SARS-CoV-2 Result Interpretation based on Image Analysis of Lateral Flow Devices

TL;DR

This paper introduces an automated image analysis method using computer vision and machine learning to interpret SARS-CoV-2 lateral flow device results, reducing human error and bias.

Contribution

It presents a novel automated approach for interpreting LFD results based on image analysis, improving accuracy and efficiency over manual reading.

Findings

Accurately classifies LFD results as positive, negative, or inconclusive

Reduces human involvement and perception bias in test interpretation

Demonstrates effectiveness of machine learning in medical image analysis

Abstract

The widely used gene quantisation technique, Lateral Flow Device (LFD), is now commonly used to detect the presence of SARS-CoV-2. It is enabling the control and prevention of the spread of the virus. Depending on the viral load, LFD have different sensitivity and self-test for normal user present additional challenge to interpret the result. With the evolution of machine learning algorithms, image processing and analysis has seen unprecedented growth. In this interdisciplinary study, we employ novel image analysis methods of computer vision and machine learning field to study visual features of the control region of LFD. Here, we automatically derive results for any image containing LFD into positive, negative or inconclusive. This will reduce the burden of human involvement of health workers and perception bias.