Design, Manufacture and Validation of a Spirometry Device aimed for Low-Resource Settings

Design, Manufacture and Validation of a Spirometry Device aimed for Low-Resource Settings | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Design, Manufacture and Validation of a Spirometry Device aimed for Low-Resource Settings James Wallace, Pedro Checa Rifá, Thanusan Kannathasan, Charles F Hayfron-Benjamin, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6812011/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Medical devices are essential for maintaining resilient health systems worldwide. However, their distribution does not reflect this, with as many as 76% of devices being used by 13% of the world’s population. Most devices, being designed for high-income countries and failing to consider the local needs of other settings, often fail when deployed in other contexts. A frugal approach can be used to design more resilient systems. Spirometers show promise for developments in this area, owing to the high burden of respiratory conditions in low- and middle-income countries and the unavailability of frugal versions that can enable local care in such settings. Therefore, this work aims to develop an affordable spirometer for low-resource settings. Methods: A frugal approach was used to design a Venturi-style spirometer based on Bernoulli’s principles, leveraging 3D-printed parts and components such as an Arduino Uno R3 and differential pressure sensor (MXP5010DP). The accompanying software to process the signal and calculate relevant variables was developed on Arduino IDE and MATLAB. The device was validated via a hand-pump test and with an initial usability study on two subjects together with a CE-marked benchmark. The mean signed percentage error was then calculated to evaluate accuracy. To determine whether the input flow affected the measured volume Spearman’s rho and p values were calculated. Results: The device was successfully 3D printed from PLA, and then assembled with electronic components. All the software functioned as expected when it was run. In the two-litre pump test, the device achieved a mean reading of 1.983 L (true value 2L) and an accuracy of 1.53% (mean absolute percentage error). The Spearman’s Rho test revealed that there was no significant correlation between flow speed and volume. In the usability study, the device achieved readings similar to those of the CE marked device, and all values fell within normative values. Conclusions: This study designed, prototyped and validated a spirometer that can be used as a lung screening tool that achieves high accuracies comparable to those of other portable spirometers. This study also validated that frugal engineering could reduce the cost of devices without affecting the clinical accuracy. Medical device Spirometer Spirometry Respiratory diseases Asthma COPD Low-resource settings Resource-Limited Settings Low- and Middle-Income Countries Frugal innovation Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Medical devices and health technologies are crucial in building resilient and effective health systems. They can enhance diagnosis and treatment, improve patient safety and care quality, and boost efficiency and productivity. Modern medical devices increasingly generate data and integrate with consumer technologies, expanding their capabilities (1). These technologies are used across a wide range of functions—from diagnosing and monitoring conditions to supporting individuals with disabilities and treating both acute and chronic illnesses. Currently, the global market includes an estimated 2 million different types of medical devices, which are grouped into over 7,000 generic categories (1). However, inequities persist between and within regions in the application of medical devices and technologies in healthcare. For instance, high-income countries (HICs), home to only 13% of the world’s population, account for 76% of global medical device use (2). Regional disparities in healthcare expenditure—US$641.30 per capita in low- and middle-income countries (LMICs) compared with US$6,943.08 per capita in HICs—as of 2022 (3) significantly affect the affordability, accessibility, and effective use of medical devices and technologies. In LMICs, approximately 80% of medical devices are donated by international governmental and nongovernmental organisations (4). However, only 10–30% of these donated devices become fully operational, largely because they often fail to meet the specific contextual needs of recipient countries—whether due to inappropriate technical specifications, lack of supporting infrastructure, or insufficient staff training (5, 6). This heavy reliance on external donations can also compromise long-term health system sustainability by creating supply chain vulnerabilities, hindering the development of local capacity for innovation and maintenance, and reducing autonomy in healthcare planning and delivery (7). Additionally, the anticipated global population growth, particularly in LMICs (8), underscores the need for proportional universalism in the development and distribution of medical devices, ensuring increased availability and contextually tailored designs in regions with the highest need. The application of frugality in the design of medical devices for managing conditions, such as respiratory ones, is crucial for improving access to life-saving treatments, strengthening health systems and reducing the environmental impact, especially in resource-limited settings where the burden of such conditions is greater (9, 10). This approach focuses on creating cost-effective and reliable solutions that address the specific needs and constraints of resource-limited settings (11), as proven in these specific examples (12)(13)(14)(15). Respiratory conditions remain a crucial global health issue, with lower respiratory infections ranking 5 th among the leading causes of death in 2021 (16). The burdens of these conditions are greater in LMICs (17). For example, lower respiratory tract infections were the number one cause of death in LICs in 2021 (16). A staggering 90% of chronic obstructive pulmonary disease (COPD) deaths among those aged under 70 years occurred within LMICs in 2021(18). Similarly, a staggering 96% of global asthma deaths also occurred in LMICs in 2019 (19). Increasing access to low-cost, context-specific medical devices in LMIC settings could help narrow regional disparities by reducing underdiagnosis/misdiagnosis and enhancing early detection and more effective disease management (20). Currently, the most commonly used devices for diagnosing respiratory diseases include peak flow meters, spirometers and pulse oximeters (21). Among these, spirometers are the most versatile, offering a more comprehensive assessment of lung function by generating full volume time curves, rather than isolated values such as peak flow. They primarily measure key indicators such as forced expiratory volume in one second (FEV1), forced vital capacity (FVC) and peak expiratory flow (PEF), providing the most reliable data for differentiating between conditions such as COPD and asthma (22). Despite their diagnostic value, spirometers remain underutilised in low-resource settings (LRS) because of relatively high costs and the need for trained personnel to operate and interpret the results (23). These challenges make spirometers strong candidates for frugal innovation—specifically, the development of affordable, durable, and easy-to-use versions tailored to the needs of LMICs for the effective diagnosis and management of respiratory diseases. There remains a lack of portable, affordable spirometers suitable for clinical use in LRS. Studies highlight not only cost barriers but also issues with data interpretation and the need for more user-friendly, technologically integrated devices (24-26). A review by Carpenter et al. (25) assessed 16 portable spirometers (priced $99–$1,390 US dollars), revealing limited information on data security, accuracy, and patient outcomes. Most devices failed to account for users’ understanding, making results such as FEV1 and PEF difficult to interpret without reference values, and also faced usability and interface challenges. Ferreira Nunes et al. (27) developed a Venturi-based home spirometer using an ESP32 microcontroller, a high-precision Honeywell SSCMRRN005PDAA3 pressure sensor, and 3D-printed components. While the device demonstrated excellent technical performance, showing high correlation with a gold-standard spirometer (intraclass correlation coefficient of 0.987 for FVC), its suitability for LRS was limited because of the expensive pressure sensor (£74.17), lack of onboard data display, and limited environmental robustness. Open ports, for example, could allow dust intrusion. These issues underscore the need for frugal, user-centred designs that improve affordability, durability, and ease of use in LMIC settings. Our study, therefore, aimed to develop an affordable, durable and effective spirometry device tailored for resource-limited settings. Adopting a frugal innovation approach, we leveraged modern technologies such as 3D-printing and affordable electronics to create a cost-efficient solution. Our design is grounded in a user-centred methodology, prioritising the needs of primary care physicians to ensure accuracy, contextual relevance, ease of use, and broad accessibility. We focus on LRS in LMICs, particularly sub-Saharan Africa, but recognise the potential for use in low-resource healthcare systems of HICs. Methods This section details the methods used to design, prototype, and validate the hardware and software components of the spirometer. Hardware Contextualised design Building on prior research and expertise in designing medical devices for LRSs (7, 11) (and the examples mentioned in the introduction), the spirometer was developed with a user-driven approach, specifically tailored to the needs and constraints of such environments. Two focus groups were conducted to consult 10 biomedical engineering and medicine students from both HICs and LMICs, to inform the co-creation of this device. This was supported by discussions with clinicians from LMICs who use spirometry to treat patients with respiratory conditions. Key decisions from these consultations included adopting a differential pressure sensing system to eliminate the need for moving parts, such as turbines, thereby increasing the device's durability and suitability for low-resource environments. While turbine flow and rolling seal systems were considered, they were ultimately dropped due to their fragility and susceptibility to dust and humidity. Another important decision from the consultation was to minimise the number of components and ensure that most parts could be locally sourced or produced via 3D-printing. Finally, the design prioritised ease of use and maintenance to enhance usability in diverse settings. Device key components A 3D-printed venturi tube to generate the pressure difference; A 3D-printed casing that ensures airtight seals for the device and locks together to protect components; An electronic board comprising of an Arduino UNO R3, differential pressure NXP sensor (model number:MPX5010DP) and indicator electronics such as light emitting diodes (LEDs) and a liquid crystal display (LCD). Tube design A custom tube was designed using Autodesk Fusion 360, which was aligned with our key design priorities. The tube design follows on Bernoulli’s principle (28), incorporating two distinct internal cross-sectional areas, so that when air flows through it, a differential pressure is created at 2 points, relative to the different cross-sectional areas. This differential pressure is correlated with the to a flow rate, as per Equation 1 (28). Where: Q is the volumetric flow rate (m 3 /s), A 1 and A 2 are the cross-sectional areas of the cylindrical tube at the level of two different ports (m 2 ), p 1 and p 2 are the pressures at the same points (Pa), and rho is the density of air, set at 1.204 kg/m 3 under standard conditions (i.e., 20° C, 101.325kPa) (29) . Based on Equation 1, the two inner diameters of the tube can be found in order to work efficiently with the selected pressure sensor, namely MXP5010DP, which has a pressure range of 10 kPa. The correct design of this would lead to reduced error and better resolution, reducing the amount of pre- and post-processing needed. It was deemed essential to ensure that the maximum pressure difference effectively present between the two ports of the tube was as compatible as possible with the range of the sensor, avoiding situations in which the sensor could either be damaged (over pressure) or could give out readings largely affected by the accuracy. For this purpose, factors such as the maximum user PEF, disposable mouthpiece size, and range of the sensor were all considered. Another key requirement was the minimum internal diameter of the Venturi tube. The diameter should not be too small to avoid creating a large air flow resistance during exhalation, making the test uncomfortable and affecting measurement accuracy. In addition, the chosen diameter needed to promote smooth air flow through the device, minimising turbulence that would increase sensor noise. Turbulent flow introduced by non-smooth transitions could increase the noise in the sensor and further increase the perceived resistance and accuracy of the measurements. This was done by following the Venturi Tube specifications from the British Standards Institute (30). In particular three main parameters were used: first, the diameter of the throat/restrictive part should not be less than 0.224D (D being the entrance diameter) of the entrance diameter; the conical converging section should have a taper of 10° ± 5°; and finally, the device should have a divergent outlet of not less than 5° and have a minimum length of 1.5 times the throat diameter. All these factors are key to ensuring high accuracy. Electronic Design An Arduino UNO R3 board was chosen to acquire the data via the MXP5010DP differential pressure sensor. These boards are extremely cost-effective solutions compared with traditional microprocessors, costing just £15. Given that they are resilient and have low power consumption, they are ideal for frugal innovation purposes. The circuit for this design would be relatively simple since the sensor only requires 3 pins to be connected: Vout, which would connect to the analogue input pins in the Arduino; a ground pin; and a 5V pin. These would be interfaced by a breadboard. Additionally, to increase the usability of the design, 3 LEDs were added to act as indicators for the user as they perform the spirometry test. A red LED to indicate that there is no serial connection between the device and the software, a yellow LED to indicate that the device has established a serial connection and should press the start button to begin data collection, and a green LED to indicate that the user should start exhalation. An LCD screen was used in conjunction with to indicate more specific commands and return spirometry variables at the end. Software Software was developed to collect and process the data from the hardware and return the spirometry values to the user. To do this, an overall flow diagram was developed to describe the steps from data collection to the return of processed data (Figure 1). With respect to software, two sets of code were developed, one within the Arduino for reading and transferring data and one within MATLAB for data analysis and display. Using this consistent format, returned spirometry data will always be comparable with those of other tests. Similarly, using the workspace feature associated with MATLAB, trials can be run multiple times, changing input variables within the code to determine the optimum settings for spirometry flow analysis. Testing and Validation To test the accuracy of the device and software, two tests were planned, namely one to measure its accuracy in a standard lab test using a two-litre handpump, and one to perform an initial usability study for the device. Two-Litre handpump Test This test was performed to measure the accuracy of the device at measuring a known volume of air. To do this, the two-litre handpump was connected to the manufactured spirometry tube via a 3D-printed friction-fitted attachment that was designed to eliminate air loss between the pump and the tube. The tube was secured firmly to a table so that it lay horizontally. This eliminated possible fictitious readings that could be generated by moving the tube through the air; similarly, objects behind the device were removed to prevent any reflections of air back into the device. With the experimental setup complete, the handpump was fully extended, and the MATLAB program was run. Moreover, the pump handle was compressed until the air was fully expelled from the handpump