Smart REASSURED Sensors via Machine-Augmented Printable On-Paper Arrays

Abstract

A major challenge in point-of-care (PoC) diagnostics is developing low-cost, scalable sensing platforms that provide high sensitivity, multiplexing capability, and intelligent data interpretation—without dependence on bulky instrumentation. In this perspective, we focus on pattern-recognition-based printable on-paper PoC sensors, rather than conventional lock-and-key receptor-specific systems, as a more practical and adaptable strategy for cellulose substrates. Paper's intrinsic properties—biodegradability, capillarity, and affordability—combined with its limited molecular selectivity make it ideally suited for cross-reactive sensor arrays, where analyte discrimination arises from collective response patterns rather than single-site binding. We discuss how these systems leverage compatibility with scalable printing techniques and explore surface modifications and material strategies to overcome challenges such as roughness, thermal instability, and moisture sensitivity. The Perspective further reviews key printing methods spanning accessible prototyping to high-throughput fabrication and highlights the shift toward array-based sensing coupled with machine learning (ML) for data interpretation. Core ML approaches—including preprocessing, classification, clustering, and regression—are discussed in the context of multidimensional signal analysis and model validation. Together, these insights outline a pathway toward intelligent, scalable, and REASSURED-aligned PoC diagnostic platforms.