Pre-post prosthetic and rehabilitation procedures are common before the amputee is provided with a permanent prosthesis [1, 4]. Generally, at the beginning, amputees are provided with information pertaining to the expected condition and timings of the rehabilitation procedure. After the surgery is finalized and the amputees have been forwarded to the occupational therapists, the pre-post prosthetic process begins [1, 4].
The amputees are provided with a temporary prosthesis and are required to perform common simple tasks before proceeding with general tasks such as picking items up, opening a door, zipping a shirt [3, 8]. This procedure is required to train the remaining muscle that is still active and allows sufficient time for the residual limb to become well shaped, which usually takes up to 6 weeks [1, 4].
The amputees are then provided with a permanent prosthesis and the rehabilitation of common daily task activities will continue for up to six weeks. Within the period, the amputees need to continue to perform general rehabilitation activities. It is common for amputees to revisit their therapist complaining of uncomfortable fitting within the first six to twelve months [4]. This problem usually occurs as a result of changes in the size and condition of the residual limb, which occur as a result of increases/decreases in the amputee’s body weight and height (Figure 2).In this study, the amputees and the therapist do not have to worried about the socket changing procedures. The design of air splint socket system help to auto-adjust the socket to the desired size and fitting needed by the amputees. At the same time, it reduced the consultation time for both parties and help to self-maintain the prosthetics. Figure 3 give the full flowchart on how the initial and final process in developing the air splint prostheses.
The air splint socket system basically uses a FSR pressure sensor [18] (Figure 4), which is placed on the surface of the air splint socket, to transfer any pressure detection data to the microprocessor and microcontroller-based system as the input data. The FSR pressure sensor is one of the most accurate and reliable measurement tools available to determine any contact pressure between the residual limb and the socket surface [18, 19]. The FSR pressure sensors use the received pressure wave to retain the input of contact within 0 kPa to 100 kPa in order to maintain the air splint system pressure accordingly to clinical principle [20, 21]. If the pressure increase more than 40 kPa, the blood system will be interrupted [20, 21]. A full illustration of the mechanism is shown in Figure 5. With the air splint system, the patient does not need to worry about changing the socket size and fitting, since the socket will change the size and fit accordingly within the desired contact of the residual limb.
The FSR pressure sensor that functions as the input will then send the generated data to the microcontroller system that is placed inside the upper elbow part. This part of the transhumeral also consists of an oscillometric pump that will generate the air volume that is required for the air splint or otherwise maintain it at 40 kPa [20, 21]. The power supply for the system comes from 9 V batteries, which are widely available, lightweight and long lasting.
The new prosthetic component was conceived to overcome the limitations imposed at the socket. The required length and mass of the prosthetic device was designed according to Drill’s and Contini’s anthropometry theorem [22]. The development mechanism is the result of a rigorous approach, which made it possible to optimize the functionality of the socket. The articulation consisted of the air splint, which replaced the thermoplastic as the main socket part (Figure 5). The air splint incorporated a silicon liner surface in order to provide the residual limb with increased gripping force. For their own comfort and satisfaction, the amputees can use a stocking net or add another silicon liner to the residual limb; this depends on the user themselves, since the air splint socket system will adjust the size according to the required size and fitting. The electronic parts were placed at the bottom part of the air splint socket. The microcontroller, the power supply and the motor controller were placed together at a convenient joint that could be readily accessed in the event that there was a need for service or reboot. The socket was also fitted with an USB cable port that could allow the user to restart or reboot the system in the event of any problems.
The minor part of the device consisted of an oscillometric air pump. The oscillometric air pump was lightweight and integrated in the upper elbow part. The weight and size of all the devices were designed according to the exact measurements of the normal hand [22]. This was intended to counter any advance force that was applied if the amputee used a prosthetic hand. The current body-powered prostheses eliminate this factor and cause a major defect to the shoulder size [6–9].
Ethics
The experimental protocol for this work was approved by the Ethical Community of the University Malaya Medical Centre (UMMC), Kuala Lumpur Malaysia. Written informed consent was granted by the participants from the authors for the publication. Approval ID: 829:15. One registered prosthetist fabricated all the prostheses to avoid alterations due to manufacturing, alignment and fitting. All the procedure of socket making and fitting involves the Certified Prosthetics and Orthotics (CPO) which had been recognized by International Society of Prosthetics and Orthotics (ISPO).
