AI-Driven Improvements in Electrochemical Biosensors for Effective Pathogen Detection at Point-of-Care

The rapid and accurate detection of pathogens is vital for effective disease management and control. This paper introduces a novel framework for integrating artificial intelligence (AI) into electrochemical biosensors for pathogen detection. Real-world samples often present unwanted noise in the signal, particularly when utilizing portable point-of-care devices. To overcome this challenge, a framework using AI for noise reduction from a portable potentiostat is proposed in this work. This approach involves employing a denoising autoencoder (DAE) to effectively remove noise from the electrochemical signals generated from a portable potentiostat by utilizing training datasets generated from benchtop potentiostat for training the DAE, bringing the performance of portable devices on par with their benchtop counterparts. This enhancement is crucial for point-of-care applications where environmental and operational factors often compromise data quality. Smartphones are often used as interfaces for portable electrochemical devices, the proposed framework can leverage the computational capabilities of smartphones to run the DAE model for processing electrochemical signals in real-time, thus making it compatible with fully point-of-care solution. The proposed system has been validated using COVID-19 and dengue DPV data, demonstrating its potential as a powerful tool in the rapid and accurate detection of SARS-CoV-2 and other pathogens. The integration of AI into electrochemical biosensing offers a more reliable and accessible option for healthcare professionals and researchers.