From lab to fab: overcoming the design–manufacture divide in microfluidic product development

1 Introduction

Microfluidic technologies now sit at the core of many advances in diagnostics and life-science tools, yet only a minority of laboratory concepts make the transition into manufacturable products. At the same time, the need for robust, deployable microfluidic systems is increasing. Ageing populations, rising chronic-disease burden, and overstretched hospital infrastructures are pushing diagnostics toward decentralised settings. Pandemic preparedness likewise depends on technologies that can scale rapidly and operate outside conventional laboratory environments. Microfluidics is one of the few engineering approaches capable of shrinking complex biochemical workflows into formats suitable for distributed care, real-time surveillance, and high-throughput response.

Yet despite this strategic importance, the broader vision of a portable, automated, inexpensive “lab-on-chip” remains limited to very few examples. Across the projects we have worked on within the Microfluidics Innovation Hub (MIH)—a European initiative established to close the lab-to-fab gap—we have seen an almost universal dynamic: early prototypes demonstrate compelling science, but the moment devices transition to industrial materials, scalable fabrication, or regulated workflows, fragility emerges. Across projects supported within the Microfluidics Innovation Hub, commercialisation timelines have consistently exceeded initial expectations, with essentially all projects experiencing delays of one to 2 years or more relative to original roadmaps, illustrating the systemic difficulty of translating promising prototypes into market-ready products. Architectures built for proof-of-concept performance often rely on laboratory processes that cannot survive the realities of market and manufacturing. Many developers continue to underestimate the product development effort required to transition from functional prototypes to reliable, mass-producible devices. The result is a frequent disconnect between core scientific innovation and engineering realities.

Here we aim to synthesise the technical, organisational, and market-alignment pitfalls we have encountered while supporting microfluidic product development at scale. We draw on recurring patterns seen across real-world projects in diagnostics, environmental testing, and life-science tools. Our aim is to provide researchers, founders, and engineers with a framework for anticipating the design–manufacture conflicts that impede scale-up, and to outline practical strategies to mitigate them early. We argue that successful translation of microfluidic technologies requires the concurrent maturation of three interdependent dimensions: technological readiness, manufacturing readiness, and application or market readiness. While this Perspective focuses primarily on product design and manufacturing challenges, these dimensions evolve together in practice. Throughout this work, we therefore consider lab-to-fab translation as a co-evolutionary process, while concentrating specifically on the design and manufacturing challenges that most frequently limit progress.

2 Alignment to market and user needs

The first and most frequent reason why microfluidics ventures struggle or fail is poor alignment to actual market and user needs. Academic research, where many microfluidic products originate, is exploratory by nature, whereas commercialisation requires a precise understanding of use context, business model, and market potential. Most academic microfluidics research is not intended to produce near-term commercial products, and this distinction is both appropriate and necessary: foundational advances in physics, materials, and biological methods provide the knowledge base upon which future products depend, even when individual projects are not themselves translated. However, when projects pursue explicit translational or product-oriented objectives, additional constraints beyond scientific performance must be addressed. In this context, development efforts often retain a “technology-push” rather than “market-pull” mindset, with market needs considered too late or insufficiently defined. Successful commercialisation therefore emerges from the interaction of scientific innovation, technology-transfer structures, industrial engineering, and investment ecosystems. Academic laboratories, technology-transfer offices, incubators, industrial partners, and investors each operate under different incentives and constraints that collectively shape outcomes. The observations presented here are therefore not intended as criticism of academic research practice, but as reflections on recurring product-development challenges encountered when scientific concepts transition toward real-world deployment.

POCT and IVD were long assumed to be fertile ground for “killer applications” (Blow, 2007). Many developments are framed as broad platform technologies with multiple potentials, which obscures the absence of a clear use case or market entry point. Molecular and immunodiagnostic platforms are inherently generic, and adaptable systems such LumiraDx (integrated cartridge-reader architecture), Fluxergy (multi-assay platform strategy), and QuantuMDx (focused clinical deployment pathway) deliberately support multiple detection modalities. Yet all ultimately commercialise by focusing on a few well-defined use cases, where they outperform incumbents, before expanding into additional analytes.

