According to the National Limb Loss Information Center (NLLIC) and the Limb Loss Research and Statistics Program (LLR&SP), there are approximately 133,235 to 185,000 amputations each year.1,2 The Amputee Coalition reported that there are currently two million individuals living with a limb loss in the United States.2 Over half of limb losses occurred due to vascular diseases.2 This may include diabetes and peripheral arterial diseases, but it is not exclusive.3 The incidence of upper-limb amputations was reported to be 5 in every 100,000 US population and accounts for 10% to 20% of all amputations.1,4 Within that population, 90% of upper-limb amputations are due to trauma.5 Other causes of upper-limb amputations include vascular diseases, osteosarcoma, and congenital deficiency. Congenital upper-limb loss occurs 1.6 times more often than lower limb and accounts for 1 in every 3846 live births.2 It was recently reported that 8% of all upper-limb loss are considered major, requiring appropriate prosthetic management as it influences the quality of life far more than lower-limb amputations.6 However, approximately 30% to 50% of individuals with upper-limb amputations do not wear their prosthesis regularly, and it is identified because of lack of education about options and care, lack of training, discomfort, poor cosmetic, and cost.4,7–10
Some of the most advanced upper-limb prostheses can cost a patient from $20,000 to $100,000 upon completion.11,12 As a result, the state-of-the art technology is not readily available to patients who can benefit from it. Instead, the majority of patients are still receiving conventional prostheses that are outdated, cosmetically unappealing, and cumbersome.13
The design of a conventional prosthesis can be dated back to the pre–Civil War era.8 The prosthesis is fabricated with a metal terminal device, the end piece. The terminal device is often a hook, which can be opened or closed voluntarily using a cable system worn around either one or both shoulders. When the cable is extended, the tension of the cable operates a lever that will either open or close the terminal device. Slack on the cable will allow the terminal device to return to its original position.5 The designs of conventional prostheses are bulky in nature and not uniform in comparison to newer myoelectric prosthetic devices. These advanced prostheses are designed with all of the components built directly into the terminal device that resembles a human hand. They are operated with a myoelectric control system, in which electrical signals produced by the muscles are detected and used as an output to control the prosthetic device.14 This control system is both a simple and intuitive interface for control of the prosthesis with high user satisfaction ratings.15,16 Myoelectric prostheses also have the potential to eliminate the shoulder harness and limit the use of the cable system, effectively simplifying the device.17 Simplification is beneficial for users of different learning capacities.18 In a comparative study between a conventional and a myoelectric hand, more than 60% of the subjects accepted the myoelectric prosthesis in preference to the conventional prosthesis.19 The recent myoelectric prostheses, although slower, have a better functional range of motion and provide more prehension types than the conventional prostheses.20,21 Having a better range of motion, or better movement potential around a joint or axis, is important to upper-limb prosthetic users. They lack the same degrees of movement in comparison to able-bodied individuals.22
Within the past decade, the introduction of three-dimensional (3D) printing has been highlighted as the technology that will change the future. Three-dimensional printing takes the concept of standard printing in one dimension further by enabling the printer to print 3D objects.23 Some of the different types of 3D printing are stereolithography (SLA), digital light processing (DLP), selective laser sintering (SLS), selective laser melting (SLM), electronic beam melting (EBM), laminating object manufacturing (LOM), and fused deposition modeling (FDM). Although there are many types and methods of 3D printing, the most common method is FDM, in which hot plastic is extruded through a nozzle to build an object in a predetermined, computer-guided way.23 Preliminary research demonstrated that 3D-printed prosthetic components such as the socket, the portion that interfaces directly with the patient, is comfortable and can be a viable method of fabricating prosthetic.24 Three-dimensional printing technologies have also gained public interest as online communities formed to share their ideas and innovations. One such community took particular interest in the development of affordable upper-limb prostheses that can be printed by average consumers. Their designs include both mechanically driven and advanced myoelectric devices, with the most costly prosthesis estimated to be under $500 to print and produce.25 These 3D-printed prosthetic hands can make functional hands more accessible to those who cannot afford the commercially made products.26
Previous research indicated that the myoelectric prosthesis allowed for easier completion of activities of daily living in comparison to a conventional prosthesis.27 In this pilot study, we compared the efficiency and functionality of a commercially available prosthetic hand, the i-limb, to a 3D-printed myoelectric hand, the Limbitless Arm. The purpose was to determine if 3D-printed myoelectric hands are a viable option by using the box and blocks test (BBT).28 If a more cost-effective prosthetic hand can be produced, it may allow for a greater number of people to have access to myoelectric terminal devices.
