Figure 3 | Scientific Reports

Figure 3

From: Machine learning-assisted high-content imaging analysis of 3D MCF7 microtissues for estrogenic effect prediction

Figure 3

2D image profiling and differential image feature analysis. (A) Data matrix structure of the original and re-grouped setting of E2 and PPT concentration. The original data went through an optimal normalization selection process, and the data were normalized using the top-ranking method and used for the PCA plot. (B) After re-grouping, multiple normalization methods achieved minimal performance, and the top-ranking method was used to normalize the data for the PCA plot. (C) Violin plot of the full two ranking differential image features in response to E2 treatment. (D) PCA analysis for regrouped E2 dataset using selected differential image features. (E)Violin plot of the top two ranking differential image features in response to PPT treatment. (F) PCA analysis for regrouped PPT dataset using selected differential image features.

Back to article page