Successful translation depends on identifying a concrete problem, quantifying its value, and designing the workflow that solves it. Where solutions enable entirely new workflows (e.g., new diagnostics or drug-discovery steps), adoption requires changing user processes—benefits must be demonstrated clearly to all relevant stakeholders. Diagnostic testing ventures seem particularly affected by poor product-market fit based on our observation and empirical evidence (Peabody, 2021). Stakeholder disparity further complicates these markets: users include clinicians, nurses, laboratory staff, or patients, whereas payers and decision-makers (health systems, insurers) have distinct incentives. In contrast, areas such as sequencing, cell analysis, bioprocessing, and veterinary diagnostics typically involve more aligned user–buyer groups. In all cases, developers must first identify and then satisfy the requirements of all stakeholders.

2.1 Users

Developers must understand who operates the device, in which environment, and under what workflow constraints. End users rarely care about technological novelty: simplicity, robustness, and reliability dominate. Yet we often see these considerations addressed far too late. This leads to poor reception, substantial redesign loops, and delayed development. As an example, for a device based on an optical measurement cell, ease-of-use requires intuitive handling, protection of optical interfaces, and fail-safe insertion features. Early engagement with users and environments, supported by Usability Engineering and Human-Factors methods (required under IVDR, MDR, and FDA), ensures user-centric design from the outset.

2.2 Stakeholders (payers, decision-makers)

These groups tend to be even more technology-agnostic. They prioritise analytical performance that matches or exceeds existing laboratory solutions—still a challenge for many POCT systems (Kahn, 2024)—and strict cost targets. Failure to meet cost constraints remains the single most common reason microfluidic products miss market alignment. The next sections address the two major cost drivers: complexity and manufacturability.

A useful counterexample illustrates how early alignment between application, technology, and manufacturability can substantially de-risk translation. The sepsis diagnostics platform developed by Cytovale demonstrates a design approach in which product constraints were implicitly addressed from the outset. Rather than relying on complex on-chip sensor integration, the system exploits a microfluidic phenomenon—cell deformation under inertial focusing flow—combined with external optical monitoring, limiting critical failure modes largely to controllable manufacturing parameters such as dimensional tolerances and chip bonding, while maintaining robustness to material variability. This enabled a tractable manufacturing architecture based on injection-moulded polymer chips with comparatively simple bonding processes. Equally important was the market context: sepsis triage represents a high-value clinical need with limited alternatives, allowing actionable but non-quantitative outputs to provide clinical value, where a simplified categorical (“red–orange–yellow”) result proved sufficient for decision support. The combination of a clearly defined clinical problem, reliance on robust physical principles rather than fragile integration, and manufacturing-compatible design choices enabled rapid progression from start-up formation to clinical studies within approximately 3 years, culminating in FDA clearance after 9 years (Cytovale, 2025). This example highlights how successful translation often emerges not from maximal technological sophistication, but from early convergence of market need, physical principle, and manufacturable system architecture.

3 Complexity and integration

Integration in microfluidics is often framed as chip-in-lab vs. lab-on-chip. Chip-in-lab devices miniaturise or enhance specific steps (droplet formation, flow control, compartmentalisation) but depend on sophisticated instruments for actuation, thermal control, and detection (Streets and Huang, 2013). For many applications—cell manipulation, research tools, sequencing, industrial bioprocess monitoring—such architectures can be highly effective and commercially successful (e.g., Illumina NGS flow cells or Lumicks tweezers). Miniaturisation is applied only where microfluidics provides unique benefit.

By contrast, lab-on-chip devices aim to integrate all steps—sample-to-answer—into a single autonomous disposable. These devices offer the clearest route to decentralised healthcare (Gurkan et al., 2024), but the less the user does, the more complexity the device must absorb. Manual steps most constrained by microfluidic physics (mixing, metering) are the hardest to automate. The few devices that have reached real scale—blood-gas cartridges, glucose strips—are far less sophisticated than the field’s academic imagination.

Numerous publications showcase highly integrated cartridges with mixing, metering, valving, pumping, reagent storage, and detection, but these often serve as demonstrations of capability rather than manufacturable products. Integration that is trivial at prototype scale becomes disproportionately difficult to industrialise.

The field is advancing, and scalable solutions increasingly exist, but packing all desired functions into a single cartridge almost always produces excessive complexity. The first safeguard is translating user needs into quantifiable input requirements and revisiting them regularly. Tools such as critical-to-quality (CTQ) flowdown or quality function deployment (QFD) help balance user needs against complexity before constraints solidify.

