There is a need for fast identification of potential treatments for COVID-19 in human cells. We propose an approach that enables cellular image-based analysis in a supervised manner with no domain knowledge required using the Regression Plane (RP) concept (Szkalisity, Abel, et al.). RP is a methodology for modeling and analyzing biological processes using 2-dimensional (2D) regression. Our approach is based on the ideas presented in the original paper by Szkalisity, Abel, et al., however, we perform naïve modeling of the problem space. The goals of this study are to identify potential treatments for COVID-19 and analyze the effect of the treatment’s dosage on the treated cells. The suggested approach is used to reanalyze a published phenotypic drug discovery dataset to achieve these goals. An additional goal of this study is to examine the robustness of the proposed method by analyzing how different modeling affects the biological conclusions drawn.