Twenty-four subjects with an average age of 26.1 ± 4.2 years old were enrolled in the present study. Subjects were students recruited by word of mouth at Loma Linda University in Loma Linda, California. The group of subjects included 14 men and 10 women. The mean height was 169.4 ± 9.1 cm, mean weight was 75.3 ± 16.1 kg, and mean body mass index (BMI) was 26.2 ± 4.8 kg/m2 (Table 1). All participants completed the BBT with the Limbitless Arm on visit 1 and the i-limb on visit 2. Subjects were tested on a first-come, first-serve basis. Inclusion criteria included all healthy right-hand dominant individuals between the ages of 18 and 60 years. Individuals were excluded from the study if they had any upper-limb motor impairment. The Institutional Review Board of Loma Linda University approved all procedures, and all participants were provided with a study information sheet before orally consenting to volunteer.
The i-limb Ultra, as shown in Figure 1, is a commercially available myoelectrically controlled prosthetic hand developed by Touch Bionics (Bioengineering Centre of the Princess Margaret Rose Hospital, Edinburgh, Scotland). The digits of the prosthetic hand are motorized independently to simulate the movements and dexterity of the natural human hand. For the purpose of this study, the prosthetic hand was attached to a custom fabricated device made from one-fourth–inch polypropylene to be used by able-bodied participants. There was one myoelectric sensor placed at the proximal anterior part of the forearm. Although there was the capability to place two sensors, one sensor was selected because it produced a similar function to the three-electrode placement of the Limbitless Arm. The pinch-grip pattern was selected. It was programmed to alternate between opening and closing upon detection of an electromyography (EMG) signal.
The Limbitless Arm, developed by UCF Armory (University of Central Florida, Orlando, FL, USA), is the first 3D printable myoelectrically controlled prosthetic hand (Figure 2). The hand was built according to the manufacturer's recommendation using an Arduino microcontroller, myoelectric muscle sensors, lithium polymer batteries, and a hobby servo. Instruction and support for this device was provided by e-NABLE, an online community dedicated to the advancement of 3D-printed prosthetic devices. Modifications were made to the Limbitless Arm's original design to allow it to be attached to a custom-fabricated device made from one-fourth–inch polypropylene for usage with able-bodied participants. Three electrodes were placed on the targeted flexor muscles as directed by the Limbitless Arm's Operating Manual. The grip pattern was modified to replicate the pinch grip of the i-limb ultra. It was programmed to alternate between opening and closing upon detection of an EMG signal.
In order for the subjects to use the i-limb and the Limbitless Arm, two similar custom devices were fabricated. These devices encompassed the participant's whole arm, starting just distal to the elbow joint. An opening was created on the medial side for the participant to don and doff the device. A square shell was formed at the distal portion to accommodate the hand of the participants. The prosthetic hands were mounted on the devices as shown in Figure 3. The devices were created and standardized using a mold of a male arm, which we believe was an average size. Both devices were created from the same mold and followed the same trimlines. They were made out of one-fourth–inch polypropylene plastic with padding on the inside for comfort. For suspension, two 2-inch Velcro straps were used at the forearm.
MAKERBOT 3D PRINTER
The MakerBot Replicator 2 (MakerBot Industries, Brooklyn, NY, USA) was used to print the Limbitless Arm (Figure 4). The printer is capable of printing polylactic acid (PLA) plastic and thermoplastic elastomer (TPE). The PLA filaments used for print were MakerBot's own, and the TPE filament was purchased from NinjaFlex (Fenner Drives Inc, USA). Print files were downloaded from e-NABLE, and all print settings were adjusted as directed by the Limbitless Arm's Operating Manual.