A recurring trap is the assumption that integration challenges can be solved later. When teams reach the stage of developing scalable manufacturing processes, they often discover they have “designed themselves into a corner.” A more robust strategy is to design lean, manufacturable devices that still solve the core user problem. As Whitesides put it: “Be cool, simple, and cheap.” (Whitesides, 2013) Additional functionality can be developed in parallel rather than embedded prematurely.

Some biological workflows can be executed with remarkably simple passive microfluidics (from classical lateral-flow tests to more sophisticated capillary-driven immunoassays like Burst Diagnostics). Within MIH, we repeatedly saw value in strategically off-loading functions to a compact reader to reduce cartridge complexity. Genspeed Biotech exemplifies this: its reader stores and dispenses reagents, enabling a disposable cartridge built around a single capillary channel—simplifying manufacturing while preserving automation (Figures 1a,b).

In other applications, on-chip actuation (valving, pumping) is unavoidable. Modular designs help manage the resulting complexity by integrating proven functional modules that can be separately tested and manufactured. Biflow Systems demonstrates this approach, combining injection-moulded components and PCBs for reagent storage, valving, and pumping (Nestler, 2016), enabling flexible configuration across use cases (Figures 1c,d).

Leveraging partners with validated platforms or OEM frameworks also accelerates industrialisation. In one MIH project developing a multiplex electrochemical sensor chip, the customer initially attempted manual multilayer assembly with limited success. By integrating MIH partner capabilities—simulation assisted microfluidics design, customizing a predeveloped electrochemical sensor platform, and leveraging proven fluid actuation and fluidic/electrical interfacing modules—the concept was rapidly transformed into a manufacturable product.

These examples underscore a central lesson: choosing the right level of integration requires understanding manufacturing implications from the beginning.

4 Manufacturing–design for manufacturability (DfM)

Even after choosing an appropriate device design with respect to complexity and integration level, significant pitfalls and challenges remain for transferring a basic design to manufacturing and scaling-up. Hence, in industrial product development processes usually a separate development phase is dedicated to “Design for Manufacturability” (DfM). The importance of DfM was already introduced many years ago by Holger Becker in one of his legendary focus articles (Becker, 2009). While in the previous phases the main concerns were to develop prototype devices that demonstrate the required performance, DfM activities are centred around optimizing the design towards manufacturing process, yield and costs. Therefore, a clear understanding and trajectory of the targeted production volumes and allowable costs is of utmost importance from the very first moment. When projects transition from one environment to the other, architectural mismatches become visible—sometimes requiring minor optimisations, and in extreme cases, complete redesign.

Importantly, manufacturability constraints are closely tied to early economic decisions that are often underestimated during laboratory development. A frequent misconception is that the microfluidic chip itself dominates product cost, whereas in practice substantial expenditures arise from tooling non-recurring engineering (NRE), backend process automation, quality-system implementation, usability validation, and regulatory documentation. Even at early development stages, tooling investments can be significant. Low-volume prototype tooling—such as family tools with simple monolithic inserts or aluminium moulds with relaxed tolerances on microchannel dimensions—typically incurs costs below approximately €20,000, depending on mastering complexity. Transitioning to high-volume manufacturing substantially increases investment requirements: hardened steel injection-mould tooling designed for multi-cavity production and long operational lifetime can significantly exceed €40,000, with costs strongly dependent on mould complexity, cavity count, and expected production volumes. Consequently, manufacturability considerations are not a late optimisation step but a primary design constraint that should inform architectural and material decisions from the outset. In this sense, manufacturability is not merely a downstream engineering activity but a primary architectural constraint that determines which microfluidic concepts can realistically progress toward industrial implementation.

4.1 Materials

The selection of the material for a microfluidic chip or cartridge is a fundamental decision that is governed by the requirement of functional material properties and by the application-dependent cost structure.

In some cases, technical constraints will clearly favour or exclude certain materials. For example, silicon or glass chips can be required if the application demands exceptional microscale precision, harsh organic solvents, reusability, etc.

Regarding cost structure, not only the cost of material itself needs to be considered, but also the manufacture methods that allow for its production. For most microfluidic-based products typical economy of scales rules apply. Products required in low to medium volumes (e.g., research systems, droplet sorters, organ-on-chip systems) often tolerate higher unit costs, and products required in much higher volumes (e.g., single-use/disposable IVD cartridges) are often extremely price sensitive (e.g., < 10€ COGS at > 10⁵ units per year). The latter is only feasible using polymer materials and high-volume manufacturing technologies such as injection moulding.

