Add to Favourites
To login click here
This article presents SMARTAI-LFA, a deep learning-assisted smartphone-based LFA that can efficiently and accurately provide output results. It is designed with a two-step CNN model (object finding and classification) and is trained with SARS-CoV-2 nucleocapsid protein spiked into phosphate buffered saline (PBS) buffer. The algorithm is further retrained with additional clinical data to strengthen the model and acquire excellent specificity and accuracy for testing clinical data. A blind test of untrained individuals, human experts, and SMARTAI-LFA demonstrates the feasibility of the cradle-free sample-to-answer platform with digitalized real-time connectivity.