through the device. The experiment was run 30 times at varying speeds, and the data were saved and recorded for later analysis after each trial. The varying speed helped us investigate whether the flow had a significant influence on the accuracy. This variable would also verify whether the drag equations had been modelled correctly, since the PEF is correlated with drag. The speed range investigated would be hard to determine since this was calculated after the test. However, the duration over which the full 2 litre could be monitored and recorded and would correlate to the flow rate. The range of duration investigated was from 0.3s to 2.5 seconds. The length of duration was monitored and recorded in MATLAB, by assessing the length on the time axis of the flow rate time graph that was above the threshold value during expiration. The accuracy was calculated via the mean percentage error. Moreover, further comparisons between the signed percentage error and the duration of each blow were made to investigate whether the speed of the input flow affected the accuracy. As a first step, the normality of the data was tested with the Shapiro-Wilk test (31), with a p-value lower than 0.05 indicating a non-normal distribution. The variable distribution proved to be non-normal (W = 0.6758, p-value < 0.001). Subsequently, Spearman’s rho (32) and its p-value were calculated. P-values lower than 0.05 were considered significant. Early usability study on the effectiveness of the spirometry device as a lung function test This test was performed to carry out an early usability study on the spirometer. For this purpose, the test involved recruiting two participants and asking them to simulate the use of the spirometer after receiving instructions. In particular, the user had to stand and take as deep a breath as possible and exhale this air through the tube as quickly as possible, performing the so-called “full forced expiration”. For hygienic reasons, disposable filtered mouthpieces were used (https://www.numed.co.uk/products/disposable-bacterial-viral-filters-for-spirometry-pack-of-50). The two participants were asked to repeat this test 12 times each (with a minimum of a 30 second rest between each blow), however, the first 3 trials were used as mock tests, so that the user could become accustomed to the device. The collected data were then plotted, and the spirometry values that were extracted from the device were compared with normative values for that person’s characteristics. The data was extracted by first establishing the peak flow (The maximum data point) and this became a reference point for the other variables. The Duration of each exhalation was calculated by finding the time stamp of the peak and iteratively working back through data points until the flow rate was a 4% of the maximum on either side of the peak. This percentage was determined a comparable power to that of the noise of the sensor and is negligible for the total volume. The FVC was found by measuring the total area within the duration and the FEV1 by measuring the total area under the curve within the first 1 second of the duration. For comparison, after collection with the developed device, each participant was also asked to perform 3 forced exhalations through a CE-marked spirometer (Winterthur Medical AG, model no. MSA100) which collected participant PEF and FEV1, to add further insights and allow comparisons. Results Hardware Tube Design Based on the boundary condition factors, such as the average PEF of humans (deemed to be 600 L/min (33)), the inlet diameter of the micro bacterial filter (mouthpiece available at https://www.numed.co.uk/products/disposable-bacterial-viral-filters-for-spirometry-pack-of-50) (28mm), and the maximum differential pressure range of the sensor (10 kPa), an outlet diameter of 10 mm was deemed suitable for most accurately modelling and collecting data between the two differential pressure points. This diameter would allow for the generation of a maximum of 8 kPa. These diameters were taken in conjunction with Venturi tube specifications from the British Standards Institute, and a model designed on Autodesk Fusion was made (Figures 2 and 3). In particular, there is a divergent outlet that smoothens flow dissipation, and a smooth transition region between ports to eliminate unpredictable flow. The two diameters at the inlet allow for the mouthpiece to fit comfortably within the device. Casing The casing was designed to hold electronic components secure relative to the spirometry device so that by moving the tube it would not loosen or weaken electrical component connections. The entire venturi tube and case were designed to be assembled in five components. The five components include the Venturi tube, the bottom casing, which allows electronics to be fitted and secured; the top casing to enclose the electronics from the elements; and finally two 3D printed ports which interface between the venturi tube and the differential pressure sensor. The full assembly is shown in Figure 4. The device in Figure 4 was designed to reduce the amount of printing material and support material for manufacture, so the device was manufactured with five assimilable components rather than one bulk component. This has three benefits. First, within LRS it is important to minimise the precious materials being wasted from support structures, which could be used to make other components. Second, in LRS reliable power supplies are not always guaranteed, so by splitting the device into multiple printable pieces it minimises the risk of total print failure in the event of power loss. Finally, orienting the prints individually to reduce material waste ensures a better print quality and reduces rough surfaces left over from the removed support material, which could harbour more bacteria when in use or influence air flow. The device is assembled by sliding the Venturi tube into the top casing using guide rails to lock the tube orientation in place; then it slides until it reaches inhibitor blocks in the rails. The Venturi ports that interface between the tube and the sensor are then inserted into the device to lock the Venturi tube from sliding. The electronic components can then be placed, and the sensor can be connected to the tube via plastic stubbing with an internal diameter of 4mm. For the device used in the testing and validation shown in Figure 4, the casing exhibits a hole at the bottom of the device to allow a USB cable to power and transfer data. In the future, this hole will be removed since a battery and Bluetooth will be used. This device was 3D printed using an Ultimaker 2+ 3D printer with a 2.85 mm polylactic acid (PLA) filament at a 0.1 mm layer height, a wall thickness of 1.2 mm and an infill density of 40%. The print did not require any support material as it was printed with the truncation of the tube at the bottom of the print. This meant that the layer orientation was aligned perpendicularly with the direction of air flow. This resulted in a wall roughness of 10 micrometres. Software The Arduino and MATLAB software were designed to interface together so that the MATLAB code could control when the Arduino would collect data from the sensor and when to stop collecting data from the sensor. This was implemented using the serial port so that, when the board detected a new connection, it would start collecting data. Once enough data were collected (10,000 data points lasting over 10 seconds), this connection was terminated. As mentioned previously, after setting up initial variables for detection from the sensor, the Arduino enters a wait mode where it waits for a serial connection to be established by the computer. Until then, no data is collected. Once a serial connection is established, the code starts collecting data via the analogue read function from the sensor pin every 1 ms. Within this loop, these datapoints are sent to MATLAB via serial printing. This process is iterated for approximately 10 seconds, after which Arduino receives a termination command from the serial connection executed in MATLAB. This causes the device to enter a wait mode again until it is required for the next data collection. The code operated as expected, as described within the methodology section, following the execution path shown in Figure 1. It successfully interfaced with the Arduino by establishing and disestablishing serial connections with the board to create 10000-bit arrays of data, computing these into accurate volumetric flows, which could be graphically outputted to find key spirometry values. Testing and Validation Two-Litre Hand Pump Test Table 1. Results from the two-litre hand pump validation test Trial Total volume (L) PEF (L/min) Duration (S) Signed percentage error (%) 1 1.94 475.78 0.43 -3.0 2 1.99 400.14 0.52 -0.5 3 1.98 391.37 0.54 -1.0 4 1.96 432.78 0.50 -2.0 5 1.97 441.93 0.49 -1.5 6 1.92 280.93 0.79 -4.0 7 1.99 320.96 0.60 -0.5 8 1.99 381.41 0.60 -0.5 9 1.99 329.8 0.66 -0.5 10 1.95 249.82 0.72 -2.5 11 2.03 387.24 0.90 1.5 12 2.00 389.94 0.56 0.0 13 2.03 471.43 0.57 1.5 14 2.09 109.82 1.51 4.5 15 2.00 618.00 0.40 0.0 16 2.00 599.00 0.39 0.0 17 1.96 348.98 0.58 -2.0 18 2.00 421.34 0.68 0.0 19 1.91 180.04 1.02 -4.5 20 1.97 396.26 0.53 -1.5 21 1.98 600.35 0.41 -1.0 22 1.97 544.07 0.45 -1.5 23 1.97 281.02 0.83 -1.5 24 2.02 112.33 1.48 1.0 25 2.03 72.97 2.44 1.5 26 1.95 602.91 0.40 -2.5 27 1.97 496.50 0.39 -1.5 28 1.98 576.73 0.49 -1.0 29 1.97 627.03 0.40 -1.5 30 1.97 678.35 0.33 -1.5 L: Litres L/min: Litres per minute S: Seconds %: percentage Table showing the measured forced vital capacity (FVC in L), peak expiratory flow (PEF L/min) and duration of each expiration in (s) after each expiration. The signed percentage error compares the measured volume with the expected volume of 2 litres and is the signed percentage difference between these two values. In the handpump test, the mean value for the measured volume was 1.983 L, whereas the real value was 2 L, and the mean absolute percentage error was 1.53%, with a mean percentage error of –0.87%. Table 1 contains all the data related to this. Figure 5 plots the signed percentage error versus the duration. The correlation analysis revealed a Spearman’s rho of 0.2717 and a p-value of 0.1464, indicating that there is neither a high nor statistically significant correlation between the input flow and the accuracy of the device. Early usability study on the effectiveness of the spirometry device as a lung function test Table 2. Female participant spirometry parameters from the forced expiratory study Duration (s) PEF (L/min) FVC (L) FEV1 (L) FEV1/FVC ratio 1 0.94 357.25 2.62 2.62 2.62 2 1.70 367.14 3.20 3.02 0.94 3 1.68 335.00 3.34 2.98 0.89 4 1.69 328.05 3.37 3.03 0.90 5 1.05 377.70 3.45 3.44 1.00 6 1.94 322.04 3.65 3.13 0.86 7 1.68 402.07 3.62 3.29 0.91 8 1.55 408.24 3.56 3.19 0.90 9 1.37 445.97 3.57 3.39 0.95 10 1.25 507.15 3.67 3.55 0.97 11 1.75 452.93 3.70 3.40 0.92 12 1.66 389.61 3.64 3.31 0.91 Mean (excluding trial 1,2,3) 1.55 403.75 3.58 3.30 0.92 The female participant performed a full forced expiratory flow test 12 times and each time the duration for which the total volume to be expelled was recorded in seconds (s). The FVC in litres (L), the peak expiratory flow (PEF) in litres per minute (L/min) and the forced expiratory volume in 1 second (FEV1) in litres (L) were also recorded (trials 1,2 and 3 were removed from the data analysis) Table 3. Male Participant Spirometry parameters from Forced Expiratory Study Trial Duration (s) PEF (L/min) FVC (L) FEV1 (L) FEV1/FVC ratio 1 1.63 524.00 3.93 3.70 0.94 2 1.75 581.51 4.16 3.78 0.91 3 1.30 590.48 4.03 3.85 0.96 4 1.47 564.22 3.83 3.64 0.95 5 1.44 580.23 4.04 3.84 0.95 6 1.48 536.71 4.03 3.93 0.98 7 1.40 571.45 3.89 3.69 0.95 8 1.96 600.38 4.13 3.76 0.91 9 1.56 415.78 3.88 3.61 0.93 10 1.36 581.52 4.22 4.08 0.97 11 1.58 585.35 4.23 3.90 0.92 12 1.58 578.07 4.35 4.12 0.95 Mean (excluding trial 1,2,3) 1.54 562.80 4.07 3.84 0.94 The male participant performed a full forced expiratory flow test 12 times and each time the duration of the total volume to be expelled was recorded in seconds (s). The FVC (FEV1) in litres (L) were also recorded (trials 1,2 and 3 were removed from data analysis) In the full forced expiratory flow test with participants the key spirometry parameters for both male and female subjects are reported in Tables 2 and 3. The first 3 trials for each subject were excluded from this data analysis, as the participants were using these trials to familiarise themselves with the device and would significantly affect the results. The female participant achieved a mean FVC (i.e., total volume) of 3.58 L, a mean PEF of 399.88 L/min and a mean FEV1 of 3.30 L. According to a global lung initiative calculator validated with 74,185 participants (34), the normative values of a woman of her height and age for the FVC should fall in the range of 3.05 L – 4.83 L for the lower limit of normality (LLN) and upper limits of normality (ULN) with a predicted value of 3.93 L. The FEV1 expected was between 2.68 L LLN and 4.20 L ULN, with a predicted of 3.46 L. Regarding the PEF, the normative value would be 477.60 L/min according to equations derived from R J Knudson (1993) et al (35). The male participant achieved a mean FVC of 4.06 L, a mean FEV1 of 3.84 L and a mean PEF of 554.20 L/min. The normative value of a man of his height and age for the FVC should fall in the range of 4.027 L – 6.272 L for LLN and ULN with a predicted value of 5.141 L (34). The expected FEV1 was between 3.429 LLN and 5.282 ULN, with a predicted of 4.478 L. As regards the PEF, the normative value would be 623.4 L/min (35). The full expiratory volume expelled over time for the 12 trials are shown in Figures 6 and 7 for the female and male participant respectively. The results of the CE marked device are hereby reported in Table 4 Table 4. Results from the CE marked spirometry device Female participant Male participant Trial no PEF (L/min) FEV1 (L) PEF (L/ min) FEV1 (L) 1 530.00 3.77 572.00 3.90 2 529.00 3.77 599.00 3.86 3 514.00 3.71 588.00 3.83 Mean 524.33 3.75 586.33 3.86 Table of results from the CE marked spirometry device showing the results from the full forced expiratory flow test. Three trials were used for each participant and after each expiration the peak expiratory flow (PEF) in litres per minute was recorded, as was the forced expiratory volume in 1 second (FEV1) in Litres. The mean values were then calculated and displayed. Discussion Based on the results of both the two-litre pump test, which assessed the device's accuracy, and the usability study, which evaluated its validity as a lung screening tool, the device’s findings were further compared with standard spirometry testing procedures to determine whether it met the criteria for clinical use. Diagnostic tools for spirometry are typically calibrated using a three-litre syringe as per the ISO 26782:2009 guidelines, according to which a mean percentage error difference of 3% is allowed for the measured volume compared with the actual volume. The reason for this is that the threshold set is intended guarantee reliability when a clinician uses the FEV1/FVC ratio to distinguish between respiratory disease types (obstructive/ restrictive). An error less than 3% is considered negligible and does not skew lung parameters such as the FEV1 enough to affect clinical decision making. In our case, although we used a two-litre pump, we still applied the 3% threshold relative to the measured volume. This results in a more stringent criterion due to the smaller calibration volume. Our measured mean absolute percentage error of 1.53% remains well within these limits. Deviations within the reported percentage error could be attributed to a number of factors such as user errors of the pump, filtering errors that lead to over or underfitting of noisy data, limitations of the low-cost sensor, potential air leakages within the device or even specific individuals’ coefficient of variation for each expiration. Other diagnostic tools commonly use this error band for similar reasons and is vital for standardising equipment reasonably within the available technology. The error of this device is, therefore, within the boundaries that can be attributed to random noise error equivocal to that of a standard spirometer. The performance of this device is comparable to that of other low-cost devices reported in the literature, such as those stated within the introduction. These prior studies that used methods such as differential pressure techniques or other approaches achieved accuracies similar to those of this project. Two Federal Drug Administration (FDA)-approved commercial devices: the AirSmart Spirometer (36) and the MIR Spirobank Smart (37), both reported accuracies of ±3% for volume and ±5% for flow, and retail at approximately €69 (36) and $190 (38), respectively. Additionally, a