BOX AND BLOCKS TEST
The BBT was used to assess the efficiency of each prosthetic hand. It was originally devised to assess unilateral gross manual dexterity.29,30 The test consists of transferring 2.5-cm cubes located in a rectangular box from one partition to another during a 60-second trial. Scores are calculated based on the numbers of blocks transferred. For the purpose of this study, a higher score demonstrated better efficiency, dexterity, and usage of the prosthetic hand.29 Professional medical associations such as the Parkinson's Taskforce (PD EDGE), Spinal Cord Injury Taskforce (PD EDGE), Neurology Section of the American Physical Therapy Association's Multiple Sclerosis Taskforce (MSEDGE), Traumatic Brain Injury Taskforce (TBI EDGE), Stroke Taskforce (StrokEDGE), and Vestibular Taskforce (VEDGE) have recommended the usage of this test. It has also demonstrated to have test-retest reliability (r = 0.98), interrater/intrarater reliability (r = 1.0), criterion validity (r = 0.64), and construct validity (r = 0.80).28,30–32
All participants were scheduled to complete the BBT using the Limbitless Arm first, before returning after a 2-week crossover time to complete the BBT again using the i-limb. Testing took place at the Loma Linda East Campus Outpatient Rehabilitation Center. Upon arrival on the scheduled day and time, researcher 1 greeted and escorted the participant to a private testing room free of distractions. The participant was instructed to wash his or her hands and forearms with soap and water before entering to eliminate any oil or lotion on the skin that might interfere with the equipment. Once the participant entered the room, he or she was instructed to sit at the table facing a rectangular box. The box was divided into two square compartments of equal dimensions using a partition. One compartment consisted of 150 2.5-cm colored wooden blocks while the other compartment was empty. The researcher informed the participant of the rules of the BBT. Once researcher 2 fitted the device onto the participant, researcher 3 provided training on how to operate the prosthetic hand. Each participant was allotted a maximum of 5 minutes to practice.
During the BBT, the participant was allowed 60 seconds to move as many blocks as possible from one compartment to the other. A point was awarded for each block successfully transferred. In addition, for the blocks to count, the prosthetic hand must clear the partition before the blocks are dropped into the second compartment. Multiple blocks carried over during an attempt were only awarded one point. Blocks that bounced out of the second compartment during the attempt were considered as a successful transfer and were awarded a point. The test consisted of two trials with a 30-second break in between. After the completion of the second trial, the participant was reminded to return in 2 weeks to complete the BBT with the i-limb as scheduled. All data were recorded and collected by researcher 4.
Data were analyzed using the statistical package SPSS for Windows, Version 22.0 (IBM, Armonk, NY, USA). The general characteristics of the subjects were summarized using frequencies and relative frequencies for categorical variables and mean + SD for quantitative variables. The normality of the outcome variables was examined using the one-sample Kolmogorov-Smirnov test. An independent t-test was used to compare the number of blocks between men and women. For the two trials, the mean number of blocks between the Limbitless Arm and the i-limb was compared using the paired t-test. The level of significance was set at P ≤ 0.05.
