Automatic quantification of living cells via a non-invasive achromatic colorimetric sensor through machine learning-assisted image analysis using a smartphone

Abstract

Several cell quantification techniques including optical and electrochemical methods have been developed. However, these techniques have certain limitations such as the requirement of considerable effort and time or insufficient feasibility and accessibility. Moreover, cell quantification is generally performed in an invasive manner, which makes performing continuous cell experiments difficult and leads to the requirement of large number of samples. In this study, a simple, rapid, and accurate in-situ cell quantification method using an achromatic colorimetric sensor system is proposed. A non-invasive achromatic colorimetric cell quantitation sensor (Ach-CQ sensor) is developed based on the subtractive color model. The color of the sensor changes from black to orange owing to the pH change occurring due to the release of acid molecules such as proton donors and lactate during cellular respiration. The Ach-CQ sensor exhibits a higher color contrast difference and effectively determines the number of cells through color changes within 20 s compared with that of a conventional pH indicator such as bromothymol blue. Moreover, machine learning-assisted, saturation-based image analysis is applied for the facile quantification of living cells to achieve automatic quantification of living cells in a non-invasive manner with a limit of detection of 51 × 104 cells in the case of HeLa cells using a smartphone. Additionally, a 3D-printed Ach-CQ sensor kit is fabricated for on-site cell quantification.