Accurate and rapid diagnostic tests that are simultaneously affordable remain elusive, typically due to trade-offs between performance and cost. Conventional nucleic acid tests offer high sensitivity but require complex, expensive steps such as amplification and purification, whereas lateral flow assays are simple and low-cost but lack the necessary sensitivity for many applications. New analytical approaches provide an opportunity to bridge this gap, raising sensitivity but not material expense. We present a miniaturized and simplified chip-based platform that combines three components into a single diagnostic pipeline: we use (i) spectrally distinct silver and gold nanoparticles (NPs) that form analyte-dependent clusters with unique spectral fingerprints, (ii) a one-pot, enzyme- and purification-free assay on a chip integrated with a high-throughput automated low-cost microscope, and (iii) a morphology-guided convolutional Graph Neural Network that embeds morphology information into convolutional kernels and performs graph-based relational learning across particle-level features. This integration captures spectral, spatial, and morphological quantification at the particle level, rather than relying on bulk spectral shifts, thereby addressing key limitations observed in representative nanoparticle (NP) assays and image-level deep learning baselines under particle-parallel analysis. Mc-GNN is able to process up to 5,000 particle crops per image in a single computational step, using less than 2 GB of GPU memory. This allows us to achieve femtomolar sensitivity with 98.2% recall for synthetic DNA and 94.8% for SARS-CoV-2 RNA from whole viruses, within in-domain experimental variability arising from different NP combinations and assay conditions. By embedding morphological information into the biosensing pipeline, our diagnostic platform is computationally efficient with millisecond-scale neural compute per raw image on a single consumer-grade GPU and is readily extensible to new analytes and multiplexing, offering a scalable solution for a fieldable diagnostic tool.
Jhawar et al. (Tue,) studied this question.