Experimental setup
The experiment had been conducted to discover and compare the dynamic interface pressure using static socket and air splint socket. The experimental setup used the Tekscan F-Socket sensor (9811E). The reason of using the F-Socket was to determine the pressure surface for its flexibility, rectangular printed, and the lowest thickness. Tekscan pressure mapping systems and measurement systems were used in this experiment to provide tactile pressure mapping and force data. The pressure mapping reacts as an input was first calibrated by using the Tekscan bladder. The applied pressure will then appear in the measurement system in the computer, and all related data, such as pressure, force, and time, will be calculated. The printed circuit with a thickness just about 0.18 mm made the sensor to easily fit in the gap between the socket and the surface of the amputation level. Figure 5 shows the the F-Socket which attached to the residual limb. For the transhumeral part, only two F-Socket needed to cover almost the entire socket surface attached to the residual limb. F-Socket sensor used the F-Scan software to operate and be connected to the rear of the PC via 762 mm long cable and a cuff unit (98 mm × 64 mm × 29 mm). The cable functioned as the converter of the analogue signal to digital signal so that the data may be read by the PC. In order to have the most accurate and reliable results, some precautions were considered such as by verifying all the connections were well-organised. The cable wire was tightened to the amputee’s body part to make sure that the surface of the sensor was not disrupted.
F-socket sensor calibrations
In this study, it was noted that every time the trial was done, the sensor was not changed, nor that a calibration was made out. For the transition between the socket and the amputation limit, the sensors that were already on the residual limb were only removed if any of the sensors was found to be defective. Before the F-Socket was fitted with the prostheses, the sensor was equilibrated and calibrated. The process of equilibrating the sensor is where the whole sensor point shares the equal amount of pressure to ensure that all 96 senses have a common output. The F-Socket was put into a pressure bladder in order to ensure that each area on the F-Socket had the similar criteria. The sensor was placed in the middle of the bladder and then was subjected to a pressure of 100 kPa by taking the specifications from the manufacturer.
Experimental procedure
After the process was completed, the sensor was then attached to the amputee residual limb so that the position of the sensor was stable (Figure 6). Silicone liners were used for both sockets, which require no reattachment when changing the socket. The sensor was attached by using the spray adhesive, a type of strong glue. As mentioned earlier, only two sensors were required to cover the area of the residual limb. The F-Socket attached only at the part of the humerus bones that were still left. During the installation of the F-socket to the amputee’s upper elbow, the main part was to confirm that the humerus of the upper elbow was well-attached to each sensor. Since the amputation part was only 40% left, the F-Socket sensor was trimmed horizontally to reduce the length of the sensor. This step was done to accommodate the subjects with shorter limb in order to obtain a tidier sensor placement, as well as to ensure there was no overlapped sensor. After the stockinet was fully fitted into the residual limb, then the socket was fitted into the stockinet. However, the position and the liner of the sensor stability must be validated so that the data collection was not interrupted.
After the amputee was comfortable with the fitting of the socket, the F-socket sensor connects to the portable to collect some data. The value recording has a vulnerable due to the external noise that may occur. This was due to the sensitivity of the sensor and the dimensions that were physically thin but to be fitted into a small interface space. Some unwanted noises usually occurred because of the bending position for the sensor itself. There were several methods to reduce the noise distraction. The first method is by setting up the noise reduction threshold in the Tekscan’s F-Scan. The value was set up to level 3 so that any values or data below or at this level will be filtered automatically. The second method is by removing any data that were collected without applying the pressure to the sensor. When the F-Scan detected the presence of any data of unmoving pressure, the data may be diminished and the calibration of the sensor was set to zero at that level. The third way to handle this problem is by applying individual measurement to each point of the sensor. Sometimes, one of the sensors gave a high pressure and surrounded by lower pressure points. To make it stable, all of the points can detect using the F-Scan and assigned to be in a level position to each other. Therefore, the data of pressure on the interface socket can be collected precisely and correctly.