AIDeveloper: Deep Learning Image Classification in Life Science and Beyond

Publications on artificial intelligence (AI)-based image analysis have increased drastically in recent years. However, all applications use individual solutions highly specialized for a particular task. Here, we present an easy-to-use, adaptable, open source software, called AIDeveloper (AID) to train neural nets (NN) for image classification without the need for programming. The software provides a variety of NN-architectures that can be simply selected for training. AID allows the user to apply trained models on new data, obtain metrics for classification performance, and export final models to different formats. The working principles of AID are first illustrated by training a convolutional neural net (CNN) on a large dataset consisting of images of different objects (CIFAR-10). We further explore the potential of AID by training a model to distinguish areas of differentiated and non-differentiated mesenchymal stem cells (MSCs) in culture. Additionally, we compare a conventional clinical whole blood cell count with a whole blood cell count performed by an NN-trained, using a dataset of more than 1.2 million images obtained by real-time deformability cytometry, delivering comparable results. Finally, we demonstrate how AID can be used for label-free classification of B- and T-cells derived from human blood, which currently requires costly and time-consuming sample preparation. Thus, AID can empower anyone to develop, train, and apply NNs for image classification. Moreover, models can be generated by non-programmers, exported, and used on different devices, which allows for an interdisciplinary use.

[1]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[2]  John A. Swets,et al.  Signal Detection Theory and ROC Analysis in Psychology and Diagnostics: Collected Papers , 1996 .

[3]  A. Ferrante,et al.  The Application of Dextran Sedimentation as an Initial Step in Neutrophil Purification Promotes Their Stimulation, due to the Presence of Monocytes , 2017, Journal of immunology research.

[4]  J. L. de la Pompa,et al.  A novel source of arterial valve cells linked to bicuspid aortic valve without raphe in mice , 2018, eLife.

[5]  Chao Yang,et al.  A Survey on Deep Transfer Learning , 2018, ICANN.

[6]  Polina Golland,et al.  CellProfiler Analyst: data exploration and analysis software for complex image-based screens , 2008, BMC Bioinformatics.

[7]  Alexei A. Efros,et al.  Image-to-Image Translation with Conditional Adversarial Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Luca Maria Gambardella,et al.  Mitosis Detection in Breast Cancer Histology Images with Deep Neural Networks , 2013, MICCAI.

[9]  Rob Fergus,et al.  Visualizing and Understanding Convolutional Networks , 2013, ECCV.

[10]  Yoshua Bengio,et al.  How transferable are features in deep neural networks? , 2014, NIPS.

[11]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[12]  Johannes E. Schindelin,et al.  Fiji: an open-source platform for biological-image analysis , 2012, Nature Methods.

[13]  YongKeun Park,et al.  Identification of non-activated lymphocytes using three-dimensional refractive index tomography and machine learning , 2017, Scientific Reports.

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

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

[16]  Alex Krizhevsky,et al.  Learning Multiple Layers of Features from Tiny Images , 2009 .

[17]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

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

[19]  J. Galipeau,et al.  Mesenchymal Stromal Cells: Clinical Challenges and Therapeutic Opportunities. , 2018, Cell stem cell.

[20]  Philipp Rosendahl,et al.  Using real-time fluorescence and deformability cytometry and deep learning to transfer molecular specificity to label-free sorting , 2019, bioRxiv.

[21]  Gerhard Ehninger,et al.  Mesenchymal Stem Cells Can Be Differentiated Into Endothelial Cells In Vitro , 2004, Stem cells.

[22]  L. Menschner,et al.  Detection of human disease conditions by single-cell morpho-rheological phenotyping of blood , 2018, eLife.

[23]  U. Keyser,et al.  Real-time deformability cytometry: on-the-fly cell mechanical phenotyping , 2015, Nature Methods.

[24]  Luca Maria Gambardella,et al.  A comparison of algorithms and humans for mitosis detection , 2014, 2014 IEEE 11th International Symposium on Biomedical Imaging (ISBI).

[25]  Thomas Podgorski,et al.  The plasma protein fibrinogen stabilizes clusters of red blood cells in microcapillary flows , 2014, Scientific Reports.

[26]  Roberto Cipolla,et al.  SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[27]  G. Lip,et al.  Fibrinogen: biochemistry, epidemiology and determinants. , 2003, QJM : monthly journal of the Association of Physicians.

[28]  Qiang Xu,et al.  nuScenes: A Multimodal Dataset for Autonomous Driving , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[29]  Ullrich Köthe,et al.  Ilastik: Interactive learning and segmentation toolkit , 2011, 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[30]  Jonghee Yoon,et al.  Label-Free Identification of Lymphocyte Subtypes Using Three-Dimensional Quantitative Phase Imaging and Machine Learning. , 2018, Journal of visualized experiments : JoVE.

[31]  Oliver Otto,et al.  Real-time fluorescence and deformability cytometry , 2018, Nature Methods.

[32]  A. Prystupa,et al.  Erythrocyte sedimentation rate – an old marker with new applications , 2011 .

[33]  Sebastian Thrun,et al.  Dermatologist-level classification of skin cancer with deep neural networks , 2017, Nature.

[34]  Fumihito Arai,et al.  Intelligent Image-Activated Cell Sorting , 2018, Cell.

[35]  Olaf Wolkenhauer,et al.  Label‐Free Identification of White Blood Cells Using Machine Learning , 2019, Cytometry. Part A : the journal of the International Society for Analytical Cytology.

[36]  Fabian J Theis,et al.  Prospective identification of hematopoietic lineage choice by deep learning , 2017, Nature Methods.

[37]  Roland Vollgraf,et al.  Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms , 2017, ArXiv.

[38]  A. Bøyum,et al.  Separation of blood leucocytes, granulocytes and lymphocytes. , 2008, Tissue antigens.

[39]  Kevin W Eliceiri,et al.  NIH Image to ImageJ: 25 years of image analysis , 2012, Nature Methods.

[40]  Nir Shavit,et al.  Deep Learning is Robust to Massive Label Noise , 2017, ArXiv.

[41]  S. Gerson,et al.  Phenotypic and functional comparison of cultures of marrow‐derived mesenchymal stem cells (MSCs) and stromal cells , 1998, Journal of cellular physiology.

[42]  M. Herbig,et al.  Reliable isolation of human mesenchymal stromal cells from bone marrow biopsy specimens in patients after allogeneic hematopoietic cell transplantation. , 2019, Cytotherapy.