The demographic characteristics of the subjects are displayed in Table 1. Twenty-four subjects, mean age 26.1 ± 4.2 years, participated in the study. There were 14 men (58.3%) and 10 women (41.7%). There were two trials in the study; all participants completed both trials. For either trial, the mean number of blocks using the Limbitless Arm was significantly lower than i-limb (trial 1: 8.4 ± 3.6 vs. 12.9 ± 3.3, P < 0.001; trial 2: 8.3 ± 3.6 vs. 13.8 ± 4.1, P < 0.001; Figure 5). The number of blocks successfully transferred by the i-limb improved by 53.57% in trial 1 and 66.27% in trial 2 over the Limbitless Arm as shown in Figure 6. Similar findings were obtained when we ran the analyses separately for men (trial 1: 9.1 ± 3.3 vs. 12.9 ± 3.7, P = 0.01; trial 2: 9.6 ± 3.2 vs. 14.1 ± 4.7, P = 0.02) and women (trial 1: 7.5 ± 3.9 vs. 12.8 ± 2.9, P = 0.00; trial 2: 6.3 ± 3.4 vs. 13.4 ± 3.2, P = 0.00). In trial 2, using the Limbitless Arm, there was a significant difference in the mean number of blocks between men and women (9.6 ± 3.0 vs. 6.3 ± 3.4, P = 0.03; Figures 7, 8, and 9).
The current study compared the Limbitless Arm to the i-limb using the BBT to determine their efficiency. The results showed that i-limb was significantly more efficient than the Limbitless Arm. Furthermore, the results showed similar findings when statistical analysis was performed based on sex. Results also indicated that the differences in the number of blocks for trial 2 were more significant for women than men.
Overall differences in results can be attributed to the early design of the 3D-printed prosthesis. The Limbitless Arm is still in its early stage of development. It was one of the few working prostheses available. It was also fabricated and operated with parts such as servos, batteries, sensors, and microcontrollers that were available at a local hobby store. The servos and mechanics of the Limbitless Arm are easily outperformed by the industrial servos and parts found within the i-limb. With further refinements from the Internet community and future versions, there is potential for the efficiency of the Limbitless Arm to improve.
During testing trials for the i-limb, 25% of the participants reported that the device was heavier than the Limbitless Arm, causing faster fatigue. Another factor that possibly caused fatigue was the sensors were not detecting the muscle contractions accurately. This resulted in the participant overworking the muscles while trying to activate the hand. Surprisingly, this did not impact their performance, as participants performed more efficiently using the i-limb. In particular, five of six participants who reported the weight difference were women, yet results indicated that women had a higher significance in mean number of blocks successfully transferred. The research team hypothesized that despite the weight difference, the overall size and shape of the i-limb was more compatible to the female body shape and size, which may have influenced their performance. Previous studies have documented the importance of cosmetic appearance and the psychological impact it has on rehabilitation for patients with amputation(s).33 Furthermore, a study indicated that most patients prefer a prosthesis that is lifelike.34 One recent study indicated that there was a high level of attraction to prostheses that have a high level of human likeness.35
The differences between EMG signal detection for the two prosthetic hands are important aspects of their function. The Limbitless Arm required the correct positioning of three electrodes to create a signal input, whereas the i-limb uses a single sensor that has three built-in electrodes to produce the same outcome. The placement of electrodes is crucial to the successful operation of the devices. Previous studies stressed the importance of electrode placement, indicating that shifting during movement can interfere with EMG signals.36–38 The Limbitless Arm required three separate attachment sites allowing for a higher potential for inaccuracy. Slight misplacement of these electrodes can alter the effectiveness in which they detect muscle activities, thereby influencing the operation and performance of the device. Unlike the Limbitless Arm, the electrode for the i-limb was a single unit, which allowed for easier adjustments. The potential for error was reduced, as the researcher only had to concentrate on the placement of a single sensor. In addition, the i-limb's sensor was created for extended usage and embedding in a prosthetic device. Since the Limbitless Arm used disposable sensors, a new set of sensors had to be applied each time it was taken off. Reapplication of the old sensors also reduced its effectiveness in detecting muscle activities.
Because of the motorized digits of the i-limb, the digits were tightly fixed and firmer than the digits of the Limbitless Arm. We suggest that this allowed the participant to grab the blocks with better efficiency and more accuracy. In addition, the rubber ends of the fingertips provided grip to grasp the blocks during the process. On the contrary, the joints of the digits of the Limbitless hand appeared limp and lacked rigidity. We suggest that this contributed to the participants not being able to grab the blocks, or maintain the grip to complete a successful transfer. Another benefit of the motorized digits in the i-limb is the ability of the grasp to operate proportionally with the duration in which the muscle signal is detected. The participants were able to have better control and precise movements. The Limbitless Arm did not have this feature.
