Automated Counting of Cancer Cells by Ensembling Deep Features

High-content and high-throughput digital microscopes have generated large image sets in biological experiments and clinical practice. Automatic image analysis techniques, such as cell counting, are in high demand. Here, cell counting was treated as a regression problem using image features (phenotypes) extracted by deep learning models. Three deep convolutional neural network models were developed to regress image features to their cell counts in an end-to-end way. Theoretically, ensembling imaging phenotypes should have better representative ability than a single type of imaging phenotype. We implemented this idea by integrating two types of imaging phenotypes (dot density map and foreground mask) extracted by two autoencoders and regressing the ensembled imaging phenotypes to cell counts afterwards. Two publicly available datasets with synthetic microscopic images were used to train and test the proposed models. Root mean square error, mean absolute error, mean absolute percent error, and Pearson correlation were applied to evaluate the models’ performance. The well-trained models were also applied to predict the cancer cell counts of real microscopic images acquired in a biological experiment to evaluate the roles of two colorectal-cancer-related genes. The proposed model by ensembling deep imaging features showed better performance in terms of smaller errors and larger correlations than those based on a single type of imaging feature. Overall, all models’ predictions showed a high correlation with the true cell counts. The ensembling-based model integrated high-level imaging phenotypes to improve the estimation of cell counts from high-content and high-throughput microscopic images.

[1]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[2]  Pietro Perona,et al.  Pedestrian Detection: An Evaluation of the State of the Art , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Sajith Kecheril Sadanandan Deep Neural Networks and Image Analysis for Quantitative Microscopy , 2017 .

[4]  Shaogang Gong,et al.  Crowd Counting and Profiling: Methodology and Evaluation , 2013, Modeling, Simulation and Visual Analysis of Crowds.

[5]  Andrew L. Maas Rectifier Nonlinearities Improve Neural Network Acoustic Models , 2013 .

[6]  Vijay S. Pande,et al.  Using Deep Learning for Segmentation and Counting within Microscopy Data , 2018, ArXiv.

[7]  Pascal Vincent,et al.  Representation Learning: A Review and New Perspectives , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  P J Sjöström,et al.  Artificial neural network-aided image analysis system for cell counting. , 1999, Cytometry.

[9]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[10]  François Chollet,et al.  Keras: The Python Deep Learning library , 2018 .

[11]  Qian Liu,et al.  A Novel Convolutional Regression Network for Cell Counting , 2019, 2019 IEEE 7th International Conference on Bioinformatics and Computational Biology ( ICBCB).

[12]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[13]  Dinesh Manocha,et al.  Modeling, Simulation and Visual Analysis of Crowds , 2013, The International Series in Video Computing.

[14]  Leo Breiman,et al.  Stacked regressions , 2004, Machine Learning.

[15]  Nuno Vasconcelos,et al.  Counting People With Low-Level Features and Bayesian Regression , 2012, IEEE Transactions on Image Processing.

[16]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[17]  David H. Wolpert,et al.  Stacked generalization , 1992, Neural Networks.

[18]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[19]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[20]  Hod Lipson,et al.  Understanding Neural Networks Through Deep Visualization , 2015, ArXiv.

[21]  Pekka Ruusuvuori,et al.  Computational Framework for Simulating Fluorescence Microscope Images With Cell Populations , 2007, IEEE Transactions on Medical Imaging.

[22]  J. Ellenberg,et al.  High-throughput fluorescence microscopy for systems biology , 2006, Nature Reviews Molecular Cell Biology.

[23]  Andrew Zisserman,et al.  Microscopy cell counting with fully convolutional regression networks , 2015 .

[24]  Xiaogang Wang,et al.  Cross-scene crowd counting via deep convolutional neural networks , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[25]  Hans Clevers,et al.  Organoids in cancer research , 2018, Nature Reviews Cancer.

[26]  D. Opitz,et al.  Popular Ensemble Methods: An Empirical Study , 1999, J. Artif. Intell. Res..

[27]  Antony Orth,et al.  Microscopy, Meet Big Data. , 2017, Cell systems.

[28]  Lior Rokach,et al.  Ensemble-based classifiers , 2010, Artificial Intelligence Review.

[29]  Andrew Zisserman,et al.  Learning To Count Objects in Images , 2010, NIPS.

[30]  R. Polikar,et al.  Ensemble based systems in decision making , 2006, IEEE Circuits and Systems Magazine.

[31]  Yutaka Suzuki,et al.  Combined Mutation of Apc, Kras, and Tgfbr2 Effectively Drives Metastasis of Intestinal Cancer. , 2018, Cancer research.

[32]  Anne E Carpenter,et al.  Annotated high-throughput microscopy image sets for validation , 2012, Nature Methods.

[33]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.