Automatic Cancer Cell Taxonomy Using an Ensemble of Deep Neural Networks

Abstract

Microscopic image-based analysis has been intensively performed for pathological studies and diagnosis of diseases. However, mis-authentication of cell lines due to misjudgments by pathologists has been recognized as a serious problem. To address this problem, we propose a deep-learning-based approach for the automatic taxonomy of cancer cell types. A total of 889 bright-field microscopic images of four cancer cell lines were acquired using a benchtop microscope. Individual cells were further segmented and augmented to increase the image dataset. Afterward, deep transfer learning was adopted to accelerate the classification of cancer types. Experiments revealed that the deep-learning-based methods outperformed traditional machine-learning-based methods. Moreover, the Wilcoxon signed-rank test showed that deep ensemble approaches outperformed individual deep-learning-based models (p < 0.001) and were in effect to achieve the classification accuracy up to 97.735%. Additional investigation with the Wilcoxon signed-rank test was conducted to consider various network design choices, such as the type of optimizer, type of learning rate scheduler, degree of fine-tuning, and use of data augmentation. Finally, it was found that the using data augmentation and updating all the weights of a network during fine-tuning improve the overall performance of individual convolutional neural network models.

Keywords: cancer cell taxonomy; convolutional neural network; deep learning; ensemble approach.

Figure 1

Workflow of the proposed approach.

Figure 2

Image preprocessing step. (a) Captured microscope image, (b) grayscale image, (c) noise removed image, (d) identified cell contour, (e) segmented image patches.

Figure 3

Example of data augmentation: (A) original, (B) rotation, (C) translation, (D) vertical flip, (E) all.

Figure 4

Degree of fine-tuning.

Figure 5

Overview of ensemble approaches.

Figure 6

Classification accuracy per epoch: (A) training accuracy, (B) validation accuracy.

Figure 7

Loss per epoch: (A) training loss, (B) validation loss.

Figure 8

Network design choices (the statistical significance is represented using * (p < 0.05), ** (p < 0.01), and *** (p < 0.001)): (A) optimizer, (B) data augmentation, (C) learning rate scheduler, (D) degree of fine-tuning.

Figure 9

Performance change according to the ensemble configuration: (A) single-arch ensemble, (B) multi-arch ensemble.