LIMITATIONS TO THE STUDY
There were some limitations to the study. First, the custom attachment was a one-size-fits-all. It allowed our able-bodied participants to use the prosthetic hands. However, it did have limitations such as being bulky, heavy, and did not fit everyone properly. The fitting issues encountered included some participants' forearms being too big for the device, or their arms being too short. The opening and straps helped by accommodating these fitting issues. Second, the wooden handle, which the participants grasped during the tests, was causing pain. Therefore, an adjustment was made to the handle by creating a foam shell around it. This provided padding and comfort for the participants during the tests.
The results of the study and the observations made by the researchers suggested that the i-limb was more efficient than the Limbitless Arm. However, this does not eliminate the potential of 3D-printed prostheses as a viable option for prosthetic care. With 3D printing increasing in technological advancement, improvements are imminent. This pilot study opens the door for future research opportunities to help advance the knowledge of 3D-printed prostheses. Improvements in 3D-printed prosthetic hands, coupled with the growing knowledge of prosthetists, can be a cost-effective clinical approach to providing whole-person care. However, at this point, myoelectric hands such as the i-limb are more practical and effective. Further research is needed to compare the two prosthetic hands on individuals with amputations to better understand the capabilities and efficiency of the 3D-printed device.
We thank our principal investigator Michael Davidson, MPH, CPO, for supporting and assisting us throughout our research, providing valuable feedback, insight, and comments to our study. We also thank Gurinder Bains, MD, PhD, for guidance as our research instructor, and Noha Daher, DrPh, for assisting with our statistical analysis.
1. Dillingham TR, Pezzin LE, MacKenzie EJ. Limb amputation and limb deficiency: epidemiology and recent trends in the United States. South Med J
2. NLLIC, LLR&SP. Limb loss in the united states. Amputee Coalition of America
2007. Available at: http://www.amputee-coalition.org/fact_sheets/limbloss_us.pdf
3. Ziegler-Graham K, MacKenzie EJ, Ephraim PL, et al. Estimating the prevalence of limb loss in the United States: 2005 to 2050. Arch Phys Med Rehabil
4. McFarland LV, Hubbard Winkler SL, Heinemann AW, et al. Unilateral upper-limb loss: satisfaction and prosthetic
-device use in veterans and servicemembers from Vietnam and OIF/OEF conflicts. J Rehabil Res Dev
5. Smith DG, Michael JW, Bowker JH. Atlas of Amputations and Limb Deficiencies: Surgical, Prosthetic, and Rehabilitation Principles
. American Academy of Orthopaedic Surgeons: Rosemont, IL; 2004.
6. Behrend C, Reizner W, Marchessault JA, Hammert WC. Update on advances in upper extremity prosthetics. J Hand Surg [Am]
7. Melendez D, Leblanc A. Survey of arm amputees not wearing prostheses: implications for research and service. J Assoc Child Prosthet Orthot Clin
8. Billock J. Upper limb prosthetic
terminal devices: hands versus hooks. Clin Prosthet Orthot
9. Biddiss EA, Chau TT. Upper limb
prosthesis use and abandonment: a survey of the last 25 years. Prosthet Orthot Int
10. Biddiss E, Chau T. Upper-limb prosthetics: critical factors in device abandonment. Am J Phys Med Rehabil
11. McGimpsey G, Bradford TC. Limb prosthetics services and devices. Critical unmet need: market analysis white paper. Bioengineering Institute Center for Neuroprosthetics - Worcerster Polytechnic Institution
12. Myoelectric prosthesis offer advantages. Orthopaedic Surgery News at UCSF Medical Center
