1. Introduction
Pressure sensors are used extensively in surgical procedures, help to provide real-time feedback on the forces applied to the characteristics of tissues, and help to reduce accidents and the number of surgical instruments used [
1]. Piezoresistive sensors have been shown to be extremely valuable in the medical field, with sensors being used for pressure monitoring and mapping, feedback systems, and force sensing and analysis. Pressure mapping is used to monitor body positioning, including monitoring pressure injury (PI) development. In the literature, it is heavily documented that pressure sensors have been integrated or retrofitted into exiting mattresses to allow for monitoring, such as Mater et al. [
2], who designed a bed sheet containing textile pressure sensors that demonstrated a promising use in posture identification. The use of these types of sensors to monitor PIs in surgery has not been documented; however, the prevalence of PIs in surgical patients undergoing spinal surgery in the prone position was 23% [
3]. This is due to morphological changes, the patient’s immobility, and the lack of opportunity to reposition the patient. This supports the need to develop sensors that can be used in the surgical environment, specifically on soft foam surfaces that are used to support the patient’s positioning during surgery.
2. Materials and Methods
2.1. Design of Sensor Matrices
The developed sensor design follows a sandwich-type structure where, in the most basic description, there are three layers: a top electrode, a piezoresistive middle layer, and a bottom electrode layer (
Figure 1).
Inter-spitzen AG (FRTI) in St. Gallen, Switzerland manufactured and produced eight flat 4 by 4, flexible textile sensor matrices using a consistent manufacturing process and varied materials for comparison. The matrices, sized 230 mm L × 160 mm W in a rhombus layout, were integrated into Baxter Healthcare Corporation’s spinal table. The sensor matrices have a total of 16 sensor intersection points (
Figure 2). Each sensor intersection point’s sensing area is 10 mm
2. The maximum force applied was 200N over the entire matrix (and bolster during testing). Tests were based on an average body weight of 65 Kg, assigning 20 Kg per bolster, aligning with a maximum of 200N during testing.
2.2. Test Methodology
Eight prototypes were tested using a Shimadzu Mechanical Tester AG-X Plus (Shimadzu Corporation, Kyoto, Japan) with a 1KN load cell. Each prototype was secured onto a custom load cell rig to prevent movement, ensuring consistent testing methods. The setup remained identical for both foam and no-foam tests. A microcontroller (Arduino) was integrated with Processing IDE, facilitating data collection from all 16 sensor intersection points, recording analogue signal readings every 1000 milliseconds (
Figure 3).
2.3. Sensor Characterisation
In order to investigate the effect that a foam surface has on sensor outputs, the prototypes were subjected to compressive loads with data captured to compare the sensitivity, span, and nonlinearity. These three main sensor errors are comparable, whether the prototypes are on foam or on solid surfaces. Hysteresis was another important characterisation. However, these data were impossible to produce on foam, due to the foam recovery when compression was unloaded. Similarly, the intended use of the sensor does not require frequent on- and offloading of forces; therefore, hysteresis was excluded from the analysis.
3. Results
Data from all 16 sensors were collected; however, sensor intersection 7 (highlighted in red in
Figure 2) was analysed in more detail due its central position in the matrix, minimising the potential signal deviation found at the foam pad’s edge. Analysing the voltage output (
Figure 4a,b) compared between hard surfaces and foam showed notable differences. Overall, a non-linearity error decrease was observed across all prototypes when evaluated on foam, except for prototype 4 (
Table 1).
Table 1 highlights all the results for nonlinearity for all prototypes. These finding show that due to a load being applied to a foam structure, the foam will yield elastically and absorb the force being applied. The rate of compression for the foam tests were 20 mm/min, which was considerably faster than with no foam compression (0.5 mm/min). This could account for the instability in the signal, as the foam does not have enough time to recover.
Table 2 shows the sensitivity and span of the sensor outputs with and without the foam surface. The results from this are varied, with the sensitivity increasing and decreasing over multiple different prototypes. However, for prototype 4, the sensitivity is very similar for both the foam and no-foam results; similarly, the span is stable for both results.
However, the nonlinearity results of prototype 4 contradicted all the other results, which indicate that nonlinearity improves the foam structure. If we look at the prototype 2 and 3 results, we can see that they have the least nonlinearity error, and the span decreases and the sensitivity increases (the smaller value, the greater the sensitivity) when tested on foam. If we compare this to other sensor results, such as prototype 1, 7, and 8, where the sensitivity worsens on foam and the span increases, we can see that these sensors suffer from a high nonlinearity error. Therefore, in this investigation, we can assume that the discrepancies in the span and sensitivity are due to nonlinearity.
4. Discussion and Conclusions
In this investigation, the sensor characteristics of eight 4 by 4 textile pressure sensors were explored. The effect of foam on these custom sensors had varying effects due to the specific sensor error tests conducted in this investigation. From the results, two possible prototypes (prototype 2 and 3) were identified to continue to be developed for use on foam surfaces due to their characteristic of having a better nonlinearity error on foam. The implications of this research will help to further the development of custom textile pressure sensors to monitor PU development and monitoring during surgery.
Author Contributions
Conceptualisation, N.D.S.; methodology, N.D.S.; software and manufacture, N.D.S., P.S. and J.F.; validation, N.D.S.; formal analysis, N.D.S.; writing—original draft preparation, N.D.S.; writing—review and editing, P.B., S.L. and Y.W.; supervision, P.B.; funding acquisition, C.D. and F.L.S. All authors have read and agreed to the published version of the manuscript.
Funding
This research was funded by BAXTER HEALTHCARE. Match-funded Ph.D. NTU Sponsor License Number: 7EVPXBF3X.
Data Availability Statement
The data are available on request and are not currently downloadable as they are commercially sensitive.
Acknowledgments
Manufacturing of sensor matrices by FRTI. Testing supported by the technicians at Nottingham Trent University.
Conflicts of Interest
Authors Julia Fleischer and Pascal Stark were employed by the company Inter-Spitzen AG. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
References
- King, C.H.; Culjat, M.O.; Franco, M.L.; Lewis, C.E.; Dutson, E.P.; Grundfest, W.S.; Bisley, J.W. Tactile feedback induces reduced grasping force in robot-assisted surgery. IEEE Trans. Haptics 2009, 2, 103–110. [Google Scholar] [CrossRef]
- Matar, G.; Lina, J.M.; Kaddoum, G. Artificial Neural Network for in-Bed Posture Classification Using Bed-Sheet Pressure Sensors. IEEE J. Biomed. Health Inf. 2020, 24, 101–110. [Google Scholar] [CrossRef] [PubMed]
- Lin, S.; Hey, H.W.; Lau, E.T.; Tan, K.A.; Thambiah, J.S.; Lau, L.L.; Kumar, N.; Liu, K.P.; Wong, H.K. Prevalence and Predictors of Pres-sure Injuries From Spine Surgery in the Prone Position: Do Body Morphological Changes During Deformity Correction Increase the Risks? SPINE 2017, 42, 1730–1736. [Google Scholar] [CrossRef] [PubMed]
| Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).