If technical requirements demand materials (e.g., silicon, glass) that are inherently more costly or require batch or wafer scale processing, cost saving strategies usually involve reducing the chip footprint to its core functionality, to reduce materials consumption and increase the number of chips per batch/wafer. For chip functions that do not depend on the material itself, hybrid solutions can be designed where such functionalities are “out-sourced” to a cheaper, often polymer-based material.

An important factor that often complicates early materials selection is a marked disconnect between prototyping and series manufacturing materials. While for high volume manufacturing almost exclusively different types of thermoplastic polymer resins (e.g., COC, COP, PMMA, PS, etc.) are used, often entirely different materials are used in early development and prototyping stages. It is naturally preferable to fix and maintain a material choice as an outcome of early product development processes to avoid material changes across different stages. Changing materials almost always requires considerable design adaptions and bear a high risk that performance might be compromised (e.g., due to different surface properties, interactions with biological compounds, leaching of interfering components, etc.). For polymer-based chips, it is key to identify low-volume, flexible prototyping processes, such as hot embossing, or family-tool based injection-moulding, that enable end material testing prior to investing in costly, unflexible production methods.

There are some situations where a change of the material was found to be appropriate when moving into another use case or application with a different volume and cost structure. For example, a microfluidic cell sorting structure originally designed in silicon for a low-volume research application was successfully transferred into a polymer chip with only minor redesign, employing high-precision, variotherm injection-moulding, to enable its use in clinical applications typically requiring a much lower cost structure.

The most widely used prototyping material in academic microfluidics is PDMS, a silicone elastomer valued for its optical clarity, ease of replication, and rapid turnaround. While the microfluidics industry almost universally regards PDMS as unsuitable for large-scale, regulated products—particularly when transitioning to thermoplastic mass production—the notion that it is inherently “non-scalable” is an oversimplification. For specific classes of devices, PDMS and related silicone polymers offer unique functional advantages such as elasticity, gas permeability, and tunable surface properties that are difficult to reproduce in thermoplastics, and when these properties are central to device performance (e.g., organ-on-chip systems, cell-culture platforms, elastomeric valve architectures) PDMS-based products can indeed be commercialised. Companies such as Fluigent have demonstrated this with elastomer-based components, and advances in automated casting and demoulding now enable PDMS-like materials to be produced at relatively high volumes (Filion et al., 2025). However, many devices continue to be designed and even optimised in PDMS primarily because of prototyping convenience, despite the substantial changes in geometry, bonding behaviour, surface interactions, and long-term stability that typically arise during scale-up. In practice, PDMS remains most appropriate when elastomeric functionality or gas permeability is intrinsic to device operation, whereas thermoplastic materials become the preferred choice once requirements such as long-term storage stability, adsorption and leaching control, solvent compatibility, regulatory robustness, and low cost-per-part at production volume dominate design considerations. Switching to thermoplastics early therefore avoids redesign penalties, and for applications requiring elastomeric behaviour (e.g., valving), alternative thermoplastic materials such as elastomeric COC (Gleichweit, 2018) now offer similar functional properties while remaining compatible with standard high-volume manufacturing methods.

A similar misconception exists regarding adhesive tape - a versatile and often essential material for prototyping and early-stage microfluidics development (Smith et al., 2024). For certain microfluidic products (e.g., simple electrochemical test strips), adhesive tape lamination can be an optimal high volume series manufacturing solution. However, for most microfluidic products the adhesive tape approach has to be replaced by more elaborate (e.g., thermocompression, solvent, etc.) bonding methods for scaling.

Another abundantly used prototyping materials are photocurable resins. 3D printing is almost universally used nowadays and is one of the most versatile means for prototyping microfluidic devices (Au et al., 2016). However, results must be carefully interpreted as resin materials are significantly different from typically used injection moulding resin and surface qualities remain incomparable.

4.2 Manufacturing processes

4.2.1 Prototyping vs. serial manufacture

As with materials, manufacturing processes inevitably shift when moving from laboratory prototyping to series production. A wide range of small-scale fabrication methods—micromilling, laser machining, hot embossing, and others—can support early development, but the choice should already anticipate the material and process constraints of the intended high-volume manufacturing route.

