The application of convolutional neural network to stem cell biology

Abstract

Induced pluripotent stem cells (iPSC) are one the most prominent innovations of medical research in the last few decades. iPSCs can be easily generated from human somatic cells and have several potential uses in regenerative medicine, disease modeling, drug screening, and precision medicine. However, further innovation is still required to realize their full potential. Machine learning is an algorithm that learns from large datasets for pattern formation and classification. Deep learning, a form of machine learning, uses a multilayered neural network that mimics human neural circuit structure. Deep neural networks can automatically extract features from an image, although classical machine learning methods still require feature extraction by a human expert. Deep learning technology has developed recently; in particular, the accuracy of an image classification task by using a convolutional neural network (CNN) has exceeded that of humans since 2015. CNN is now used to address several tasks including medical issues. We believe that CNN would also have a great impact on the research of stem cell biology. iPSCs are utilized after their differentiation to specific cells, which are characterized by molecular techniques such as immunostaining or lineage tracing. Each cell shows a characteristic morphology; thus, a morphology-based identification system of cell type by CNN would be an alternative technique. The development of CNN enables the automation of identifying cell types from phase contrast microscope images without molecular labeling, which will be applied to several researches and medical science. Image classification is a strong field among deep learning tasks, and several medical tasks will be solved by deep learning-based programs in the future.

Keywords: Artificial intelligence; Deep learning; Endothelial cell; Image recognition; Induced pluripotent stem cell; Machine learning; Stem cell.

Fig. 1

a Structure of simple perceptron. x₁, x₂, x₃ … xi represent the output signals of each upstream neuron and each signal is multiplied by each weight: w₁, w₂, w₃ …wi. Multiplied signals, which comprise the input signal, are summed and calculated by activation function. y is the output of the perceptron. b Neural network consisting of multiple layers of perceptrons converts input signal to final output signal, which is called the predictive value. Predictive value is compared with the objective value, and error is calculated by loss function. Each neuron signal weight is adjusted to minimize the error with the optimizer method, which is based on the backward propagation method

Fig. 2

Concept of a morphology-based cell identification system. Each cell shows a unique morphology. The machine can identify the cell type solely from phase contrast images, which humans cannot do

Fig. 3

Strategy to identify iPSC-ECs by a deep neural network. iPSCs are differentiated to endothelial cells, and phase contrast microscope images are captured. Input blocks are cropped from phase contrast images and inputted into the neural network. The neural network predicts whether target blocks are “unstained” or “stained.” Target blocks that include the target cells to be examined are cropped from binary images of CD31-immunostaining to generate correct answers, which are determined by the white pixel ratio of target blocks. Predictions are compared with the correct answers, and weights of the network are adjusted automatically to increase the predictive value of the deep neural network