Posted in | News | Medical Sensor

Graphene and AI combine for rapid COVID-19 detection

A cutting-edge biosensor fuses nanotech and AI to detect COVID-19 faster and more accurately, opening the door for future portable, affordable pandemic-ready diagnostics. 

 Image Credit: PeopleImages/Shutterstock.com

In a Scientific Reports study, researchers present a graphene-silver metasurface biosensor combined with machine learning for highly accurate, label-free COVID-19 detection. The device offers 400 GHz/RIU sensitivity, real-time detection, and scalable, low-cost fabrication suited for point-of-care diagnostics.

Why Are We Still Looking for Solutions to COVID-19?

The emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in late 2019 triggered a global pandemic, revealing weaknesses in surveillance and diagnostic systems.

While antigen tests and reverse transcription polymerase chain reaction (RT-PCR) were crucial, they were limited in two key areas. RT-PCR required labs and time, while antigen tests lacked sensitivity. At a time of urgency, these faults highlighted the demand for innovative, accessible diagnostic solutions globally.

Surface plasmon resonance (SPR) became the solution, offering label-free, highly sensitive, and real-time biomolecular interaction detection. SPR biosensors are fast, require limited sample preparation, and can be miniaturized for portable testing.

Studies have demonstrated that SARS-CoV-2 antigens can be targeted using aptamers, molecularly imprinted polymers, and antibodies. Improvements in thermal control, advanced signal processing, and microfluidics have reduced noise and improved sensitivity.

The integration of artificial intelligence and machine learning also minimized false positives and boosted diagnostic accuracy. Incorporating materials like graphene and perovskites with architectures like nanorings has achieved sensitivities up to 2000 GHz/RIU.

These systems show potential across environmental, industrial, and biomedical domains. Yet, translating prototypes into clinical tools presents challenges, particularly in ensuring scalability, stability, reproducibility, and cost-effective manufacturing for real-world deployment.

The Study

In this work, the researchers proposed a high-performance graphene-silver hybrid metasurface plasmonic biosensor that integrates optimized resonator geometries with machine learning algorithms and advanced nanomaterials. The sensor was designed for scalability, with potential for industrial-scale production exceeding 10,000 units per wafer and an estimated per-sensor cost of between $2 and $5.

Design For Enhanced Plasmonic Resonance

To enhance plasmonic resonance, the metasurface featured a silver-coated central circular ring resonator (3.8 μm inner, 4.3 μm outer diameters). This ring was flanked by two larger rectangular resonators (10 μm × 1.3 μm), and two rectangular components (3 μm × 1.3 μm) were placed methodically on each side of the central circular ring.

These smaller rectangles flank larger ones to form plus-shaped resonators. All rectangular components were silver-coated for resonance enhancement and consistent optical conductivity. These resonators were fixed to a square base covered with a 0.34 nm thick graphene monolayer, enhancing the electromagnetic field tunability and confinement.

Finally, the full structure was mounted on a 21 μm × 21 μm × 1.5 μm silicon dioxide substrate, ensuring mechanical support and terahertz optical transparency. Fabrication tolerance was assessed using 1000 Monte Carlo simulations, accounting for deviations in silver thickness (±8 nm), rectangle dimensions, ring radius (±50 nm), and −5 % graphene coverage.

The Sensor

The graphene-silver metasurface sensor was fabricated using a carefully controlled multi-step nanofabrication process. First, a high-purity silicon dioxide substrate was prepared, cleaned using a standard semiconductor approach, and dried using nitrogen gas.

A graphene monolayer, grown separately by CVD on copper foil, was transferred on the substrate using a poly(methyl methacrylate) (PMMA) support layer, copper etching with iron chloride, and subsequent removal of PMMA by acetone. The sample was then annealed in a hydrogen/argon atmosphere to enhance adhesion and remove residues.

The metasurface pattern, including circular rings, plus-shaped resonators, and rectangular components, was defined using high-resolution electron beam lithography on a PMMA-coated substrate.

After pattern development, silver was deposited through electron beam evaporation to form 50-100 nm thick resonators. A lift-off process in acetone removed excess metal and resist, yielding precisely defined silver-coated structures. Optional annealing improved metal conductivity and crystallinity.