2006. Available at: https://www.ucsfhealth.org/newsletters/orthopaedic_surgery_news/winter_2006/myoelectric/
13. Berning K, Cohick S, Johnson R, et al. Comparison of body-powered voluntary opening and voluntary closing prehensor for activities of daily life. J Rehabil Res Dev
14. Scott RN. Myoelectric control of prostheses and orthoses. Bull Prosthet Res
15. Dosen S, Markovic M, Somer K, et al. EMG Biofeedback for online predictive control of grasping force in a myoelectric prosthesis. J Neuroeng Rehabil
16. Engdahl SM, Christie BP, Kelly B, et al. Surveying the interest of individuals with upper limb
loss in novel prosthetic
control techniques. J Neuroeng Rehabil
17. Weaver SA, Lange LR, Vogts VM. Comparison of myoelectric and conventional prostheses for adolescent amputees. Am J Occup Ther
18. Bouwsema H, van der Sluis CK, Bongers RM. Learning to control opening and closing a myoelectric hand
. Arch Phys Med Rehabil
19. Stein RB, Walley M. Functional comparison of upper extremity amputees using myoelectric and conventional prostheses. Arch Phys Med Rehabil
20. Dalley SA, Bennett DA, Goldfarb M. Functional assessment of the Vanderbilt Multigrasp myoelectric hand
: a continuing case study. Conf Proc IEEE Eng Med Biol Soc
21. Fougner AL, Stavdahl O, Kyberd PJ. System training and assessment in simultaneous proportional myoelectric prosthesis control. J Neuroeng Rehabil
22. Major MJ, Stine RL, Heckathorne CW, et al. Comparison of range-of-motion and variability in upper body movements between transradial prosthesis users and able-bodied controls when executing goal-oriented tasks. J Neuroeng Rehabil
23. Kaur S. How is "Internet of the 3D printed products" going to affect our lives. Iete Tech Rev
24. Herbert N, Simpson D, Spence WD, Ion W. A preliminary investigation into the development of 3-D printing of prosthetic
sockets. J Rehabil Res Dev
26. Zuniga J, Katsavelis D, Peck J, et al. Cyborg beast: a low-cost 3d-printed prosthetic
hand for children with upper-limb differences. BMC Res Notes
27. Probsting E, Kannenberg A, Conyers DW, et al. Ease of activities of daily living with conventional and multigrip myoelectric hands. J Prosthet Orthot
28. Desrosiers J, Bravo G, Hebert R, et al. Validation of the box and block test as a measure of dexterity of elderly people: reliability, validity, and norms studies. Arch Phys Med Rehabil
29. Dawood A. Rehab measures: Box and block test. Rehabilitation Measures Database
2010. Available at: http://www.rehabmeasures.org/Lists/RehabMeasures/DispForm.aspx?ID=917
30. Mathiowetz V, Volland G, Kashman N, Weber K. Adult norms for the Box and Block Test of manual dexterity. Am J Occup Ther
31. Chen HM, Chen CC, Hsueh IP, et al. Test-retest reproducibility and smallest real difference of 5 hand function tests in patients with stroke. Neurorehabil Neural Repair
32. Lin KC, Chuang LL, Wu CY, et al. Responsiveness and validity of three dexterous function measures in stroke rehabilitation. J Rehabil Res Dev
33. Bhuvaneswar CG, Epstein LA, Stern TA. Reactions to amputation: recognition and treatment. Prim Care Companion J Clin Psychiatry
34. Marks LJ, Michael JW. Science, medicine, and the future: artificial limbs. BMJ
35. Sansoni S, Wodehouse A, McFadyen A, Buis A. The aesthetic appeal of prosthetic
limbs and the uncanny valley: the role of personal characteristics in attraction. Int J Design
36. Beck TW, Housh TJ, Cramer JT, et al. Electrode shift and normalization reduce the innervation zone's influence on EMG. Med Sci Sports Exerc
37. Beck TW, Housh TJ, Cramer JT, Weir JP. The effect of the estimated innervation zone on EMG amplitude and center frequency. Med Sci Sports Exerc
38. Rainoldi A, Melchiorri G, Caruso I. A method for positioning electrodes during surface EMG recordings in lower limb muscles. J Neurosci Methods