Truace
Subscribe

Smartphone AI reads Rift Valley fever strips better than the naked eye in blind tests

Study title: Smartphone AI-Enabled Lateral Flow Immunoassay Platform Using Advanced Quantum Dots for Intelligent Quantitative Diagnostics

Lateral flow immunoassays (LFIAs) are widely used in point-of-care diagnostics for their low cost and simplicity, but conventional formats rely on subjective visual readout and lack quantitative accuracy and sensitivity. Here, we present a smartphone-guided, artificial-intelligence-enhanced LFIA platform that unites a rationally engineered dual-modal nanoprobe with on-device machine learning, demonstrated for quantitative detection of Rift Valley fever virus (RVFV).

Read the study abstract

Lateral flow immunoassays (LFIAs) are widely used in point-of-care diagnostics for their low cost and simplicity, but conventional formats rely on subjective visual readout and lack quantitative accuracy and sensitivity. Here, we present a smartphone-guided, artificial-intelligence-enhanced LFIA platform that unites a rationally engineered dual-modal nanoprobe with on-device machine learning, demonstrated for quantitative detection of Rift Valley fever virus (RVFV). The platform combines three synergistic princi…

Truvace record2 min read1 sourceEvidence: Moderate · 52/100 — Benchmark or lab evaluation · 1 independent sourcePeer-reviewedTRV-2026-1156 · cite or check it
Smartphone AI-Enabled Lateral Flow Immunoassay Platform Using Advanced Quantum Dots for Intelligent Quantitative Diagnostics
Hospital Universitari Doctor Peset, València 02 by 19Tarrestnom65. CC BY-SA 4.0 · https://creativecommons.org/licenses/by-sa/4.0

In brief

By coupling dual-modal amplification and magnetic enrichment with AI-driven analysis on widely available hardware, the AQD-LFIA platform offers a scalable route toward next-generation point-of-care and at-home diagnostics.

Main points

  1. Lateral flow immunoassays (LFIAs) are widely used in point-of-care diagnostics for their low cost and simplicity, but conventional formats rely on subjective visual readout and lack quantitative accuracy and sensitivity.
  2. Here, we present a smartphone-guided, artificial-intelligence-enhanced LFIA platform that unites a rationally engineered dual-modal nanoprobe with on-device machine learning, demonstrated for quantitative detection of Rift Valley fever virus (RVFV).
  3. The platform combines three synergistic principles: advanced magnetic quantum dot nanoparticles (AQDs) that integrate a magnetic core and a quantum dot shell in a single label to provide both colorimetric and fluorescence signals while enabling magnetic preconcentration of the target antigen; multi-illumination imaging that turns an ordinary smartphone camera into a quantitative reader; and a machine-learning model that extracts features that hold up under different lighting to convert signal into concentration, removing user subjectivity and lighting sensitivity.

The gain

A smartphone AI test for Rift Valley fever virus beat naked-eye reading in blind testing, quantified the virus in serum without pretreatment, and was more sensitive than conventional gold-nanosphere strips.

The full story

Lateral flow immunoassays (LFIAs) are widely used in point-of-care diagnostics for their low cost and simplicity, but conventional formats rely on subjective visual readout and lack quantitative accuracy and sensitivity. Here, we present a smartphone-guided, artificial-intelligence-enhanced LFIA platform that unites a rationally engineered dual-modal nanoprobe with on-device machine learning, demonstrated for quantitative detection of Rift Valley fever virus (RVFV).

The platform combines three synergistic principles: advanced magnetic quantum dot nanoparticles (AQDs) that integrate a magnetic core and a quantum dot shell in a single label to provide both colorimetric and fluorescence signals while enabling magnetic preconcentration of the target antigen; multi-illumination imaging that turns an ordinary smartphone camera into a quantitative reader; and a machine-learning model that extracts features that hold up under different lighting to convert signal into concentration, removing user subjectivity and lighting sensitivity. This design markedly improves sensitivity over conventional gold-nanosphere LFIA in both modes, enables reliable quantification directly in serum without any pretreatment, and outperforms naked-eye interpretation in blind testing.

Sources

  1. Peer-reviewedACS Applied Materials & Interfaces2026-09-21

The debate