Gold Nanograss and AI Detect Drugs in Urine Within Five Minutes

Gold Nanograss Detects Drugs in Five Minutes

A single drop of raw urine can now reveal illicit drug use within five minutes through a new platform combining nanotechnology and artificial intelligence without chemical pretreatment.

Rising global drug-related crime drives an urgent need for fast, reliable point-of-testing technologies. Traditional laboratory testing provides accuracy but demands multiple preparation steps and specialized equipment, while rapid kits often suffer from insufficient volume and cross-reactivity with similar drugs.

Dr. Eun-Kyung Lim led a research team at the Bionanotechnology Research Center of the Korea Research Institute of Bioscience and Biotechnology (KRIBB) to bridge this gap. Their findings were published in ACS Measurement Science Au on September 24, 2026.

“This study demonstrates that combining nanoengineering with explainable AI algorithms can enable rapid, precise classification of illicit drugs even in complex bodily fluids,” Lim said.

Nanostructured Substrate Amplifies Faint Signals

The research team formed densely packed, grass-like nanopillars on a thin gold surface and roughened them by selective etching. When drug molecules settle on this nanostructured substrate, their characteristic vibrational light signals amplify by millions of times.

This substrate-enhanced Raman scattering approach allows even trace amounts of narcotics to be clearly identified. The team collaborated with Professor Tae Hwan Kim at Daegu Catholic University, Professor Joon Hee Lee at Seoul National University Bundang Hospital, and the AI Research Center at StoneLab.

LightGBM Artificial Intelligence Isolates Signatures

Distinguishing drug-specific signals from metabolic waste products and medication metabolites in urine has long challenged researchers. The team solved this by deploying an artificial intelligence model called LightGBM.

In tests involving 155 real clinical urine specimens, the AI-assisted platform identified drug-positive samples with approximately 80% accuracy. This success rate held steady even when testing samples from patients taking non-narcotic analgesics or multiple medications.

In vivo experiments using rat models further demonstrated that drug components could be analyzed in urine collected up to six hours after administration. The primary study abstract notes that principal component analysis allowed the successful classification of four illicit drugs during this window.

Portable Diagnostic Devices Await Further Development

The platform merges the advantages of large laboratory instruments with the speed of field test kits. Researchers plan to scale the core technology into a compact, portable measurement device suitable for forensic and medical settings.

Law enforcement agencies could utilize the tool for on-site drug screening, while hospital emergency departments could rapidly assess drug intoxication. “By further developing the platform into a portable diagnostic device for on-site use, we hope to transform drug screening and emergency diagnostics,” Lim said.

Common Questions About the New Screening Technology

How much urine does the new screening platform require?

The platform requires only a single drop of raw urine to complete the screening process.

Does the urine sample need chemical preparation before testing?

No chemical pretreatment or sample preparation steps are necessary before applying the sample to the gold nanograss substrate.

What is the accuracy rate of the AI-assisted screening platform?

In tests utilizing clinical urine specimens, the LightGBM-assisted platform identified drug-positive samples with approximately 80% accuracy.

Which academic and research institutions participated in the study?

The research was conducted by teams from the Korea Research Institute of Bioscience and Biotechnology (KRIBB), Daegu Catholic University, Seoul National University Bundang Hospital, StoneLab, and the Korea Advanced Institute of Science and Technology (KAIST).