Cortex's AI identifies compounds effective against ovarian cancer cells
Our team was provided with high-content imaging data, to be used as training data for our AI.
The data represented measurement of the effect of various compounds on the viability of low grade ovarian serous adenocarcinoma cells.
Our predictions were tested on withheld compounds (unpublished data hidden from us). Pr. Kaylene Simpson, Head of the Victorian Centre for Functional Genomics, Peter MacCallum Cancer Centre, Australia, provided us with the data and conducted this proof of concept.
The results showed that our predictions were as accurate as a repeat experiment in identifying compounds potentially killing cancer cells.
The above curve shows our prediction's accuracy as an ROC curve (performance at all points of the sensitivity/specificity tradeoff), compared to the accuracy obtained from a repeated experiment (in vitro baseline).