custom spirometry device developed by Gupta et al. (39) reported a standard deviation of 8% and a maximum deviation error of 2.5% compared with a standard spirometer. Since the developed device obtained accuracies similar to those for FVC and FEV1 of commercial standard devices, it reinforces the validity of the device as a spirometry instrument. The production cost of the device is estimated to be approximately £45, covering the Arduino board, differential pressure sensor, 3D printing, and additional electronic components. Large-scale manufacturing could further reduce costs through bulk purchasing. Compared with the Venturi-based spirometry device by Ferreira Nunes et al (27), which was most similar to the device reported in this paper, our device, as per this paper narrative, is the result of a frugal, user-driven and contextualised design process, which is paramount for medical devices to be deployed and used effectively in LRSs and LMICs. In terms of performance, their device achieved a mean absolute percentage error of 1.94% (compare to ours of 1.53%) and a signed percentage error of 0.08% when a 3 L cylinder was measured. This was achieved using a Venturi-based system (diameter 32 mm constricting to 10 mm), an ESP32 microcontroller and a Honeywell SSCMRRN005PDAA3 differential pressure sensor. Although the full prototype price was not reported, the retail price of this sensor is £74 (40), and with this high price, the sensor has a large operating range of -34 to 34 kPa and a %Vfss of 0.25%. Compared with MXP5010DP, which has an operating range of 0-10kPa and a %Vss of 5%, the Honeywell sensor is technically better. However, within the context of LRS, the NXP sensor is more suitable. Its lower cost of £20 (41) makes the device more affordable and accessible, with only a minimal sacrifice in performance. The MXP5010DP sensor also has several technical advantages in these settings. The NXP sensor is an analogue component that outputs voltage relative to pressure, so it is compatible with any microcontroller, the Honeywell sensor is digital and thus requires a much more precise configuration, requiring a digital communication setup, specialised software and library requirements, and can be damaged easily by supplying the wrong voltage to the sensor. This makes debugging in the LRS more difficult. The Honeywell sensor is factory calibrated in comparison to the NXP device, which requires user calibration to take into account possible voltage offset; however, by allowing user calibration each device can automatically calibrate with the available data making the device more suitable for specific settings, and when the NXP is paired with a suitable low pass filter it can reduce noise to a comparable level to that of the digital sensor. Overall, although the NXP sensor is less technically precise, it still has an error within ±0.5 kPa, which is clinically acceptable for spirometry whilst maintaining several advantages in LRS over the SSCMRRN005PDAA3 (41). This device differs from the spirometers noted above in its specific design and suitability for LRS. In comparison, it is a fraction of the price at just £45 for base components, and the components used are all affordable and available within LRS. Furthermore, 3D printing can be implemented, and recyclable materials such as PLA can be used. With access to a 3D printer becoming more commonly available, devices can be created locally. This will reduce reliance on imports for components and allow communities to maintain and repair medical devices without external interventions. This was supported by the open-source nature of the project and the design process, which aims to create an easily assimilable device and minimised postproduction processes that lead to increased waste. The device does not have high power requirements and can be connected to either a laptop or smartphone to power the device and interpret data. Future iterations will involve the use rechargeable batteries. Furthermore, the software required does not need continuous internet connectivity to work, supporting use within more rural locations and mobile deployments. As an alternative frugal spirometry device for implementation in LRS, our device comprises of fewer and more affordable components than its available counterparts do, it incorporates user feedback systems that instruct the user how to use it, and it returns the values of FEV1, FVC and PEF back to the user instantly via an LCD screen or via the MATLAB script which provides a visual representation of the data. It uses entirely 3D printed components or off-the-shelf bought electronics, so it can be easily replicated in LRS. Since the design is simplistic and primarily focusing on the basic needs of low-income and remote areas, while maintaining clinical accuracy and addressing financial barriers, it can, therefore, be considered a good solution to enhance respiratory care in LRS. While this work proved to have value as an effective spirometry device, it also has limitations. The device was designed to operate in LRS with minimal resources, and as such, the selected components were chosen to reflect this. Therefore, the individual limits of the components tended to be greater because of constraints such as cost. Most notably, the differential pressure sensor (MXP5010P) chosen had the most limitations, which could be attributes to its error. The sensor could measure pressures in the range of 0 to10 kPa (i.e., 75 mmHg). And the datasheet of the sensor states that its accuracy is ±5% on the full-scale voltage, i.e., 4.5 V, which means ±0.225 V. This means that, when small values in the lower portion of the sensor range are read, the measurements will be affected more by inaccuracies. For example, with an input of 1 mmHg, the sensor reading without offset is 0.14 volts, which is a comparable number to the stated accuracy of the device. Another possible factor is that at very small and very large readings it is possible that the device has a non-linearity error, such that where the device normally reads linearly, at the extremes, this deviation can lead to different results (42). Finally, the sensor can experience a factor called offset drift. This is when the nominal offset voltage, which is calibrated at the start of each blow, can change over time, affecting the accuracy. All these factors could be attributed to the observations shown in Figure 5 where it can be seen that at longer durations (lower flow rates) there tends to be a greater percentage error. However, there was non-significant correlation between the two variables. These differences may be due to the limits of the sensor itself and to its accuracy, as mentioned above. With respect to the usability study, it was observed that several factors could affect air flow within the device and, therefore, the spirometry results. Each participant used a microbacterial filter to prevent the spread of bacteria and viruses within the device, as well as prevent the inhalation of particles from the 3D printed parts. Each of these factors has an impedance resistance to the flow that may affect the flow speed. This effect should be minimal, however, and since impedance does not occur across the measured zone. A more prevalent issue was the fit between the joints. Although the device was designed to have a friction-fitted seal, it may not inherently be airtight, especially if, when the device is used, it moves as the person exhales. This could lead to some air not being directed down the spirometry tube. It should be noted that this effect, however, should also be easily detectable by the person during each blow, as well as identifiable on the data as there will be a significant drop-in the flow rate. Technique significantly affected the readings from the participants; this is seen by the increase in the total volume in the female participant after each trial. It should be noted this was this participant first time using the device. Within the first 3 trials the volume increased from 2.61 L to 3.0 5L to 3.34 L and finally peaked at 3.73L at trial 11. Therefore, the greatest blows should only be considered when comparing forced expiration blows. Finally, since the design uses 3D printed parts, these parts influence fluid flow within the device. The feature that was considered was the effect of wall drag due to 3D printed layer structures; this effect was mitigated by calculating this pressure drop within the code estimating wall roughness from printer settings. The wall thickness of the structures was increased to 1.8bmm thick to limit the possibility of air permeating through the structure, at lower thicknesses more air may escape from flaws within the structure. Resulting in a smaller pressure drop. Given the greater implications of these devices, this study shows promise in developing a real-world solution to better manage respiratory disease in LRS. By reducing costs, focusing on more open-source designs and orienting the design around the contextualised needs, a more effective device can be produced. This could increase the rates of early diagnosis as well as better monitor disease progression locally. Future work to consider includes investigating the robustness of the casing for electronics, shifting towards the use of rechargeable batteries, and Bluetooth monitoring. Finally, more data should be collected from a greater number of participants to validate the device further. Within this data collection, there would be a secondary aim to balance the data collection between healthy participants and people with respiratory diseases. This could then be used in future studies to train a machine learning model to help identify different respiratory diseases. By doing so, the device can shift to becoming more of clinical decision-making tool, which would help less skilled user determine the appropriate course of action when dealing with their lung health. The device should further focus on integrating more generalised policies surrounding spirometry devices, focusing on how it can implement more international standards and FDA legislation so it can be more widely adopted. Scaling these frugal spirometry devices for wider adoption further aligns with the World Health Organisation (WHOs) sustainable development goals, goal 3: Good health and wellbeing to combat respiratory diseases and sustainable development goals 7 and 9 which support frugal engineering as a form of sustainable development. Conclusion This study designed, developed and validated a novel and frugal 3D-printed spirometry device aimed at addressing the contextual needs and challenges of LRS that are often neglected in the design of medical devices. Our study explored unique data processing methods to model fluid flow within a 3D-printed constricted spirometry tube. The device was validated through two sets of tests. First, its accuracy in volume measurement was assessed by pumping a known volume of 2 litres through the device and recording the output. The device demonstrated a mean absolute percentage error of 1.53% compared with the known volume, indicating a high level of measurement accuracy. Second, the device's usability for assessing lung function was evaluated. The male participant achieved a mean FVC of 4.03 L and a mean PEF of 557 L/min, while the female participant achieved a mean FVC of 3.44 L and a mean PEF of 395 L/min. When predicted normal values based on the participants' physiological characteristics were compared with a validated Global Lung Initiative calculator (35), the measurements fell within expected ranges. Comparison with the CE-marked spirometer further supported the validity of the device. The CE-marked device recorded mean PEFs of 586.33 L/min and 524.33 L/min and FEV1 values of 3.86 L and 3.75 L for the male and female participants, respectively. These findings suggest that the developed device provides measurements that are comparable to those of a certified spirometry system, supporting its potential for use in lung function monitoring. This study demonstrates the implications that frugal engineering could have, when applied to medical devices. It specifically addresses the challenging environments that often hinder resource constrained settings and enables designs to adapt to a diverse range of contexts so that they are effective and sustainable. This study further underpins that by considering only the required contextual needs, devices can be developed at fraction of the cost without influencing clinical accuracy. Overall, by developing and adopting more frugal devices, the gap between HIC and LMICs can be reduced. This provides better access to health care for LMICs’ and places less reliance upon HICs for donations or aid. Furthermore, the developments for LMICs can also benefit HICs as they address problems uniquely that might not have been identified. This could lead to the adoption of context designed devices into HICs, especially within more impoverished areas within HICs. This reverse innovation could provide affordable solutions to address consistent health problems that have been affecting underrepresented populations within HICs, leading to better worldwide health at all levels of health care. Abbreviations High Income Countries (HICs) Low Middle Income countries (LMICs) Chronic Obstructive Pulmonary Disease (COPD) Forced Expiratory Volume in one second (FEV1) Forced Vital Capacity (FVC) Polylactic acid (PLA) lower limit of normality (LLN) upper limits of normality (ULN) low resource settings (LRSs) Federal Drug Administration (FDA) Liquid Crystal Display (LCD) Light Emitting Diodes (LEDs) World Health Organisation (WHOs) Declarations Ethics Approval and Consent to Participate This study was approved by the University of Warwick Biomedical and Scientific Research Ethics Committee (BSREC) Application Reference: BSREC 145/23-24. Before the study started, investigators introduced the purpose of the survey to the participants who volunteered to participate in the study and obtained their informed consent. Consent for Publication Not applicable. Availability of Data and Material The data is available upon a reasonable request. Please put in this request to the corresponding author James Wallace ( [email protected] ). Competing Interests The authors declare no competing interests. Funding James Wallace was funded by the EPSRC (EP/W524645/1). Philip Anyanwu and Charles F Hayfron-Benjamin contributions to this paper are part of their work within the African Research Universities Alliance - Multimorbidity Cluster of Research Excellence (ARUA-AMCORE), a collaborative initiative led by ARUA and the Guild. Author Contributions James Wallace: Conceptualisation, Methodology, Software, Validation, Formal Analysis, Investigation, Data Curation, Writing-Original Draft, Writing- Review and Editing, Visualisation. Davide Piaggio: Conceptualisation, Methodology, Validation, Investigation, Resources, Writing – Review and Editing, Supervision, Project Administration, Funding Acquisition. Philip Anyanwu: Writing – Review and Editing, Supervision, Project Management, Funding Acquisition . Pedro Rifa-Checa: Conceptualisation, Methodology, Software, Investigation, Writing – Original Draft. Thanusan Kannathasan: Software, Writing – Original Draft. Charles F Hayfron-Benjamin: Conceptualisation, Methodology, Writing – Review and Editing. Acknowledgements The authors would like to thank Prof Petr Denissenko at the University of Warwick for providing useful insight into the usability of project and guiding fluid dynamic considerations. The authors would also like to thank the School of Engineering Build space for allowing the use of their 3D printers. Finally, the authors thank Alvin Vyapooree (Medical graduate from Warwick Medical School) for his inputs to the medical aspects of the project. References Organisation WH. WHO global model regulatory framework for medical devices including in vitro diagnostic medical devices. Geneva; 2017. Arasaratnam A, Humphreys G. Emerging economies drive frugal innovation. World Health Organization Bulletin of the World Health Organization. 2013;91(1):6. Bank W. Current Health Expenditure Per Capita, PPP (Current International $) 2025 [Available from: https://data.worldbank.org/indicator/SH.XPD.CHEX.PP.CD?most_recent_value_desc=false. Organisation WH. Medical Devices: Managing the Mismatch: An Outcome of the Priority Medical Devices Project. Geneva; 2010. Ongaro AE, Ndlovu Z, Sollier E, Otieno C, Ondoa P, Street A, et al. Engineering a sustainable future for point-of-care diagnostics and single-use microfluidic devices. Lab on a Chip. 2022;22(17):3122-37. Organisation WH. Medical device donations: considerations for solicitation and provision. second edition ed. Geneva2024. Ibrahim NH, Wallace J, Piaggio D, Pecchia L. Validation of a framework for assessing healthcare facilities in low-resource settings: a field study in selected facilities in Ethiopia. Health and Technology. 