The sensor was characterized using scanning electron microscopy (SEM) and atomic force microscopy (AFM) to assess its surface structure and geometry. Raman spectroscopy confirmed graphene quality, and terahertz time-domain spectroscopy (THz-TDS) assessed its optical performance. 

The sensor could also be integrated with microfluidics via soft lithography. Its electromagnetic behavior was modeled using plasmonic resonance, Kubo formalism, Maxwell’s equations, and coupled-mode theories.

Significance of the Study

Through parametric optimization using COMSOL Multiphysics, the sensor achieved 400 GHz/RIU sensitivity, a 5.000 RIU-1 figure of merit (FOM), and a 12.7 Q factor within a 1.334-1.355 RIU refractive index range.

The study’s simulations also demonstrated strong electric field confinement at the resonance frequency of 1.08 THz, emphasizing its detection accuracy.

A machine learning algorithm was used to improve predictive reliability, with a 0.90 coefficient of determination across different refractive indices. This result was achieved using a degree-3 polynomial regression model validated across multiple refractive index scenarios. The algorithm was supported by residual and correlation analysis, ensuring high performance and minimal overfitting.

Download your PDF now!

However, it is important to note that these results are based on simulations; experimental validation and multi-site clinical testing are proposed in the next phase of the study.

With its integrated fabrication strategy, the combination of the graphene–silver metasurface with ML-based modeling enabled fast, label-free, and highly accurate detection. This sensor outperformed conventional RT-PCR and antigen tests by offering superior sensitivity, accuracy, and potential cost-effectiveness compared to existing optical and terahertz biosensors.

The authors evaluated the sensor’s performance under environmental conditions, including temperature (15-40 °C), humidity (30-90 %), and electromagnetic interference. The results showed stable sensitivity and minimal drift, confirming its robustness for real-world settings.

Conclusion

To summarize, this study demonstrated that the proposed device could be effective as a portable, cost-effective, and novel diagnostic tool for next-generation pandemic preparedness. Future development will focus on multiplexed detection arrays, expanded respiratory pathogen panels, and regulatory readiness (FDA, CE) to move toward clinical use.

Journal Reference

Natraj, N. A., Mubarakali, A., Alagarsamy, M., H., M. Y., Dhivya, R. (2025). Next-generation COVID-19 detection using a metasurface biosensor with machine learning-enhanced refractive index sensing. Scientific Reports, 15(1), 1-20. DOI: 10.1038/s41598-025-18753-w, https://www.nature.com/articles/s41598-025-18753-w

Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.

Samudrapom Dam

Written by

Samudrapom Dam

Samudrapom Dam is a freelance scientific and business writer based in Kolkata, India. He has been writing articles related to business and scientific topics for more than one and a half years. He has extensive experience in writing about advanced technologies, information technology, machinery, metals and metal products, clean technologies, finance and banking, automotive, household products, and the aerospace industry. He is passionate about the latest developments in advanced technologies, the ways these developments can be implemented in a real-world situation, and how these developments can positively impact common people.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Dam, Samudrapom. (2025, October 10). Graphene and AI combine for rapid COVID-19 detection. AZoSensors. Retrieved on October 10, 2025 from https://www.azosensors.com/news.aspx?newsID=16647.

  • MLA

    Dam, Samudrapom. "Graphene and AI combine for rapid COVID-19 detection". AZoSensors. 10 October 2025. <https://www.azosensors.com/news.aspx?newsID=16647>.

  • Chicago

    Dam, Samudrapom. "Graphene and AI combine for rapid COVID-19 detection". AZoSensors. https://www.azosensors.com/news.aspx?newsID=16647. (accessed October 10, 2025).

  • Harvard

    Dam, Samudrapom. 2025. Graphene and AI combine for rapid COVID-19 detection. AZoSensors, viewed 10 October 2025, https://www.azosensors.com/news.aspx?newsID=16647.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit

Sign in to keep reading

We're committed to providing free access to quality science. By registering and providing insight into your preferences you're joining a community of over 1m science interested individuals and help us to provide you with insightful content whilst keeping our service free.

or

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.