Journal: Biomaterials
Article Title: Deep learning identification of stiffness markers in breast cancer
doi: 10.1016/j.biomaterials.2022.121540
Figure Lengend Snippet: Young’s modulus (global stiffness) characterization and composition analysis of breast tissue. a, Heatmap including (columns) 15 tissue samples from 6 patients (P#) clustered using Euclidean distance with complete linkage by (rows) related features. Each parameter is normalized using a z score. The values within each feature are color coded by low to high. The heatmap key on the left denotes the following color-coded parameters of each feature: cell component, extracellular matrix (ECM) component, pathologic feature, or mechanical measurement. b, Univariate analysis comparing Young’s modulus (global stiffness; kPa) to the percent composition of cell component class: (i) blood vessels, (ii) tumor cells; c, extracellular matrix combined; d, (i) straight collagen and (ii) fibrotic tissue; e, straight collagen from patients who received neoadjuvant chemotherapy; and f, percent breast density. g, Highest correlated pair of tissue composition classes with Young’s Modulus. The Pearson Correlation (r), p-value, r2 value, and error is listed at the top of plots b-e and g. One-way ANOVA was used to perform statistics in f. h, Table of top five correlated tissue composition pairs from bivariate analysis using normal distribution and identity link using MATLAB’s glmfit and glmval functions. Rank ordered by correlation. The error is the fit-error. i, Plot of Spearman Correlation (ρs) versus the p-value for all cellular and extracellular classes versus straight collagen. Values below the dashed line where p = 0.05 are significant. j, Plots showing the monotonic relationship between straight collagen and (i) blood vessels, (ii) tumor cells, (iii) wavy collagen, and (iv) fibrotic tissue. Plots in j show the r2 value and root mean squared error (RMSE) at the top of the plot. Plots with square data points represent luminal A patients who have not received chemotherapy. Plots with circles represent patients who received neoadjuvant chemotherapy. Each data point is color coded by patient. The lines denote the best fit trend line. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Article Snippet: The Pearson Correlation (r), p-value, r2 value, and error is listed at the top of plots b-e and g. One-way ANOVA was used to perform statistics in f. h, Table of top five correlated tissue composition pairs from bivariate analysis using normal distribution and identity link using MATLAB’s glmfit and glmval functions.
Techniques:
MathWorks Inc is a verified supplier 