2025:1-9. Leong D, Teo K, Rangarajan S, Lopez-Jaramillo P, Avezum Jr A, Orlandini A. World Population Prospects 2019. Department of Economic and Social Affairs Population Dynamics. New York (NY): United Nations; 2019 (https://population. un. org/wpp/Download/, accessed 20 September 2020). The decade of healthy ageing. Geneva: World Health Organization. World. 2018;73(7):362. Tran VT, Ravaud P. Frugal innovation in medicine for low resource settings. BMC Med. 2016;14(1):102. Montesinos L, Checa Rifá P, Rifá Fabregat M, Maldonado-Romo J, Capacci S, Maccaro A, et al. Sustainability across the medical device lifecycle: a scoping review. Sustainability. 2024;16(4):1433. Piaggio D, Castaldo R, Cinelli M, Cinelli S, Maccaro A, Pecchia L. A framework for designing medical devices resilient to low-resource settings. Globalization and health. 2021;17(1):64. Piaggio D, Namm G, Melillo P, Simonelli F, Iadanza E, Pecchia L. Pupillometry via smartphone for low-resource settings. Biocybernetics and Biomedical Engineering. 2021;41(3):891-902. Williams E, Piaggio D, Andellini M, Pecchia L. 3D-printed activated charcoal inlet filters for oxygen concentrators: a circular economy approach. Development Engineering. 2022;7:100094. Piaggio D, Hyland S, Maccaro A, Iadanza E, Pecchia L. A 3D-printed condom intrauterine balloon tamponade: Design, prototyping, and technical validation. Plos one. 2024;19(6):e0303844. Piaggio D, Castaldo R, Garibizzo G, Iadanza E, Pecchia L. A smartphone-based tool for screening diabetic neuropathies: A mHealth and 3D printing approach. Biomedical Signal Processing and Control. 2024;89:105807. Organisation WH. The Top 10 Causes of Death: World Health Organisation; 2024 [Available from: https://www.who.int/news-room/fact-sheets/detail/the-top-10-causes-of-death. Vos T, Lim SS, Abbafati C, Abbas KM, Abbasi M, Abbasifard M, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. The lancet. 2020;396(10258):1204-22. Organisation WH. Chronic Obstructive Pulmonary Disease: World Health Organsiation; 2023 [Available from: https://www.who.int/news-room/fact-sheets/detail/chronic-obstructive-pulmonary-disease-(copd). Meghji J, Mortimer K, Agusti A, Allwood BW, Asher I, Bateman ED, et al. Improving lung health in low-income and middle-income countries: from challenges to solutions. Lancet. 2021;397(10277):928-40. Yadav H, Shah D, Sayed S, Horton S, Schroeder LF. Availability of essential diagnostics in ten low-income and middle-income countries: results from national health facility surveys. Lancet Glob Health. 2021;9(11):e1553-e60. THOMAS PS, NG C, BENNETT M. Peak expiratory flow at increased barometric pressure: comparison of peak flow meters and volumetric spirometer. Clinical Science. 2000;98(1):121-4. Moore V. Spirometry: step by step. Breathe. 2012;8(3):232-40. Vanjare N CS, Madas S, Kodgule R, Gogtay J, Salvi S. . Use of spirometry among chest physicians and primary care physicians in India. 2016. Zhou J, Li X, Wang X, Yu N, Wang W. Accuracy of portable spirometers in the diagnosis of chronic obstructive pulmonary disease A meta-analysis. NPJ Primary Care Respiratory Medicine. 2022;32(1):15. Carpenter DM, Jurdi R, Roberts CA, Hernandez M, Horne R, Chan A. A review of portable electronic spirometers: implications for asthma self-management. Current allergy and asthma reports. 2018;18:1-10. Wu Z, Huang R, Zhong L, Gao Y, Zheng J. Technical performance analysis of different types of spirometers. BMC Pulmonary Medicine. 2022;22:1-7. Ferreira Nunes M, Plácido da Silva H, Raposo L, Rodrigues F. Design and Evaluation of a Novel Venturi-Based Spirometer for Home Respiratory Monitoring. Sensors. 2024;24(17):5622. Igwe Johnson O. Development and Test Performance of a Benchtop Venturi Tube. NIPES-Journal of Science and Technology Research. 2023;5(2). Caretto L. Bernoulli Equation: California State University, Northridge; 2008 [Available from: https://www.csun.edu/~lcaretto/me390/03-Bernoulli.pdf. Boyes W. Instrumentation reference book: Butterworth-Heinemann; 2009. Shapiro SS, Wilk MB. An Analysis of Variance Test for Normality (Complete Samples). Biometrika. 1965;52(3/4):591-611. Schober P, Boer C, Schwarte LA. Correlation coefficients: appropriate use and interpretation. Anesthesia & analgesia. 2018;126(5):1763-8. Clinic C. Peak flow meter: how to use, measurement & results Cleveland (OH): Cleveland Clinic; 2024 [Available from: https://my.clevelandclinic.org/health/treatments/peak-flow-meter. Bowerman C, Bhakta NR, Brazzale D, Cooper BR, Cooper J, Gochicoa-Rangel L, et al. A race-neutral approach to the interpretation of lung function measurements. American journal of respiratory and critical care medicine. 2023;207(6):768-74. Knudson RJ, Lebowitz MD, Holberg CJ, Burrows B. Changes in the normal maximal expiratory flow-volume curve with growth and aging. American Review of Respiratory Disease. 1983;127(6):725-34. Ramos Hernandez C, Nunez Fernandez M, Pallares Sanmartin A, Mouronte Roibas C, Cerdeira Dominguez L, Botana Rial MI, et al. Validation of the portable Air-Smart Spirometer. PLoS One. 2018;13(2):e0192789. Ramsey RR, Plevinsky JM, Milgrim L, Hommel KA, McDowell KM, Shepard J, et al. Feasibility and preliminary validity of mobile spirometry in pediatric asthma. J Allergy Clin Immunol Pract. 2021;9(10):3821-3. Research MMI. Spirobank Smart – Personal Bluetooth Spirometer 2024 [Available from: https://shopusa.spirometry.com/product/spirobank-smart/. Gupta S, Chang P, Anyigbo N, Sabharwal A. mobileSpiro: accurate mobile spirometry for self-management of asthma. Proceedings of the First ACM Workshop on Mobile Systems, Applications, and Services for Healthcare; Seattle, Washington: Association for Computing Machinery; 2011. p. Article 1. Components R. Honeywell Piezoresistive Pressure Sensor, 34kPa Operating Max, Surface Mount, 8-Pin, 206kPa Overload Max, SMT 2024 [Available from: Honeywell Piezoresistive Pressure Sensor, 34kPa Operating Max, Surface Mount, 8-Pin, 206kPa Overload Max, SMT. Components R. NXP Low Pressure Sensor MPX5010DP. 2025. Lukat R. The 3 Main Errors That Affect Pressure Sensor Accuracy 2025 [Available from: https://blog.wika.com/us/knowhow/pressure-sensor-accuracy-3-errors/. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6812011","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":467167007,"identity":"a3c15ea2-4184-41f4-8712-a1407f7ac85a","order_by":0,"name":"James Wallace","email":"data:image/png;base64,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","orcid":"","institution":"University of Warwick","correspondingAuthor":true,"prefix":"","firstName":"James","middleName":"","lastName":"Wallace","suffix":""},{"id":467167008,"identity":"ce1d5b3b-6bc1-4a77-a50c-7cb90aa09492","order_by":1,"name":"Pedro Checa Rifá","email":"","orcid":"","institution":"University of Warwick","correspondingAuthor":false,"prefix":"","firstName":"Pedro","middleName":"Checa","lastName":"Rifá","suffix":""},{"id":467167009,"identity":"4739e1bf-e08d-4696-9a0a-816d935c9d71","order_by":2,"name":"Thanusan Kannathasan","email":"","orcid":"","institution":"University of Warwick","correspondingAuthor":false,"prefix":"","firstName":"Thanusan","middleName":"","lastName":"Kannathasan","suffix":""},{"id":467167010,"identity":"842507d7-f4e0-4b8c-b7fb-b0c01023f3ea","order_by":3,"name":"Charles F Hayfron-Benjamin","email":"","orcid":"","institution":"University of Ghana Medical School","correspondingAuthor":false,"prefix":"","firstName":"Charles","middleName":"F","lastName":"Hayfron-Benjamin","suffix":""},{"id":467167011,"identity":"41c5dde4-29c2-48c8-be10-760498d761df","order_by":4,"name":"Philip Anyanwu","email":"","orcid":"","institution":"University of Warwick","correspondingAuthor":false,"prefix":"","firstName":"Philip","middleName":"","lastName":"Anyanwu","suffix":""},{"id":467167012,"identity":"51bc4fe1-374d-4f2f-b291-ce1d6957e991","order_by":5,"name":"Davide Piaggio","email":"","orcid":"","institution":"University of Warwick","correspondingAuthor":false,"prefix":"","firstName":"Davide","middleName":"","lastName":"Piaggio","suffix":""}],"badges":[],"createdAt":"2025-06-03 13:53:24","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6812011/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6812011/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":84375656,"identity":"71a147b2-5e7d-461b-be5c-1da82c5aa74f","added_by":"auto","created_at":"2025-06-11 08:19:56","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":556667,"visible":true,"origin":"","legend":"\u003cp\u003eFlowchart of steps for data processing within MATLAB.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6812011/v1/ea7ff273db6bb20446643ff7.png"},{"id":84375646,"identity":"9c150cf6-798f-409c-9047-0b77ee6a36eb","added_by":"auto","created_at":"2025-06-11 08:19:55","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":149987,"visible":true,"origin":"","legend":"\u003cp\u003eDrawing of the proposed Venturi tube used to create differential pressure units in mm.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6812011/v1/ec3eda25f6d286139e14c340.png"},{"id":84375660,"identity":"9914974c-f9f7-4d1d-9e85-2994f2299220","added_by":"auto","created_at":"2025-06-11 08:19:56","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1220756,"visible":true,"origin":"","legend":"\u003cp\u003e(top), 3 (Bottom). 3D models of the spirometry tube.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6812011/v1/bf008d455745f598707b6cb1.png"},{"id":84375655,"identity":"3d34122c-7867-45ff-b79a-ee680b950284","added_by":"auto","created_at":"2025-06-11 08:19:56","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":401535,"visible":true,"origin":"","legend":"\u003cp\u003e(top) Technical drawing of the casing including the venturi tube, (bottom) Full assembly of the 3D printed casing. The sizes measures are reported in mm.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6812011/v1/a1c6e0a93810551ee56e3ab0.png"},{"id":84375648,"identity":"5537089c-124f-480b-8a02-09635b5e1fdd","added_by":"auto","created_at":"2025-06-11 08:19:55","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":62729,"visible":true,"origin":"","legend":"\u003cp\u003eGraph showing the signed percentage error (SPE) vs the duration of the 2L measured volume.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-6812011/v1/0cfa3e0a23fe8377e0934a7f.png"},{"id":84375657,"identity":"3607aa2e-f2b3-4195-9d79-421a0b02e678","added_by":"auto","created_at":"2025-06-11 08:19:56","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":282295,"visible":true,"origin":"","legend":"\u003cp\u003eVolume against time graph showing 12 trials for the forced expiratory flow test for female participants\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-6812011/v1/19fbb9c7f951b0f9c4b6b19c.png"},{"id":84375659,"identity":"381b992c-50ac-4eb0-97cd-47ea4da3729b","added_by":"auto","created_at":"2025-06-11 08:19:56","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":252541,"visible":true,"origin":"","legend":"\u003cp\u003eVolume against time graph showing 12 trials for the forced expiratory flow test for male participant\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-6812011/v1/dc8a9c00499367a617298a96.png"},{"id":84376629,"identity":"7a6cc751-f0ef-4385-85e9-6ccd3f0551b1","added_by":"auto","created_at":"2025-06-11 08:27:58","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4459509,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6812011/v1/0b30b92c-15e9-4b08-8afb-9a774c3d5460.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eDesign, Manufacture and Validation of a Spirometry Device aimed for Low-Resource Settings\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMedical devices and health technologies are crucial in building resilient and effective health systems. They can enhance diagnosis and treatment, improve patient safety and care quality, and boost efficiency and productivity. Modern medical devices increasingly generate data and integrate with consumer technologies, expanding their capabilities (1). These technologies are used across a wide range of functions\u0026mdash;from diagnosing and monitoring conditions to supporting individuals with disabilities and treating both acute and chronic illnesses.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCurrently, the global market includes an estimated 2 million different types of medical devices, which are grouped into over 7,000 generic categories (1). However, inequities persist between and within regions in the application of medical devices and technologies in healthcare. For instance, high-income countries (HICs), home to only 13% of the world\u0026rsquo;s population, account for 76% of global medical device use (2). Regional disparities in healthcare expenditure\u0026mdash;US$641.30 per capita in low- and middle-income countries (LMICs) compared with US$6,943.08 per capita in HICs\u0026mdash;as of 2022 (3) significantly affect the affordability, accessibility, and effective use of medical devices and technologies. In LMICs, approximately 80% of medical devices are donated by international governmental and nongovernmental organisations (4). However, only 10\u0026ndash;30% of these donated devices become fully operational, largely because they often fail to meet the specific contextual needs of recipient countries\u0026mdash;whether due to inappropriate technical specifications, lack of supporting infrastructure, or insufficient staff training (5, 6). This heavy reliance on external donations can also compromise long-term health system sustainability by creating supply chain vulnerabilities, hindering the development of local capacity for innovation and maintenance, and reducing autonomy in healthcare planning and delivery (7). Additionally, the anticipated global population growth, particularly in LMICs (8), underscores the need for proportional universalism in the development and distribution of medical devices, ensuring increased availability and contextually tailored designs in regions with the highest need.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe application of frugality in the design of medical devices for managing conditions, such as respiratory ones, is crucial for improving access to life-saving treatments, strengthening health systems and reducing the environmental impact, especially in resource-limited settings where the burden of such conditions is greater (9, 10). This approach focuses on creating cost-effective and reliable solutions that address the specific needs and constraints of resource-limited settings (11), as proven in these specific examples (12)(13)(14)(15).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRespiratory conditions remain a crucial global health issue, with lower respiratory infections ranking 5\u003csup\u003eth\u003c/sup\u003e among the leading causes of death in 2021 (16). The burdens of these conditions are greater in LMICs (17). For example, lower respiratory tract infections were the number one cause of death in LICs in 2021 (16). A staggering 90% of chronic obstructive pulmonary disease (COPD) deaths among those aged under 70 years occurred within LMICs in 2021(18). Similarly, a staggering 96% of global asthma deaths also occurred in LMICs in 2019 (19). Increasing access to low-cost, context-specific medical devices in LMIC settings could help narrow regional disparities by reducing underdiagnosis/misdiagnosis and enhancing early detection and more effective disease management (20).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCurrently, the most commonly used devices for diagnosing respiratory diseases include peak flow meters, spirometers and pulse oximeters (21). Among these, spirometers are the most versatile, offering a more comprehensive assessment of lung function by generating full volume time curves, rather than isolated values such as peak flow. They primarily measure key indicators such as forced expiratory volume in one second (FEV1), forced vital capacity (FVC) and peak expiratory flow (PEF), providing the most reliable data for differentiating between conditions such as COPD and asthma (22). Despite their diagnostic value, spirometers remain underutilised in low-resource settings (LRS) because of relatively high costs and the need for trained personnel to operate and interpret the results (23). These challenges make spirometers strong candidates for frugal innovation\u0026mdash;specifically, the development of affordable, durable, and easy-to-use versions tailored to the needs of LMICs for the effective diagnosis and management of respiratory diseases.