Image
|
Figure Caption
Fig. 4
Classification modeling performance.
We used RandomForest to build a classification model for 6 FRC classes: Other, OtherBrominated, PolyBrominated Diphenyl Ether (PBDE), aryl phosphate ester (APE), brominated phenol (BP), and chlorinated phosphate ester (CPE). The balanced accuracy (BA) for each of the FRC classes, and their sensitivity, specificity, and positive predictive value for the models (A) 16 parameters of 13 chemicals with the LEC for all 3 zebrafish assays and (B) 14 parameters with the BMD for the zebrafish morphology only. Both models yield a BA of 91.67%.
Acknowledgments
This image is the copyrighted work of the attributed author or publisher, and
ZFIN has permission only to display this image to its users.
Additional permissions should be obtained from the applicable author or publisher of the image.
Reprinted from Reproductive toxicology (Elmsford, N.Y.), 96, Truong, L., Marvel, S., Reif, D.M., Thomas, D., Pande, P., Dasgupta, S., Simonich, M.T., Waters, K.M., Tanguay, R.L., The multi-dimensional embryonic zebrafish platform predicts flame retardant bioactivity, 359-369, Copyright (2020) with permission from Elsevier. Full text @ Reprod. Toxicol.