Convolutional Neural Networks for Placenta Cell Classification

The placenta is a complicated organ that plays several roles during fetal evolution. There was a little awareness regarding the connection of anatomical placental disorders with fetal biology. In this particular work, we introduce an open source mathematically traceable deep learning pipeline to examine cell-level placenta histology using neural convolution networks. Also, we have a tendency to learn deep embedded encoding makeup information that's capable of each stratifying 5 distinct cell populations and learn interclass makeup variance. We anticipate that the automation of such a pipeline to population-scale placenta histology research does have the potential to benefit our understanding of the basic cellular placental biology, substantially its role in forecasting adverse birth results. The objective of the present work is to classify the cell populations into 5 classes using the convolution neural network.