\u003c/p\u003e\n\u003cp\u003eThere remains a lack of portable, affordable spirometers suitable for clinical use in LRS. Studies highlight not only cost barriers but also issues with data interpretation and the need for more user-friendly, technologically integrated devices (24-26). A review by Carpenter et al. (25) assessed 16 portable spirometers (priced $99\u0026ndash;$1,390 US dollars), revealing limited information on data security, accuracy, and patient outcomes. Most devices failed to account for users\u0026rsquo; understanding, making results such as FEV1 and PEF difficult to interpret without reference values, and also faced usability and interface challenges.\u003c/p\u003e\n\u003cp\u003eFerreira Nunes et al. (27) developed a Venturi-based home spirometer using an ESP32 microcontroller, a high-precision Honeywell SSCMRRN005PDAA3 pressure sensor, and 3D-printed components. While the device demonstrated excellent technical performance, showing high correlation with a gold-standard spirometer (intraclass correlation coefficient of 0.987 for FVC), its suitability for LRS was limited because of the expensive pressure sensor (\u0026pound;74.17), lack of onboard data display, and limited environmental robustness. Open ports, for example, could allow dust intrusion. These issues underscore the need for frugal, user-centred designs that improve affordability, durability, and ease of use in LMIC settings.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOur study, therefore, aimed to develop an affordable, durable and effective spirometry device tailored for resource-limited settings. Adopting a frugal innovation approach, we leveraged modern technologies such as 3D-printing and affordable electronics to create a cost-efficient solution. Our design is grounded in a user-centred methodology, prioritising the needs of primary care physicians to ensure accuracy, contextual relevance, ease of use, and broad accessibility. We focus on LRS in LMICs, particularly sub-Saharan Africa, but recognise the potential for use in low-resource healthcare systems of HICs. \u0026nbsp;\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eThis section details the methods used to design, prototype, and validate the hardware and software components of the spirometer.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eHardware\u003c/h2\u003e\n\u003ch3\u003eContextualised design\u003c/h3\u003e\n\u003cp\u003eBuilding on prior research and expertise in designing medical devices for LRSs (7, 11) (and the examples mentioned in the introduction), the spirometer was developed with a user-driven approach, specifically tailored to the needs and constraints of such environments. Two focus groups were conducted to consult 10 biomedical engineering and medicine students from both HICs and LMICs, to inform the co-creation of this device. This was supported by discussions with clinicians from LMICs who use spirometry to treat patients with respiratory conditions. Key decisions from these consultations included adopting a differential pressure sensing system to eliminate the need for moving parts, such as turbines, thereby increasing the device's durability and suitability for low-resource environments. While turbine flow and rolling seal systems were considered, they were ultimately dropped due to their fragility and susceptibility to dust and humidity. Another important decision from the consultation was to minimise the number of components and ensure that most parts could be locally sourced or produced via 3D-printing. Finally, the design prioritised ease of use and maintenance to enhance usability in diverse settings.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDevice key components\u003c/em\u003e\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003eA 3D-printed venturi tube to generate the pressure difference;\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eA 3D-printed casing that ensures airtight seals for the device and locks together to protect components;\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eAn electronic board comprising of an Arduino UNO R3, differential pressure NXP sensor (model number:MPX5010DP) and indicator electronics such as light emitting diodes (LEDs) and a liquid crystal display (LCD). \u0026nbsp;\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003e\u0026nbsp;Tube design\u003c/p\u003e\n\u003cp\u003eA custom tube was designed using Autodesk Fusion 360, which was aligned with our key design priorities. The tube design follows on Bernoulli’s principle (28), incorporating two distinct internal cross-sectional areas, so that when air flows through it, a differential pressure is created at 2 points, relative to the different cross-sectional areas. This differential pressure is correlated with the to a flow rate, as per Equation 1 (28).\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eWhere:\u003c/em\u003e\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cem\u003eQ is the volumetric flow rate (m\u003csup\u003e3\u003c/sup\u003e/s),\u0026nbsp;\u003c/em\u003e\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eA\u003csub\u003e1\u0026nbsp;\u003c/sub\u003eand A\u003csub\u003e2\u0026nbsp;\u003c/sub\u003eare the cross-sectional areas of the cylindrical tube at the level of two different ports (m\u003csup\u003e2\u003c/sup\u003e),\u0026nbsp;\u003c/em\u003e\u003c/li\u003e\n \u003cli\u003e\u003cem\u003ep\u003csub\u003e1\u003c/sub\u003e and p\u003csub\u003e2\u003c/sub\u003e are the pressures at the same points (Pa), and\u0026nbsp;\u003c/em\u003e\u003c/li\u003e\n \u003cli\u003e\u003cem\u003erho is the density of air, set at 1.204 kg/m\u003csup\u003e3\u003c/sup\u003e under standard conditions (i.e., 20° C, 101.325kPa)\u0026nbsp;\u003c/em\u003e\u003cem\u003e(29)\u003c/em\u003e\u003cem\u003e.\u003c/em\u003e\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eBased on Equation 1, the two inner diameters of the tube can be found in order to work efficiently with the selected pressure sensor, namely MXP5010DP, which has a pressure range of 10 kPa. The correct design of this would lead to reduced error and better resolution, reducing the amount of pre- and post-processing needed. It was deemed essential to ensure that the maximum pressure difference effectively present between the two ports of the tube was as compatible as possible with the range of the sensor, avoiding situations in which the sensor could either be damaged (over pressure) or could give out readings largely affected by the accuracy. For this purpose, factors such as the maximum user PEF, disposable mouthpiece size, and range of the sensor were all considered.\u003c/p\u003e\n\u003cp\u003eAnother key requirement was the minimum internal diameter of the Venturi tube. The diameter should not be too small to avoid creating a large air flow resistance during exhalation, making the test uncomfortable and affecting measurement accuracy. In addition, the chosen diameter needed to promote smooth air flow through the device, minimising turbulence that would increase sensor noise. Turbulent flow introduced by non-smooth transitions could increase the noise in the sensor and further increase the perceived resistance and accuracy of the measurements. This was done by following the Venturi Tube specifications from the British Standards Institute (30). In particular three main parameters were used: first, the diameter of the throat/restrictive part should not be less than 0.224D (D being the entrance diameter) of the entrance diameter; the conical converging section should have a taper of 10° ± 5°; and finally, the device should have a divergent outlet of not less than 5° and have a minimum length of 1.5 times the throat diameter.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAll these factors are key to ensuring high accuracy.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eElectronic Design\u003c/h3\u003e\n\u003cp\u003eAn Arduino UNO R3 board was chosen to acquire the data via the MXP5010DP differential pressure sensor. These boards are extremely cost-effective solutions compared with traditional microprocessors, costing just £15. Given that they are resilient and have low power consumption, they are ideal for frugal innovation purposes. The circuit for this design would be relatively simple since the sensor only requires 3 pins to be connected: Vout, which would connect to the analogue input pins in the Arduino; a ground pin; and a 5V pin. These would be interfaced by a breadboard.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAdditionally, to increase the usability of the design, 3 LEDs were added to act as indicators for the user as they perform the spirometry test. A red LED to indicate that there is no serial connection between the device and the software, a yellow LED to indicate that the device has established a serial connection and should press the start button to begin data collection, and a green LED to indicate that the user should start exhalation. An LCD screen was used in conjunction with to indicate more specific commands and return spirometry variables at the end.\u003c/p\u003e\n\u003ch2\u003eSoftware\u003c/h2\u003e\n\u003cp\u003eSoftware was developed to collect and process the data from the hardware and return the spirometry values to the user. To do this, an overall flow diagram was developed to describe the steps from data collection to the return of processed data (Figure 1).\u003c/p\u003e\n\u003cp\u003eWith respect to software, two sets of code were developed, one within the Arduino for reading and transferring data and one within MATLAB for data analysis and display. Using this consistent format, returned spirometry data will always be comparable with those of other tests. Similarly, using the workspace feature associated with MATLAB, trials can be run multiple times, changing input variables within the code to determine the optimum settings for spirometry flow analysis.\u003c/p\u003e\n\u003ch2\u003eTesting and Validation\u003c/h2\u003e\n\u003cp\u003eTo test the accuracy of the device and software, two tests were planned, namely one to measure its accuracy in a standard lab test using a two-litre handpump, and one to perform an initial usability study for the device.\u0026nbsp;\u003c/p\u003e\n\u003ch3\u003eTwo-Litre handpump Test\u003c/h3\u003e\n\u003cp\u003eThis test was performed to measure the accuracy of the device at measuring a known volume of air. To do this, the two-litre handpump was connected to the manufactured spirometry tube via a 3D-printed friction-fitted attachment that was designed to eliminate air loss between the pump and the tube. The tube was secured firmly to a table so that it lay horizontally. This eliminated possible fictitious readings that could be generated by moving the tube through the air; similarly, objects behind the device were removed to prevent any reflections of air back into the device.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWith the experimental setup complete, the handpump was fully extended, and the MATLAB program was run. Moreover, the pump handle was compressed until the air was fully expelled from the handpump through the device. The experiment was run 30 times at varying speeds, and the data were saved and recorded for later analysis after each trial. The varying speed helped us investigate whether the flow had a significant influence on the accuracy. This variable would also verify whether the drag equations had been modelled correctly, since the PEF is correlated with drag. The speed range investigated would be hard to determine since this was calculated after the test. However, the duration over which the full 2 litre could be monitored and recorded and would correlate to the flow rate. The range of duration investigated was from 0.3s to 2.5 seconds. The length of duration was monitored and recorded in MATLAB, by assessing the length on the time axis of the flow rate time graph that was above the threshold value during expiration.\u003c/p\u003e\n\u003cp\u003eThe accuracy was calculated via the mean percentage error. Moreover, further comparisons between the signed percentage error and the duration of each blow were made to investigate whether the speed of the input flow affected the accuracy. As a first step, the normality of the data was tested with the Shapiro-Wilk test (31), with a p-value lower than 0.05 indicating \u0026nbsp;a non-normal distribution. The variable distribution proved to be non-normal (W = 0.6758, p-value \u0026lt; 0.001). Subsequently, Spearman’s rho (32) and its p-value were calculated. P-values lower than 0.05 were considered significant.\u003c/p\u003e\n\u003ch3\u003eEarly usability study on the effectiveness of the spirometry device as a lung function test\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThis test was performed to carry out an early usability study on the spirometer.\u003c/p\u003e\n\u003cp\u003eFor this purpose, the test involved recruiting two participants and asking them to simulate the use of the spirometer after receiving instructions. \u0026nbsp; In particular, the user had to stand and take as deep a breath as possible and exhale this air through the tube as quickly as possible, performing the so-called “full forced expiration”. For hygienic reasons, disposable filtered mouthpieces were used (https://www.numed.co.uk/products/disposable-bacterial-viral-filters-for-spirometry-pack-of-50). \u0026nbsp;The two participants were asked to repeat this test 12 times each (with a minimum of a 30 second rest between each blow), however, the first 3 trials were used as mock tests, so that the user could become accustomed to the device. The collected data were then plotted, and the spirometry values that were extracted from the device were compared with normative values for that person’s characteristics. The data was extracted by first establishing the peak flow (The maximum data point) and this became a reference point for the other variables. The Duration of each exhalation was calculated by finding the time stamp of the peak and iteratively working back through data points until the flow rate was a 4% of the maximum on either side of the peak. This percentage was determined a comparable power to that of the noise of the sensor and is negligible for the total volume. The FVC was found by measuring the total area within the duration and the FEV1 by measuring the total area under the curve within the first 1 second of the duration.\u003c/p\u003e\n\u003cp\u003eFor comparison, after collection with the developed device, each participant was also asked to perform 3 forced exhalations through a CE-marked spirometer (Winterthur Medical AG, model no. MSA100) which collected participant PEF and FEV1, to add further insights and allow comparisons.\u003c/p\u003e"},{"header":"Results","content":"\u003ch2\u003eHardware\u0026nbsp;\u003c/h2\u003e\n\u003ch3\u003eTube Design\u003c/h3\u003e\n\u003cp\u003eBased on the boundary condition factors, such as the average PEF of humans (deemed to be 600 L/min (33)), the inlet diameter of the micro bacterial filter (mouthpiece available at https://www.numed.co.uk/products/disposable-bacterial-viral-filters-for-spirometry-pack-of-50) (28mm), and the maximum differential pressure range of the sensor (10 kPa), an outlet diameter of 10 mm was deemed suitable for most accurately modelling and collecting data between the two differential pressure points. This diameter would allow for the generation of a maximum of 8 kPa. These diameters were taken in conjunction with Venturi tube specifications from the British Standards Institute, and a model designed on Autodesk Fusion was made (Figures 2 and 3). In particular, there is a divergent outlet that smoothens flow dissipation, and a smooth transition region between ports to eliminate unpredictable flow. The two diameters at the inlet allow for the mouthpiece to fit comfortably within the device.\u003c/p\u003e\n\u003ch3\u003eCasing\u0026nbsp;\u003c/h3\u003e\n\u003cp\u003eThe casing was designed to hold electronic components secure relative to the spirometry device so that by moving the tube it would not loosen or weaken electrical component connections. The entire venturi tube and case were designed to be assembled in five components. The five components include the Venturi tube, the bottom casing, which allows electronics to be fitted and secured; the top casing to enclose the electronics from the elements; and finally two 3D printed ports which interface between the venturi tube and the differential pressure sensor. The full assembly is shown in Figure 4.\u003c/p\u003e\n\u003cp\u003eThe device in Figure 4 was designed to reduce the amount of printing material and support material for manufacture, so the device was manufactured with five assimilable components rather than one bulk component. This has three benefits. First, within LRS it is important to minimise the precious materials being wasted from support structures, which could be used to make other components. Second, in LRS reliable power supplies are not always guaranteed, so by splitting the device into multiple printable pieces it minimises the risk of total print failure in the event of power loss. Finally, orienting the prints individually to reduce material waste ensures a better print quality and reduces rough surfaces left over from the removed support material, which could harbour more bacteria when in use or influence air flow.