally manufacturing ready-to-use orthotic shapes does indeed reduce printing time. However, this horizontal orientation negatively affects the surface quality by producing a staircase effect on curved surfaces, as can be seen in Figure 8a. Typically, this staircase effect can be mitigated by reducing the layer thickness to smaller values [27]. However, this solution proves inefficient when dealing with the complex shapes of customized wrist–hand orthoses. As an example, no obvious improvement could be noted in Figure 9b, while the orientation more than doubled the printing time.

Another unexpected result was observed for samples 2 and 3, where the printing time at 100% infill density was less than that for 80% infill density. This can be attributed to the complexity of the orthosis shape and the specific paths calculated by the slicer, which are followed by the nozzle to build each layer.

By adjusting the value of the Support Overhang Angle parameter, one can reduce the required support structure, saving time during manufacturing (Figure 10): 11 h 50 min for a 45° overhang angle, 10 h 25 min for a 60° overhang angle, and 6 h 32 min for an 80° overhang angle (at 60% infill density). In their study, Sala et al. opted for an 80° overhang angle when creating a custom 3D-printed wrist–hand orthosis with a 4 mm thickness manufactured in a vertical orientation [17]. It is worth noting that no printing difficulties were reported, likely due to the fact that their design was more robust with fewer open pockets.

Another approach to reducing 3D printing time involves decreasing the infill density, as previously mentioned. However, it is important to carefully analyze all such solutions, as they may lead to various defects, as illustrated in Figure 11, or as discussed in prior research where mechanical resistance to flexural fatigue was investigated [13]. Therefore, the recommendation when building ready-to-use orthoses in an upright position is to use 100% infill density to enhance the orthosis’ stiffness, and to use patterns like the zig-zag that can more effectively fill each layer compared to a concentric pattern, for instance. The need for printing with 100% infill density is not applicable to flat orthoses, where lower densities can be used without encountering the previously mentioned defects. Additionally, Popescu et al. [13] showed that the resistance to flexural strength of thermoformed 3DP-WHOs is better when compared to orthoses 3D-printed in their ready-to-use shape.

The results in Table 4 for DLP-printed orthoses indicate that there is no significant difference between samples 5, 8, and 9 when analyzing 3D printing time. These orthoses have a similar thickness but differ in pocket shapes and dimensions, suggesting that pocket design does not significantly impact printing time. However, the effect on mechanical resistance, another important factor, should be further investigated for DLP orthoses, as such research is lacking.

4.2. 3DP-WHO Usability Assessment

User 1 reported that the thin Voronoi DLP orthosis (2.7 mm thickness) provided insufficient immobilization. When comparing the immobilization capabilities of orthoses with a similar design (samples 3 and 5), but produced through different methods, it was observed that the orthosis manufactured through the MEX process exhibited greater stiffness due to superior material properties. The user prioritized the functional aspect more than the fact that sample 5 was lighter, more comfortable, and had a better surface quality. Despite the weight difference being less than 3 g, user 1 was able to discern the variation in weight between samples 3 and 4. This weight difference was due to the fact that the thermoformed version (sample 4) was printed with a 60% infill density, whereas orthosis 3 was printed with 100% infill density (sample 3). The fact that the flat orthoses can be manufactured with less than 100% infill density is an advantage in comparison to orthoses 3D-printed in their ready-to-use shape, as in [13], where it was shown that orthoses with 55% infill density have the flexural strength required for ensuring functionality.

The opinion of User 2 on the design of sample 7 (3D-printed vertically in ready-to-use form) indicated that the surface quality was not as satisfactory compared to sample 6 (similar design but 3D-printed flat and then thermoformed). In particular, the removal of the support structure left noticeable marks and some rough edges, as can be seen in Figure 11c. Securing Sample 9 on the hand was accomplished with three Velcro strips glued on the top of the orthosis, and this was rated by User 2 as less satisfactory than the securing of sample 6, for instance, which used the open pockets for inserting the strips. The 3.7 mm thick Voronoi sample 9 was felt as less rigid than the 2.7 mm thick hexagonal pockets of samples 6 and 7. In order of preference, once again, the DLP-produced orthosis was ranked last due to its lower stiffness in comparison with the MEX-produced orthoses, despite the fact that User 2 expressed a greater appreciation for the design offered by this model.

4.3. Guidelines for Producing 3DP-WHOs

A general observation from the investigations in this study is that these 3D-printed orthoses are characterized by intricate shapes, numerous open pockets, and reduced thickness, which enhance aesthetics and make them favored by users. At the same time, these features complicate the task of determining the optimal solution in terms of build orientation and process parameters. Both these investigations and the literature data show that balancing stiffness, surface quality, and reduced printing time remains a non-trivial challenge.

Based on the results presented, several guidelines for producing customized 3D-printed wrist–hand orthoses can be formulated, correlating build orientation and process parameter settings with printing time and surface quality. There are two main solutions to design the orthoses, namely based on 3D scans: a case in which the orthosis is 3D-printed in its ready-to-use shape or by orthoses that can be printed flat, and a case in which they need thermoforming to fit the patient’s hand. The first method requires design knowledge, while the second approach can be based on an online app to generate the flat model. The second approach is suitable for the MEX process but not for the DLP process. Flat orthoses can be manufactured in almost half the time of their ready-to-use counterparts.

The problem of build orientation should be addressed only for orthoses 3D-printed in their ready-to-use shape, and the decision should take into account the specificity of the manufacturing process as well as the parameter settings. Here, the main recommendation is to 3D-print the orthoses at 100% infill density, and to test several orientations for finding the one which provides a balance between printing time and surface quality. Manual adjustments may be necessary to achieve the best results.

Caution is required when minimizing the volume of the support structure (with the purpose of reducing material consumption and printing time) to avoid the orthosis’ detachment from the platform. The findings in this study showed that a lower infill density can create gaps inside these thin prints (a design characteristic of these medical devices) with a negative impact on mechanical resistance. When adjusting the overhang angle in MEX-produced WHOs, careful consideration should be given to the design, thickness, and total volume of ventilation pockets.

For the analyzed orthoses, the infill patterns proved not have any relevant impact on the printing time.

For MEX-produced orthoses, decreasing the layer thickness from 0.2 mm to 0.1 mm does not significantly improve surface quality but doubles the printing time. Therefore, a lesser layer thickness is not necessary, as shown by our investigations. Orthoses are produced faster by DLP, but the range of materials available for this process is less extensive than for MEX. Additionally, the build space for DLP might be unsuitable for orthoses with large dimensions, such as those required for a male forearm.

5. Conclusions and Further Work

This study investigated several aspects related to the design and printability (by means of MEX and vat photopolymerization—DLP processes) of custom wrist–hand orthoses, covering also the assessment of usability criteria, such as immobilization strength, design, comfort, and weight.

The main findings can be summarized as follows:

• Ten orthosis configurations were analyzed, with seven 3D-printed and tested by two users.
• Optimal build orientation and process parameters have an important influence over orthoses’ quality and stiffness, as well as printing time.
• Flat and thermoformed orthoses with lower infill densities printed efficiently, while tall, thin orthoses required 100% infill for stability.
• Setting a larger overhang angle in MEX-produced orthoses reduced build time but might cause detachment issues.
• Optimized orientations from Meshmixer and Cura did not consistently reduce printing times; horizontal orientation proved the fastest, but the surface quality was negatively influenced.
• Positive feedback was received for the Voronoi design, although DLP-produced versions were less stiff than MEX-produced ones.