Development of Smartphone-Integrated Dual-mode Biosensor for Enhanced Lateral Flow Sensitivity Using Machine Learning

<p dir="ltr">The research presents the development of smartphone-integrated, dual-mode biosensor designed to enhance the sensitivity of commercially available lateral flow assays (LFAs) for foodborne pathogen detection. Conventional LFAs are widely used for rapid diagnostics but often suffer from limited sensitivity and qualitative output. To address these limitations, a portable device was developed using a Raspberry Pi platform to integrate optical and thermal sensing modalities within a unified system. The colorimetric mode quantifies line intensity changes on the test strip, while the photothermal mode analyzes speckle dynamics induced by laser heating of gold nanoparticles. Both sensing modes were paired with machine learning algorithms to enable quantitative prediction of pathogen concentration. Experiments using <i>Salmonella</i> test strips demonstrated that the dual-mode approach improved the limit of detection by one order of magnitude compared to traditional visual interpretation. The system's compact design, real-time data acquisition, and smartphone-based user interface highlight its potential for field-deployable, quantitative biosensing applications in food safety and other diagnostics. </p>