\u003c/p\u003e\n\u003cp\u003eThe device is assembled by sliding the Venturi tube into the top casing using guide rails to lock the tube orientation in place; then it slides until it reaches inhibitor blocks in the rails. The Venturi ports that interface between the tube and the sensor are then inserted into the device to lock the Venturi tube from sliding. The electronic components can then be placed, and the sensor can be connected to the tube via plastic stubbing with an internal diameter of 4mm. For the device used in the testing and validation shown in Figure 4, the casing exhibits a hole at the bottom of the device to allow a USB cable to power and transfer data. In the future, this hole will be removed since a battery and Bluetooth will be used.\u003c/p\u003e\n\u003cp\u003eThis device was 3D printed using an Ultimaker 2+ 3D printer with a 2.85 mm polylactic acid (PLA) filament at a 0.1 mm layer height, a wall thickness of 1.2 mm and an infill density of 40%. The print did not require any support material as it was printed with the truncation of the tube at the bottom of the print. This meant that the layer orientation was aligned perpendicularly with the direction of air flow. This resulted in a wall roughness of 10 micrometres.\u003c/p\u003e\n\u003ch2\u003eSoftware\u003c/h2\u003e\n\u003cp\u003eThe Arduino and MATLAB software were designed to interface together so that the MATLAB code could control when the Arduino would collect data from the sensor and when to stop collecting data from the sensor. This was implemented using the serial port so that, when the board detected a new connection, it would start collecting data. Once enough data were collected (10,000 data points lasting over 10 seconds), this connection was terminated.\u003c/p\u003e\n\u003cp\u003eAs mentioned previously, after setting up initial variables for detection from the sensor, the Arduino enters a wait mode where it waits for a serial connection to be established by the computer. Until then, no data is collected. Once a serial connection is established, the code starts collecting data via the analogue read function from the sensor pin every 1 ms. Within this loop, these datapoints are sent to MATLAB via serial printing. This process is iterated for approximately 10 seconds, after which Arduino receives a termination command from the serial connection executed in MATLAB. This causes the device to enter a wait mode again until it is required for the next data collection.\u003c/p\u003e\n\u003cp\u003eThe code operated as expected, as described within the methodology section, following the execution path shown in Figure 1. It successfully interfaced with the Arduino by establishing and disestablishing serial connections with the board to create 10000-bit arrays of data, computing these into accurate volumetric flows, which could be graphically outputted to find key spirometry values.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eTesting and Validation\u003c/h2\u003e\n\u003ch3\u003eTwo-Litre Hand Pump Test\u003c/h3\u003e\n\u003cp\u003eTable 1. Results from the two-litre hand pump validation test\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"548\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTrial\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal volume (L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePEF (L/min)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDuration (S)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSigned percentage error (%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e475.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-3.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e400.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-0.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e391.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e432.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-2.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e441.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e280.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-4.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e320.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-0.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e381.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-0.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.99\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e329.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-0.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e249.82\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-2.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e387.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e389.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e0.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e471.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e109.82\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e1.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e4.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e618.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e0.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e599.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e0.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e348.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-2.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e421.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e0.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e180.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e1.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-4.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e396.26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e600.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e544.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e281.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e112.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e1.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e2.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e72.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e2.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e26\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e602.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-2.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e496.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e576.73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e627.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 47px;\"\u003e\n \u003cp\u003e30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 113px;\"\u003e\n \u003cp\u003e1.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e678.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 198px;\"\u003e\n \u003cp\u003e-1.5\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 548px;\"\u003e\n \u003cp\u003e\u003cem\u003eL: Litres\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eL/min: Litres per minute\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eS: Seconds\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003e\u003cem\u003e%: percentage\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cem\u003eTable showing the measured forced vital capacity (FVC in L), peak expiratory flow (PEF L/min) and duration of each expiration in (s) after each expiration. The signed percentage error compares the measured volume with the expected volume of 2 litres and is the signed percentage difference between these two values.\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn the handpump test, the mean value for the measured volume was 1.983 L, whereas the real value was 2 L, and the mean absolute percentage error was 1.53%, with a mean percentage error of \u0026ndash;0.87%. Table 1 contains all the data related to this. Figure 5 plots the signed percentage error versus the duration. The correlation analysis revealed a Spearman\u0026rsquo;s rho of 0.2717 and a p-value of 0.1464, indicating that there is neither a high nor statistically significant correlation between the input flow and the accuracy of the device.\u003c/p\u003e\n\u003ch3\u003e\u0026nbsp;Early usability study on the effectiveness of the spirometry device as a lung function test\u003c/h3\u003e\n\u003cp\u003eTable 2. Female participant spirometry parameters from the forced expiratory study\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"557\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003eDuration (s)\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003ePEF (L/min)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eFVC (L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eFEV1 (L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003eFEV1/FVC ratio\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e0.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e357.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e2.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e2.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e2.62\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e367.14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e0.94\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e335.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e2.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e0.89\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e328.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e0.90\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e377.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e322.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e0.86\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e402.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e408.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e0.90\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e445.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.39\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e507.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e0.97\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e452.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e0.92\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.66\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e389.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eMean (excluding trial 1,2,3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e1.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e403.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e0.92\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eThe female participant performed a full forced expiratory flow test 12 times and each time the duration for which the total volume to be expelled was recorded in seconds (s). The FVC in litres (L), the peak expiratory flow (PEF) in litres per minute (L/min) \u0026nbsp;and the forced expiratory volume in 1 second (FEV1) in litres (L) were also recorded \u0026nbsp; (trials 1,2 and 3 were removed from the data analysis)\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Table 3. Male Participant Spirometry parameters from Forced Expiratory Study\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eTrial\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003eDuration (s)\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003ePEF (L/min)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003eFVC (L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eFEV1 (L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003eFEV1/FVC ratio\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e524.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.94\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e581.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e590.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.96\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.47\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e564.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e580.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e536.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e571.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.89\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e600.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4.13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.76\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e415.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e3.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.93\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e581.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e4.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.97\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e585.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.92\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003e12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e578.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e4.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 56px;\"\u003e\n \u003cp\u003eMean (excluding trial 1,2,3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 95px;\"\u003e\n \u003cp\u003e1.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e562.80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 123px;\"\u003e\n \u003cp\u003e4.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.84\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e0.94\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eThe male participant performed a full forced expiratory flow test 12 times and each time the duration of the total volume to be expelled was recorded in seconds (s). The FVC (FEV1) in litres (L) were also recorded \u0026nbsp; (trials 1,2 and 3 were removed from data analysis)\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;In the full forced expiratory flow test with participants the key spirometry parameters for both male and female subjects are reported in Tables 2 and 3. The first 3 trials for each subject were excluded from this data analysis, as the participants were using these trials to familiarise themselves with the device and would significantly affect the results.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe female participant achieved a mean FVC (i.e., total volume) of 3.58 L, a mean PEF of 399.88 L/min and a mean FEV1 of 3.30 L. According to a global lung initiative calculator validated with 74,185 participants (34), the normative values of a woman of her height and age for the FVC should fall in the range of 3.05 L \u0026ndash; 4.83 L for the lower limit of normality (LLN) and upper limits of normality (ULN) with a predicted value of 3.93 L. The FEV1 expected was between 2.68 L LLN and 4.20 L ULN, with a predicted of 3.46 L. \u0026nbsp;Regarding the PEF, the normative value would be 477.60 L/min according to equations derived from R J Knudson (1993) et al (35).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe male participant achieved a mean FVC of 4.06 L, a mean FEV1 of 3.84 L and a mean PEF of 554.20 L/min. The normative value of a man of his height and age for the FVC should fall in the range of 4.027 L \u0026ndash; 6.272 L for LLN and ULN with a predicted value of 5.141 L (34). The expected FEV1 was between 3.429 LLN and 5.282 ULN, with a predicted of 4.478 L. As regards the PEF, the normative value would be 623.4 L/min (35).\u003c/p\u003e\n\u003cp\u003eThe full expiratory volume expelled over time for the 12 trials are shown in Figures 6 and 7 for the female and male participant respectively.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe results of the CE marked device are hereby reported in Table 4\u003c/p\u003e\n\u003cp\u003eTable 4. Results from the CE marked spirometry device\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"501\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 226px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFemale participant\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMale participant\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTrial no\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 122px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePEF (L/min)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFEV1 (L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePEF (L/ min)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFEV1 (L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 122px;\"\u003e\n \u003cp\u003e530.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e3.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e572.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.90\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e2\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 122px;\"\u003e\n \u003cp\u003e529.