Poster W6 - Automating Scalable Manufacturing of Point–of–Care Paper-Based Nucleic Acid Detection Tests

Introduction: Achieving precise assembly and component alignment poses a significant challenge in small-scale manufacturing of medical devices, particularly during the early stages of development and prototyping. This challenge is especially evident when producing batches of 10-50 paper-based diagnostic devices for testing and laboratory applications [1]. Manual assembly of devices causes inherent variations in production batches which can lead to a lack of reproducibility[2]. To solve these challenges and enable scalable manufacturing, we are exploring automated pick-and-place assembly techniques. This advancement aims to streamline the production of point-of-care-paper-based nucleic acid detection for infectious diseases such as our lab’s MicroRAAD [3].

Materials and

Methods: A Dobot MG400 robot was used to perform the pick-and-place process at a vacuum pressure of 150kPA. Holders for components were 3D printed on a Prusa printer. Commercial lateral flow assays (LFAs, UStar Biotechnologies) were used to assess any potential damage by the robot’s suction cup end-effector at the detection zone during device assembly. The robot was programmed to lift the LFAs from their center nitrocellulose membrane and place them sequentially in a row. To assess any potential damage by the robot’s end effector at the detection zone, 3LFAs were scanned on an EPSON scanner prior to, and after relocation while dry, as well as after 50uL of water with blue dye (McCormick) was added to the sample pad. The robot also prepared LFIA placement followed by placement of a cellulose membrane with wax valve and polyethylene terephthalate glycol (PETG) sample well. An alignment test was also performed to evaluate the precision of the placement of each component, and the average x-axis displacement and standard deviation were calculated.

Results, Conclusions, and Discussions: No physical damage to the nitrocellulose LFIA was observed from the vacuum pressure of 150kPA (Figure 1a). The water was transported efficiently and evenly along the membrane, showing the control line as expected in (Figure 1b). The alignment test revealed an average displacement of 0.67 ± 0.15 mm. By optimizing 3D printed components for more tight holding of the LFA membranes prior to placement, the alignment was improved for assembly with the wax valve and sample cup (Figure 1c). Upon sequentially placing the LFA and wax valve, fluid transferred evenly through each membrane.
The Dobot’s handling demonstrated no significant damage to the LFA’s membranes and acceptable tolerances of alignment of the paper components. The results provide insight into the acceptable tolerances for ongoing device integration to enable the commercialization of laboratory diagnostic devices toward manufacturable point-of-care tests. Future work includes further improving the alignment and generating a workstation with vision-assisted pick-and-place features for more robust device production.

Acknowledgements (Optional): This project was funded by Moore Inventor Foundation grant number 9867. The authors appreciate Dr. Marina Sala de Medeiros and Guilherme Cecato Delunardo for their initial work on this project.

References:
[1] C. R. Mace and R. N. Deraney, Microfluid. Nanofluidics, vol. 16, no. 5, pp. 801–809, May 2014,
[2] N. Sritong, M. Sala De Medeiros, L. A. Basing, and J. C. Linnes, Lab Chip, vol. 23, no. 5, pp. 888–912, 2023,.
[3] E. A. Phillips et al., Lab. Chip, vol. 19, no. 20, pp. 3375–3386, 2019,.