Injection moulding remains the dominant target method for microfluidic mass production. It is mature, precise, and capable of combining macro- and micro-features with tolerances down to a few micrometres, scaling efficiently to millions of units per year at low unit cost. Injection moulding comes with well-defined design rules—such as gate placement, draft angles, and wall thickness—that cannot be retrofitted later. Bringing manufacturers into the conversation early ensures that these constraints are understood during prototyping, when design flexibility remains high. Intermediate-scale manufacturing offers a bridge between prototyping and full production. Family-tools, reconfigurable, or soft (aluminium or unhardened steel) tooling enables near-industrial validation of geometry without committing to costly hardened steel tools. In practice, several design constraints that are often underestimated during laboratory development become decisive at this stage, including demoulding considerations such as draft angles and ejection strategies, gating and flow effects that may introduce weld lines or optical artefacts, venting requirements in shallow microcavities to avoid gas trapping, replication limits linked to aspect ratio and feature density, and shrinkage or warpage effects that later drive bonding misalignment. While highly application-dependent, such constraints illustrate how manufacturing realities begin shaping device architecture long before production scale is reached.

We repeatedly found that these intermediate approaches generate manufacturable parts early while also exposing hidden material sensitivities that standard testing often misses. In one project, PMMA prototypes showed internal stress that led to cracking when exposed to certain liquids used during chip operation. In another, moulding conditions altered the material’s autofluorescence, directly impacting optical assay performance. Identifying such behaviours before committing to final tooling allows refinement of materials, bonding strategies, and process parameters, significantly reducing downstream risk and cost.

At the far end of the scalability spectrum, emerging approaches such as roll-to-roll (R2R) manufacturing highlight what the ultimate consolidation of microfluidic production might look like. By enabling continuous replication, bonding, surface treatment, and even reagent deposition on a moving web, R2R demonstrates how disparate process steps could eventually be unified into a single high-throughput workflow—offering a realistic route to ultra-high-volume, low-cost production once market demand supports it (Tören et al., 2020). Yet such technologies also underscore a broader truth: achieving manufacturable scale in microfluidics requires substantial process-development effort and capital investment. Because industrial production methods differ fundamentally from laboratory prototyping, outsourcing to specialised manufacturing partners is often unavoidable—both to access purpose-built infrastructure and to avoid unsustainable in-house CAPEX burdens.

4.2.2 Patterning vs. backend processing

An often-underestimated role is played by back-end processes that follow the actual creation of microfluidic structures—bonding, surface modification, reagent deposition, assembly, or packaging. Typical backend operations introduce distinct classes of risk, for example, bonding processes (thermal diffusion, solvent, adhesive or laser welding) that can deform channels or introduce haze and leachables; surface treatments whose functionality may decay during storage; reagent integration steps that impose repeatability constraints; assembly and alignment challenges driven by tolerance stack-up unless fiducial or poka-yoke features are incorporated; and quality-control requirements arising because many defects remain invisible until late-stage testing.

During early development many of those can be accomplished manually or with simple tools, jigs and fixtures. However, when it comes to process automation and larger scale manufacturing these processes can play an increasingly dominant role (Cong and Zhang, 2022).

Polymer replication itself rarely constitutes the main cost driver nor throughput limiting step in microfluidic manufacture; rather, expense arises from concatenating multiple, highly customised post-processing steps. As an example from one of our projects, an electrode printing step was contributing up to 10x the cost of the original polymer chip moulding step.

4.3 Supply chains

The manufacture of a microfluidic device—especially a highly integrated lab-on-chip—is an inherently exaggerated form of multistep production, combining process steps that originate from entirely different industries, from microstructure tooling, polymer replication, and bonding to surface treatment, printing, biological and chemical reagent integration, assembly, and packaging. This is reflected in the supply chains currently in place for microfluidic systems.

Although it is often intended to combine at least the last backend process steps (such as, e.g., bonding, assembly, marking, packaging) within a dedicated production line, supply chains are typically fragmented and distributed across different suppliers. There are very few real microfluidic foundries that cover entire production value chains. In most situations, biological, fluidic, electronic components, and semi-finished parts such as moulded parts or patterned surfaces, are obtained from different partners for subsequent integration and assembly steps. In the worst case, semi-finished devices are even shipped back and forth between sites for specific process steps such as surface modification or materials deposition. Aside from the required logistics and co-ordination efforts, highly fragmented supply chains also introduce additional validation and incoming goods inspection burdens that compound to overall cost.