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e3.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e599.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.86\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e3\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 122px;\"\u003e\n \u003cp\u003e514.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e3.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e588.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.83\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 86px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMean\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 122px;\"\u003e\n \u003cp\u003e524.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e3.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e586.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 85px;\"\u003e\n \u003cp\u003e3.86\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eTable of results from the CE marked spirometry device showing the results from the full forced \u0026nbsp;expiratory flow test. Three trials were used for each participant and after each expiration the peak expiratory flow (PEF) in litres per minute was recorded, as was the forced expiratory volume in 1 second (FEV1) in Litres. The mean values were then calculated and displayed.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eBased on the results of both the two-litre pump test, which assessed the device\u0026apos;s accuracy, and the usability study, which evaluated its validity as a lung screening tool, the device\u0026rsquo;s findings were further compared with standard spirometry testing procedures to determine whether it met the criteria for clinical use.\u003c/p\u003e\n\u003cp\u003eDiagnostic tools for spirometry are typically calibrated using a three-litre syringe as per the ISO 26782:2009 guidelines, according to which a mean percentage error difference of 3% is allowed for the measured volume compared with the actual volume. The reason for this is that the threshold set is intended guarantee reliability when a clinician uses the FEV1/FVC ratio to distinguish between respiratory disease types (obstructive/ restrictive). An error less than 3% is considered negligible and does not skew lung parameters such as the FEV1 enough to affect clinical decision making. In our case, although we used a two-litre pump, we still applied the 3% threshold relative to the measured volume. This results in a more stringent criterion due to the smaller calibration volume. Our measured mean absolute percentage error of 1.53% remains well within these limits.\u003c/p\u003e\n\u003cp\u003eDeviations within the reported percentage error could be attributed to a number of factors such as user errors of the pump, filtering errors that lead to over or underfitting of noisy data, limitations of the low-cost sensor, potential air leakages within the device or even specific individuals\u0026rsquo; coefficient of variation for each expiration. Other diagnostic tools commonly use this error band for similar reasons and is vital for standardising equipment reasonably within the available technology. The error of this device is, therefore, within the boundaries that can be attributed to random noise error equivocal to that of a standard spirometer.\u003c/p\u003e\n\u003cp\u003eThe performance of this device is comparable to that of other low-cost devices reported in the literature, such as those stated within the introduction. These prior studies that used methods such as differential pressure techniques or other approaches achieved accuracies similar to those of this project. Two Federal Drug Administration (FDA)-approved commercial devices: the AirSmart Spirometer (36) and the MIR Spirobank Smart (37), both reported accuracies of \u0026plusmn;3% for volume and \u0026plusmn;5% for flow, and retail at approximately \u0026euro;69 (36) and $190 (38), respectively. Additionally, a custom spirometry device developed by Gupta et al. (39) reported a standard deviation of 8% and a maximum deviation error of 2.5% compared with a standard spirometer. Since the developed device obtained accuracies similar to those for FVC and FEV1 of commercial standard devices, it reinforces the validity of the device as a spirometry instrument. The production cost of the device is estimated to be approximately \u0026pound;45, covering the Arduino board, differential pressure sensor, 3D printing, and additional electronic components. Large-scale manufacturing could further reduce costs through bulk purchasing.\u003c/p\u003e\n\u003cp\u003eCompared with the Venturi-based spirometry device by Ferreira Nunes et al (27), which was most similar to the device reported in this paper, our device, as per this paper narrative, is the result of a frugal, user-driven and contextualised design process, which is paramount for medical devices to be deployed and used effectively in LRSs and LMICs. In terms of performance, their device achieved a mean absolute percentage error of 1.94% (compare to ours of 1.53%) and a signed percentage error of 0.08% when a 3 L cylinder was measured. This was achieved using a Venturi-based system (diameter 32 mm constricting to 10 mm), an ESP32 microcontroller and a Honeywell SSCMRRN005PDAA3 differential pressure sensor. Although the full prototype price was not reported, the retail price of this sensor is \u0026pound;74 (40), and with this high price, the sensor has a large operating range of -34 to 34 kPa and a %Vfss of 0.25%. Compared with MXP5010DP, which has an operating range of 0-10kPa and a %Vss of 5%, the Honeywell sensor is technically better. However, within the context of LRS, the NXP sensor is more suitable. Its lower cost of \u0026pound;20 (41) makes the device more affordable and accessible, with only a minimal sacrifice in performance.\u003c/p\u003e\n\u003cp\u003eThe MXP5010DP sensor also has several technical advantages in these settings. The NXP sensor is an analogue component that outputs voltage relative to pressure, so it is compatible with any microcontroller, the Honeywell sensor is digital and thus requires a much more precise configuration, requiring a digital communication setup, specialised software and library requirements, and can be damaged easily by supplying the wrong voltage to the sensor. This makes debugging in the LRS more difficult. The Honeywell sensor is factory calibrated in comparison to the NXP device, which requires user calibration to take into account possible voltage offset; however, by allowing user calibration each device can automatically calibrate with the available data making the device more suitable for specific settings, and when the NXP is paired with a suitable low pass filter it can reduce noise to a comparable level to that of the digital sensor. Overall, although the NXP sensor is less technically precise, it still has an error within \u0026plusmn;0.5 kPa, which is clinically acceptable for spirometry whilst maintaining several advantages in LRS over the SSCMRRN005PDAA3 (41).\u003c/p\u003e\n\u003cp\u003eThis device differs from the spirometers noted above in its specific design and suitability for LRS. In comparison, it is a fraction of the price at just \u0026pound;45 for base components, and the components used are all affordable and available within LRS. Furthermore, 3D printing can be implemented, and recyclable materials such as PLA can be used. With access to a 3D printer becoming more commonly available, devices can be created locally. This will reduce reliance on imports for components and allow communities to maintain and repair medical devices without external interventions. This was supported by the open-source nature of the project and the design process, which aims to create an easily assimilable device and minimised postproduction processes that lead to increased waste. The device does not have high power requirements and can be connected to either a laptop or smartphone to power the device and interpret data. Future iterations will involve the use rechargeable batteries. Furthermore, the software required does not need continuous internet connectivity to work, supporting use within more rural locations and mobile deployments.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAs an alternative frugal spirometry device for implementation in LRS, our device comprises of fewer and more affordable components than its available counterparts do, it incorporates user feedback systems that instruct the user how to use it, and it returns the values of FEV1, FVC and PEF back to the user instantly via an LCD screen or via the MATLAB script which provides a visual representation of the data. It uses entirely 3D printed components or off-the-shelf bought electronics, so it can be easily replicated in LRS. Since the design is simplistic and primarily focusing on the basic needs of low-income and remote areas, while maintaining clinical accuracy and addressing financial barriers, it can, therefore, be considered a good solution to enhance respiratory care in LRS.\u003c/p\u003e\n\u003cp\u003eWhile this work proved to have value as an effective spirometry device, it also has limitations. The device was designed to operate in LRS with minimal resources, and as such, the selected components were chosen to reflect this. Therefore, the individual limits of the components tended to be greater because of constraints such as cost. Most notably, the differential pressure sensor (MXP5010P) chosen had the most limitations, which could be attributes to its error. The sensor could measure pressures in the range of 0 to10 kPa (i.e., 75 mmHg). And the datasheet of the sensor states that its accuracy is \u0026plusmn;5% on the full-scale voltage, i.e., 4.5 V, which means \u0026plusmn;0.225 V. This means that, when small values in the lower portion of the sensor range are read, the measurements will be affected more by inaccuracies. For example, with an input of 1 mmHg, the sensor reading without offset is 0.14 volts, which is a comparable number to the stated accuracy of the device. Another possible factor is that at very small and very large readings it is possible that the device has a non-linearity error, such that where the device normally reads linearly, at the extremes, this deviation can lead to different results (42). Finally, the sensor can experience a factor called offset drift. This is when the nominal offset voltage, which is calibrated at the start of each blow, can change over time, affecting the accuracy. All these factors could be attributed to the observations shown in Figure 5 where it can be seen that at longer durations (lower flow rates) there tends to be a greater percentage error. However, there was non-significant correlation between the two variables. These differences may be due to the limits of the sensor itself and to its accuracy, as mentioned above.\u003c/p\u003e\n\u003cp\u003eWith respect to the usability study, it was observed that several factors could affect air flow within the device and, therefore, the spirometry results. Each participant used a microbacterial filter to prevent the spread of bacteria and viruses within the device, as well as prevent the inhalation of particles from the 3D printed parts. Each of these factors has an impedance resistance to the flow that may affect the flow speed. This effect should be minimal, however, and since impedance does not occur across the measured zone. A more prevalent issue was the fit between the joints. Although the device was designed to have a friction-fitted seal, it may not inherently be airtight, especially if, when the device is used, it moves as the person exhales. This could lead to some air not being directed down the spirometry tube. It should be noted that this effect, however, should also be easily detectable by the person during each blow, as well as identifiable on the data as there will be a significant drop-in the flow rate.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTechnique significantly affected the readings from the participants; this is seen by the increase in the total volume in the female participant after each trial. It should be noted this was this participant first time using the device. Within the first 3 trials the volume increased from 2.61 L to 3.0 5L to 3.34 L and finally peaked at 3.73L at trial 11. Therefore, the greatest blows should only be considered when comparing forced expiration blows.\u003c/p\u003e\n\u003cp\u003eFinally, since the design uses 3D printed parts, these parts influence fluid flow within the device. The feature that was considered was the effect of wall drag due to 3D printed layer structures; this effect was mitigated by calculating this pressure drop within the code estimating wall roughness from printer settings. The wall thickness of the structures was increased to 1.8bmm thick to limit the possibility of air permeating through the structure, at lower thicknesses more air may escape from flaws within the structure. Resulting in a smaller pressure drop.\u003c/p\u003e\n\u003cp\u003eGiven the greater implications of these devices, this study shows promise in developing a real-world solution to better manage respiratory disease in LRS. By reducing costs, focusing on more open-source designs and orienting the design around the contextualised needs, a more effective device can be produced. This could increase the rates of early diagnosis as well as better monitor disease progression locally.\u003c/p\u003e\n\u003cp\u003eFuture work to consider includes investigating the robustness of the casing for electronics, shifting towards the use of rechargeable batteries, and Bluetooth monitoring. Finally, more data should be collected from a greater number of participants to validate the device further. Within this data collection, there would be a secondary aim to balance the data collection between healthy participants and people with respiratory diseases. This could then be used in future studies to train a machine learning model to help identify different respiratory diseases. By doing so, the device can shift to becoming more of clinical decision-making tool, which would help less skilled user determine the appropriate course of action when dealing with their lung health. The device should further focus on integrating more generalised policies surrounding spirometry devices, focusing on how it can implement more international standards and FDA legislation so it can be more widely adopted. Scaling these frugal spirometry devices for wider adoption further aligns with the World Health Organisation (WHOs) sustainable development goals, goal 3: Good health and wellbeing to combat respiratory diseases and sustainable development goals 7 and 9 which support frugal engineering as a form of sustainable development.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study designed, developed and validated a novel and frugal 3D-printed spirometry device aimed at addressing the contextual needs and challenges of LRS that are often neglected in the design of medical devices. Our study explored unique data processing methods to model fluid flow within a 3D-printed constricted spirometry tube.\u003c/p\u003e\n\u003cp\u003eThe device was validated through two sets of tests. First, its accuracy in volume measurement was assessed by pumping a known volume of 2 litres through the device and recording the output. The device demonstrated a mean absolute percentage error of 1.53% compared with the known volume, indicating a high level of measurement accuracy. Second, the device\u0026apos;s usability for assessing lung function was evaluated. The male participant achieved a mean FVC of 4.03 L and a mean PEF of 557 L/min, while the female participant achieved a mean FVC of 3.44 L and a mean PEF of 395 L/min. When predicted normal values based on the participants\u0026apos; physiological characteristics were compared with a validated Global Lung Initiative calculator (35), the measurements fell within expected ranges.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eComparison with the CE-marked spirometer further supported the validity of the device. The CE-marked device recorded mean PEFs of 586.33 L/min and 524.33 L/min and FEV1 values of 3.86 L and 3.75 L for the male and female participants, respectively. These findings suggest that the developed device provides measurements that are comparable to those of a certified spirometry system, supporting its potential for use in lung function monitoring.\u003c/p\u003e\n\u003cp\u003eThis study demonstrates the implications that frugal engineering could have, when applied to medical devices. It specifically addresses the challenging environments that often hinder resource constrained settings and enables designs to adapt to a diverse range of contexts so that they are effective and sustainable. This study further underpins that by considering only the required contextual needs, devices can be developed at fraction of the cost without influencing clinical accuracy.