Classification of red blood cell shapes in flow using outlier tolerant machine learning

The manual evaluation, classification and counting of biological objects demands for an enormous expenditure of time and subjective human input may be a source of error. Investigating the shape of red blood cells (RBCs) in microcapillary Poiseuille flow, we overcome this drawback by introducing a convolutional neural regression network for an automatic, outlier tolerant shape classification. From our experiments we expect two stable geometries: the so-called ‘slipper’ and ‘croissant’ shapes depending on the prevailing flow conditions and the cell-intrinsic parameters. Whereas croissants mostly occur at low shear rates, slippers evolve at higher flow velocities. With our method, we are able to find the transition point between both ‘phases’ of stable shapes which is of high interest to ensuing theoretical studies and numerical simulations. Using statistically based thresholds, from our data, we obtain so-called phase diagrams which are compared to manual evaluations. Prospectively, our concept allows us to perform objective analyses of measurements for a variety of flow conditions and to receive comparable results. Moreover, the proposed procedure enables unbiased studies on the influence of drugs on flow properties of single RBCs and the resulting macroscopic change of the flow behavior of whole blood.

[1]  J. Freund,et al.  Cellular flow in a small blood vessel , 2010, Journal of Fluid Mechanics.

[2]  Hong Zhao,et al.  A deep convolutional neural network for classification of red blood cells in sickle cell anemia , 2017, PLoS Comput. Biol..

[3]  A. Patapoutian,et al.  Piezo1 links mechanical forces to red blood cell volume , 2015, eLife.

[4]  R. Skalak,et al.  Deformation of Red Blood Cells in Capillaries , 1969, Science.

[5]  Geoffrey E. Hinton,et al.  Rectified Linear Units Improve Restricted Boltzmann Machines , 2010, ICML.

[6]  H. Noguchi,et al.  Shape transitions of fluid vesicles and red blood cells in capillary flows. , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[7]  Masakazu Matsugu,et al.  Subject independent facial expression recognition with robust face detection using a convolutional neural network , 2003, Neural Networks.

[8]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  A. Benyoussef,et al.  On the problem of slipper shapes of red blood cells in the microvasculature. , 2013, Microvascular research.

[10]  N. Bressloff,et al.  Blood flow in microvascular networks , 2009 .

[11]  C. Verschoor,et al.  The Molecular Structure of Human Red Blood Cell Membranes from Highly Oriented, Solid Supported Multi-Lamellar Membranes , 2017, Scientific Reports.

[12]  R. Skalak,et al.  Flow of axisymmetric red blood cells in narrow capillaries , 1986, Journal of Fluid Mechanics.

[13]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[14]  K S Sakariassen,et al.  Shear-induced platelet activation and platelet microparticle formation at blood flow conditions as in arteries with a severe stenosis. , 1997, Arteriosclerosis, thrombosis, and vascular biology.

[15]  Nicolas Le Roux,et al.  Ask the locals: Multi-way local pooling for image recognition , 2011, 2011 International Conference on Computer Vision.

[16]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

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

[18]  L. Kaestner,et al.  Red Blood Cell Passage of Small Capillaries Is Associated with Transient Ca2+-mediated Adaptations , 2017, Front. Physiol..

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

[20]  L. Kaestner,et al.  3D tomography of cells in micro-channels , 2017, 1709.05686.

[21]  S. Guido,et al.  Microconfined flow behavior of red blood cells in vitro , 2009 .

[22]  Alexander Kihm,et al.  Numerical-experimental observation of shape bistability of red blood cells flowing in a microchannel. , 2017, Soft matter.

[23]  J. Freund.  The flow of red blood cells through a narrow spleen-like slit , 2013 .

[24]  Radford M. Neal.  Pattern Recognition and Machine Learning , 2007, Technometrics.

[25]  Sven Behnke,et al.  Evaluation of Pooling Operations in Convolutional Architectures for Object Recognition , 2010, ICANN.

[26]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[27]  Ignacio Pagonabarraga,et al.  Rheology of red blood cells under flow in highly confined microchannels: I. effect of elasticity. , 2014, Soft matter.

[28]  M. Kubát.  An Introduction to Machine Learning , 2017, Springer International Publishing.

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

[30]  W. Helfrich.  Elastic Properties of Lipid Bilayers: Theory and Possible Experiments , 1973, Zeitschrift fur Naturforschung. Teil C: Biochemie, Biophysik, Biologie, Virologie.

[31]  O. Baskurt,et al.  New guidelines for hemorheological laboratory techniques. , 2009, Clinical hemorheology and microcirculation.

[32]  Ning Qian,et al.  On the momentum term in gradient descent learning algorithms , 1999, Neural Networks.