As with all multi-step production processes, the yield of individual process steps is of critical importance. If in a five step process every step achieves 90% yield, this compounds into a business-busting yield of ∼60%. Designing for yield therefore means integrating only what is essential for the core function, keeping submodules separable and testable, and applying quality control at critical points.

Seamless compatibility and interoperability between the different actors along the value chains cannot be taken for granted (Ducree, 2019). Compared to other industries having to deal with complex, multistep production processes such as, e.g., the semiconductor and microelectronics industries, often almost every component of microfluidic devices is highly customized and unique. Consequently, a higher level of harmonization and standardization of microfluidic designs, interfaces, manufacturing processes, and test methods would be highly desirable to streamline efficient development and manufacturability of lab-on-a-chip devices.

Respective standardization efforts are being pursued (Reyes et al., 2021) but there is a long way to go until this improves interoperability between suppliers and production process steps.

In the meantime, applying Design-for-Manufacturability (DfM) as a top-guiding principle early during development, or partnering with manufacturing organisations that provide turnkey solutions, or leveraging well established partner structures along the values chain, minimises these risks and accelerates transfer to volume production.

5 Conclusion

Microfluidic innovation has long been framed as an unfulfilled promise. We argue that this perception does not arise from a lack of scientific creativity, but from the intrinsic difficulty of translating interdisciplinary laboratory technologies into manufacturable and economically viable products. Microfluidic systems combine precision microfabrication, materials science, surface chemistry, biology, optics, and fluid dynamics within a single device, creating development challenges that extend far beyond conventional research workflows. The gap between laboratory success and industrial deployment therefore reflects not failure of the field, but the complexity of integrating these domains into robust products.

Successful translation depends on alignment rather than sequential progress. Technology maturity, manufacturability, and application readiness must evolve together, guided by clear product objectives and realistic constraints. Across projects supported within the Microfluidics Innovation Hub, recurring obstacles rarely stem from insufficient scientific performance alone, but from mismatches between design ambition and manufacturing reality, fragmented supply chains, underestimated development effort, and delayed consideration of cost and scalability. Adopting Design-for-Manufacturability (DfM) principles early—treating manufacturing constraints as design inputs rather than downstream optimisation tasks—substantially reduces these risks and helps prevent costly redesign cycles.

Equally important is recognising that commercialisation is a systemic endeavour involving scientific innovation, technology transfer structures, industrial engineering, and investment ecosystems operating under different incentives. Academic research plays a vital and distinct role in advancing foundational knowledge, and most microfluidics research is appropriately not aimed at near-term product creation. However, when projects pursue translational objectives, additional disciplines—including usability, supply-chain robustness, and production economics—must be incorporated alongside scientific performance.

An additional lesson emerging from industrial translation is the importance of early awareness of regulatory and quality-system requirements. While academic teams are not expected to possess detailed regulatory expertise, established frameworks such as ISO 13485 provide a useful reference for understanding the breadth of processes that ultimately shape product development. Early familiarity with concepts such as design controls, verification and validation, risk management, and supplier qualification can help teams anticipate constraints that otherwise emerge late in development. In one MIH-supported project, a customer initially developed a device assuming requirements aligned with MDR classification; subsequent clarification under IVDR introduced sterility requirements that necessitated concatenation of multiple chip production steps into sterile manufacturing environments. This fundamentally altered production architecture, assembly workflows, and cost structure, illustrating how regulatory understanding can directly influence manufacturability decisions from the earliest design stages.

Ultimately, microfluidics will succeed not by maximising technological integration or platform ambition, but by delivering focused solutions that solve clearly defined problems within realistic manufacturing and economic boundaries. As illustrated by successful translation efforts discussed earlier, systems grounded in robust physical principles, constrained integration, and clearly defined clinical value propositions often achieve manufacturable solutions more readily than architectures pursuing maximal technical integration. Platform value emerges only after viable products exist. If the field increasingly internalises concurrent development of design, manufacturing, and product strategy as a cultural norm, microfluidics may finally transition from an inspiring research domain into a mature industrial discipline capable of delivering on the long-standing promise of lab-on-chip technologies beyond the laboratory.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Samar Damiati, University of Sharjah, United Arab Emirates

Reviewed by: Gianluca Trotta, National Research Council (CNR), Italy

Alfredo Edoardo Ongaro, Onalabs Inno Hub, Spain