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eOverall, by developing and adopting more frugal devices, the gap between HIC and LMICs can be reduced. This provides better access to health care for LMICs\u0026rsquo; and places less reliance upon HICs for donations or aid. Furthermore, the developments for LMICs can also benefit HICs as they address problems uniquely that might not have been identified. This could lead to the adoption of context designed devices into HICs, especially within more impoverished areas within HICs. This reverse innovation could provide affordable solutions to address consistent health problems that have been affecting underrepresented populations within HICs, leading to better worldwide health at all levels of health care.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eHigh Income Countries (HICs)\u003c/p\u003e\n\u003cp\u003eLow Middle Income countries (LMICs)\u003c/p\u003e\n\u003cp\u003eChronic Obstructive Pulmonary Disease (COPD)\u003c/p\u003e\n\u003cp\u003eForced Expiratory Volume in one second (FEV1)\u003c/p\u003e\n\u003cp\u003eForced Vital Capacity (FVC)\u003c/p\u003e\n\u003cp\u003ePolylactic acid (PLA)\u003c/p\u003e\n\u003cp\u003elower limit of normality (LLN)\u003c/p\u003e\n\u003cp\u003eupper limits of normality (ULN)\u003c/p\u003e\n\u003cp\u003elow resource settings (LRSs)\u003c/p\u003e\n\u003cp\u003eFederal Drug Administration (FDA)\u003c/p\u003e\n\u003cp\u003eLiquid Crystal Display (LCD)\u003c/p\u003e\n\u003cp\u003eLight Emitting Diodes (LEDs)\u003c/p\u003e\n\u003cp\u003eWorld Health Organisation (WHOs)\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eEthics Approval and Consent to Participate\u003c/h2\u003e\n\u003cp\u003eThis study was approved by the University of Warwick Biomedical and Scientific Research Ethics Committee (BSREC) Application Reference: BSREC 145/23-24.\u003c/p\u003e\n\u003cp\u003eBefore the study started, investigators introduced the purpose of the survey to the participants who volunteered to participate in the study and obtained their informed consent.\u003c/p\u003e\n\u003ch2\u003eConsent for Publication\u003c/h2\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch2\u003eAvailability of Data and Material\u003c/h2\u003e\n\u003cp\u003eThe data is available upon a reasonable request. Please put in this request to the corresponding author James Wallace ([email protected]).\u003c/p\u003e\n\u003ch2\u003eCompeting Interests\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eJames Wallace was funded by the EPSRC (EP/W524645/1).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePhilip Anyanwu and Charles F Hayfron-Benjamin\u0026nbsp;contributions to this paper are part of their work within the African Research Universities Alliance - Multimorbidity Cluster of Research Excellence (ARUA-AMCORE), a collaborative initiative led by ARUA and the Guild.\u003c/p\u003e\n\u003ch2\u003eAuthor Contributions\u003c/h2\u003e\n\u003cp\u003e\u003cstrong\u003eJames Wallace:\u0026nbsp;\u003c/strong\u003eConceptualisation, Methodology, Software, Validation, Formal Analysis, Investigation, Data Curation, Writing-Original Draft, Writing- Review and Editing, Visualisation.\u003cstrong\u003e\u0026nbsp;Davide Piaggio:\u0026nbsp;\u003c/strong\u003eConceptualisation, Methodology, Validation, Investigation, Resources, Writing \u0026ndash; Review and Editing, Supervision, Project Administration, Funding Acquisition.\u003cstrong\u003e\u0026nbsp;Philip Anyanwu:\u0026nbsp;\u003c/strong\u003eWriting \u0026ndash; Review and Editing, Supervision, Project Management, Funding Acquisition\u003cstrong\u003e. Pedro Rifa-Checa:\u0026nbsp;\u003c/strong\u003eConceptualisation, Methodology, Software, Investigation, Writing \u0026ndash; Original Draft.\u003cstrong\u003e\u0026nbsp;Thanusan Kannathasan:\u0026nbsp;\u003c/strong\u003eSoftware,\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eWriting \u0026ndash; Original Draft. \u003cstrong\u003eCharles F Hayfron-Benjamin:\u003c/strong\u003e Conceptualisation, Methodology, Writing \u0026ndash; Review and Editing.\u003c/p\u003e\n\u003ch2\u003eAcknowledgements\u003c/h2\u003e\n\u003cp\u003eThe authors would like to thank Prof Petr Denissenko at the University of Warwick for providing useful insight into the usability of project and guiding fluid dynamic considerations. The authors would also like to thank the School of Engineering Build space for allowing the use of their 3D printers. Finally, the authors thank Alvin Vyapooree (Medical graduate from Warwick Medical School) for his inputs to the medical aspects of the project.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eOrganisation WH. WHO global model regulatory framework for medical devices including in vitro diagnostic medical devices. Geneva; 2017.\u003c/li\u003e\n\u003cli\u003eArasaratnam A, Humphreys G. Emerging economies drive frugal innovation. World Health Organization Bulletin of the World Health Organization. 2013;91(1):6.\u003c/li\u003e\n\u003cli\u003eBank W. Current Health Expenditure Per Capita, PPP (Current International $) 2025 [Available from: https://data.worldbank.org/indicator/SH.XPD.CHEX.PP.CD?most_recent_value_desc=false.\u003c/li\u003e\n\u003cli\u003eOrganisation WH. Medical Devices: Managing the Mismatch: An Outcome of the Priority Medical Devices Project. Geneva; 2010.\u003c/li\u003e\n\u003cli\u003eOngaro AE, Ndlovu Z, Sollier E, Otieno C, Ondoa P, Street A, et al. Engineering a sustainable future for point-of-care diagnostics and single-use microfluidic devices. Lab on a Chip. 2022;22(17):3122-37.\u003c/li\u003e\n\u003cli\u003eOrganisation WH. Medical device donations: considerations for solicitation and provision. second edition ed. Geneva2024.\u003c/li\u003e\n\u003cli\u003eIbrahim NH, Wallace J, Piaggio D, Pecchia L. Validation of a framework for assessing healthcare facilities in low-resource settings: a field study in selected facilities in Ethiopia. Health and Technology. 2025:1-9.\u003c/li\u003e\n\u003cli\u003eLeong D, Teo K, Rangarajan S, Lopez-Jaramillo P, Avezum Jr A, Orlandini A. World Population Prospects 2019. Department of Economic and Social Affairs Population Dynamics. New York (NY): United Nations; 2019 (https://population. un. org/wpp/Download/, accessed 20 September 2020). The decade of healthy ageing. Geneva: World Health Organization. World. 2018;73(7):362.\u003c/li\u003e\n\u003cli\u003eTran VT, Ravaud P. Frugal innovation in medicine for low resource settings. BMC Med. 2016;14(1):102.\u003c/li\u003e\n\u003cli\u003eMontesinos L, Checa Rif\u0026aacute; P, Rif\u0026aacute; Fabregat M, Maldonado-Romo J, Capacci S, Maccaro A, et al. Sustainability across the medical device lifecycle: a scoping review. Sustainability. 2024;16(4):1433.\u003c/li\u003e\n\u003cli\u003ePiaggio D, Castaldo R, Cinelli M, Cinelli S, Maccaro A, Pecchia L. A framework for designing medical devices resilient to low-resource settings. Globalization and health. 2021;17(1):64.\u003c/li\u003e\n\u003cli\u003ePiaggio D, Namm G, Melillo P, Simonelli F, Iadanza E, Pecchia L. Pupillometry via smartphone for low-resource settings. Biocybernetics and Biomedical Engineering. 2021;41(3):891-902.\u003c/li\u003e\n\u003cli\u003eWilliams E, Piaggio D, Andellini M, Pecchia L. 3D-printed activated charcoal inlet filters for oxygen concentrators: a circular economy approach. Development Engineering. 2022;7:100094.\u003c/li\u003e\n\u003cli\u003ePiaggio D, Hyland S, Maccaro A, Iadanza E, Pecchia L. A 3D-printed condom intrauterine balloon tamponade: Design, prototyping, and technical validation. Plos one. 2024;19(6):e0303844.\u003c/li\u003e\n\u003cli\u003ePiaggio D, Castaldo R, Garibizzo G, Iadanza E, Pecchia L. A smartphone-based tool for screening diabetic neuropathies: A mHealth and 3D printing approach. Biomedical Signal Processing and Control. 2024;89:105807.\u003c/li\u003e\n\u003cli\u003eOrganisation WH. The Top 10 Causes of Death: World Health Organisation; 2024 [Available from: https://www.who.int/news-room/fact-sheets/detail/the-top-10-causes-of-death.\u003c/li\u003e\n\u003cli\u003eVos T, Lim SS, Abbafati C, Abbas KM, Abbasi M, Abbasifard M, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990\u0026ndash;2019: a systematic analysis for the Global Burden of Disease Study 2019. The lancet. 2020;396(10258):1204-22.\u003c/li\u003e\n\u003cli\u003eOrganisation WH. Chronic Obstructive Pulmonary Disease: World Health Organsiation; 2023 [Available from: https://www.who.int/news-room/fact-sheets/detail/chronic-obstructive-pulmonary-disease-(copd).\u003c/li\u003e\n\u003cli\u003eMeghji J, Mortimer K, Agusti A, Allwood BW, Asher I, Bateman ED, et al. Improving lung health in low-income and middle-income countries: from challenges to solutions. Lancet. 2021;397(10277):928-40.\u003c/li\u003e\n\u003cli\u003eYadav H, Shah D, Sayed S, Horton S, Schroeder LF. Availability of essential diagnostics in ten low-income and middle-income countries: results from national health facility surveys. Lancet Glob Health. 2021;9(11):e1553-e60.\u003c/li\u003e\n\u003cli\u003eTHOMAS PS, NG C, BENNETT M. Peak expiratory flow at increased barometric pressure: comparison of peak flow meters and volumetric spirometer. Clinical Science. 2000;98(1):121-4.\u003c/li\u003e\n\u003cli\u003eMoore V. Spirometry: step by step. Breathe. 2012;8(3):232-40.\u003c/li\u003e\n\u003cli\u003eVanjare N CS, Madas S, Kodgule R, Gogtay J, Salvi S. . Use of spirometry among chest physicians and primary care physicians in India. 2016.\u003c/li\u003e\n\u003cli\u003eZhou J, Li X, Wang X, Yu N, Wang W. Accuracy of portable spirometers in the diagnosis of chronic obstructive pulmonary disease A meta-analysis. NPJ Primary Care Respiratory Medicine. 2022;32(1):15.\u003c/li\u003e\n\u003cli\u003eCarpenter DM, Jurdi R, Roberts CA, Hernandez M, Horne R, Chan A. A review of portable electronic spirometers: implications for asthma self-management. Current allergy and asthma reports. 2018;18:1-10.\u003c/li\u003e\n\u003cli\u003eWu Z, Huang R, Zhong L, Gao Y, Zheng J. Technical performance analysis of different types of spirometers. BMC Pulmonary Medicine. 2022;22:1-7.\u003c/li\u003e\n\u003cli\u003eFerreira Nunes M, Pl\u0026aacute;cido da Silva H, Raposo L, Rodrigues F. Design and Evaluation of a Novel Venturi-Based Spirometer for Home Respiratory Monitoring. Sensors. 2024;24(17):5622.\u003c/li\u003e\n\u003cli\u003eIgwe Johnson O. Development and Test Performance of a Benchtop Venturi Tube. NIPES-Journal of Science and Technology Research. 2023;5(2).\u003c/li\u003e\n\u003cli\u003eCaretto L. Bernoulli Equation: California State University, Northridge; 2008 [Available from: https://www.csun.edu/~lcaretto/me390/03-Bernoulli.pdf.\u003c/li\u003e\n\u003cli\u003eBoyes W. Instrumentation reference book: Butterworth-Heinemann; 2009.\u003c/li\u003e\n\u003cli\u003eShapiro SS, Wilk MB. An Analysis of Variance Test for Normality (Complete Samples). Biometrika. 1965;52(3/4):591-611.\u003c/li\u003e\n\u003cli\u003eSchober P, Boer C, Schwarte LA. Correlation coefficients: appropriate use and interpretation. Anesthesia \u0026amp; analgesia. 2018;126(5):1763-8.\u003c/li\u003e\n\u003cli\u003eClinic C. Peak flow meter: how to use, measurement \u0026amp; results Cleveland (OH): Cleveland Clinic; 2024 [Available from: https://my.clevelandclinic.org/health/treatments/peak-flow-meter.\u003c/li\u003e\n\u003cli\u003eBowerman C, Bhakta NR, Brazzale D, Cooper BR, Cooper J, Gochicoa-Rangel L, et al. A race-neutral approach to the interpretation of lung function measurements. American journal of respiratory and critical care medicine. 2023;207(6):768-74.\u003c/li\u003e\n\u003cli\u003eKnudson RJ, Lebowitz MD, Holberg CJ, Burrows B. Changes in the normal maximal expiratory flow-volume curve with growth and aging. American Review of Respiratory Disease. 1983;127(6):725-34.\u003c/li\u003e\n\u003cli\u003eRamos Hernandez C, Nunez Fernandez M, Pallares Sanmartin A, Mouronte Roibas C, Cerdeira Dominguez L, Botana Rial MI, et al. Validation of the portable Air-Smart Spirometer. PLoS One. 2018;13(2):e0192789.\u003c/li\u003e\n\u003cli\u003eRamsey RR, Plevinsky JM, Milgrim L, Hommel KA, McDowell KM, Shepard J, et al. Feasibility and preliminary validity of mobile spirometry in pediatric asthma. J Allergy Clin Immunol Pract. 2021;9(10):3821-3.\u003c/li\u003e\n\u003cli\u003eResearch MMI. Spirobank Smart \u0026ndash; Personal Bluetooth Spirometer 2024 [Available from: https://shopusa.spirometry.com/product/spirobank-smart/.\u003c/li\u003e\n\u003cli\u003eGupta S, Chang P, Anyigbo N, Sabharwal A. mobileSpiro: accurate mobile spirometry for self-management of asthma. Proceedings of the First ACM Workshop on Mobile Systems, Applications, and Services for Healthcare; Seattle, Washington: Association for Computing Machinery; 2011. p. Article 1.\u003c/li\u003e\n\u003cli\u003eComponents R. Honeywell Piezoresistive Pressure Sensor, 34kPa Operating Max, Surface Mount, 8-Pin, 206kPa Overload Max, SMT 2024 [Available from: Honeywell Piezoresistive Pressure Sensor, 34kPa Operating Max, Surface Mount, 8-Pin, 206kPa Overload Max, SMT.\u003c/li\u003e\n\u003cli\u003eComponents R. NXP Low Pressure Sensor MPX5010DP. 2025.\u003c/li\u003e\n\u003cli\u003eLukat R. The 3 Main Errors That Affect Pressure Sensor Accuracy 2025 [Available from: https://blog.wika.com/us/knowhow/pressure-sensor-accuracy-3-errors/.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Medical device, Spirometer, Spirometry, Respiratory diseases, Asthma, COPD, Low-resource settings, Resource-Limited Settings, Low- and Middle-Income Countries, Frugal innovation","lastPublishedDoi":"10.21203/rs.3.rs-6812011/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6812011/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMedical devices are essential for maintaining resilient health systems worldwide. However, their distribution does not reflect this, with as many as 76% of devices being used by 13% of the world’s population. Most devices, being designed for high-income countries and failing to consider the local needs of other settings, often fail when deployed in other contexts. A frugal approach can be used to design more resilient systems. Spirometers show promise for developments in this area, owing to the high burden of respiratory conditions in low- and middle-income countries and the unavailability of frugal versions that can enable local care in such settings. Therefore, this work aims to develop an affordable spirometer for low-resource settings.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA frugal approach was used to design a Venturi-style spirometer based on Bernoulli’s principles, leveraging 3D-printed parts and components such as an Arduino Uno R3 and differential pressure sensor (MXP5010DP). The accompanying software to process the signal and calculate relevant variables was developed on Arduino IDE and MATLAB. The device was validated via a hand-pump test and with an initial usability study on two subjects together with a CE-marked benchmark. The mean signed percentage error was then calculated to evaluate accuracy. To determine whether the input flow affected the measured volume Spearman’s rho and p values were calculated.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe device was successfully 3D printed from PLA, and then assembled with electronic components. All the software functioned as expected when it was run. In the two-litre pump test, the device achieved a mean reading of 1.983 L (true value 2L) and an accuracy of 1.53% (mean absolute percentage error). The Spearman’s Rho test revealed that there was no significant correlation between flow speed and volume. In the usability study, the device achieved readings similar to those of the CE marked device, and all values fell within normative values.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions:\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study designed, prototyped and validated a spirometer that can be used as a lung screening tool that achieves high accuracies comparable to those of other portable spirometers. This study also validated that frugal engineering could reduce the cost of devices without affecting the clinical accuracy.\u003c/p\u003e","manuscriptTitle":"Design, Manufacture and Validation of a Spirometry Device aimed for Low-Resource Settings","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-11 08:19:51","doi":"10.21203/rs.3.rs-6812011/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"8618eec8-b5a4-45ce-85fe-224a6f9e8e4a","owner":[],"postedDate":"June 11th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-06-11T08:19:51+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-11 08:19:51","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6812011","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6812011","identity":"rs-6812011","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}