Single Test Image-Based Automated Machine Learning System for Distinguishing between Trait and Diseased Blood Samples

We introduce a machine learning-based method for fully automated diagnosis of sickle cell disease of poor-quality unstained images of a mobile microscope. Our method is capable of distinguishing between diseased, trait (carrier), and normal samples unlike the previous methods that are limited to distinguishing the normal from the abnormal samples only. The novelty of this method comes from distinguishing the trait and the diseased samples from challenging images that have been captured directly in the field. The proposed approach contains two parts, the segmentation part followed by the classification part. We use a random forest algorithm to segment such challenging images acquitted through a mobile phone-based microscope. Then, we train two classifiers based on a random forest (RF) and a support vector machine (SVM) for classification. The results show superior performances of both of the classifiers not only for images which have been captured in the lab, but also for the ones which have been acquired in the field itself.

[1]  Antonio Criminisi,et al.  Decision Forests: A Unified Framework for Classification, Regression, Density Estimation, Manifold Learning and Semi-Supervised Learning , 2012, Found. Trends Comput. Graph. Vis..

[2]  Bernhard Schölkopf,et al.  Comparing support vector machines with Gaussian kernels to radial basis function classifiers , 1997, IEEE Trans. Signal Process..

[3]  L L Wheeless,et al.  Classification of red blood cells as normal, sickle, or other abnormal, using a single image analysis feature. , 1994, Cytometry.

[4]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[5]  Pranati Rakshit,et al.  Detection of Abnormal Findings in Human RBC in Diagnosing Sickle Cell Anaemia Using Image Processing , 2013 .

[6]  Inkyu Moon,et al.  Human red blood cell recognition enhancement with three-dimensional morphological features obtained by digital holographic imaging , 2016, Journal of biomedical optics.

[7]  M. Omair Ahmad,et al.  Tracking Biological Cells in Time-Lapse Microscopy: An Adaptive Technique Combining Motion and Topological Features , 2011, IEEE Transactions on Biomedical Engineering.

[8]  Simon Kasif,et al.  Induction of Oblique Decision Trees , 1993, IJCAI.

[9]  W. Eaton,et al.  Treating sickle cell disease by targeting HbS polymerization. , 2017, Blood.

[10]  Xianghua Xie,et al.  A multi-stage random forest classifier for phase contrast cell segmentation , 2015, 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[11]  Prof. I. A. Chintawar,et al.  Detection of Sickle Cells using Image Processing , 2016 .

[12]  K. S. Kim,et al.  Analyzing blood cell image to distinguish its abnormalities (poster session) , 2000, ACM Multimedia.

[13]  Eric R. Ziegel,et al.  The Elements of Statistical Learning , 2003, Technometrics.

[14]  Magudeeswaran Veluchamy,et al.  Feature Extraction and Classification of Blood Cells Using Artificial Neural Network , 2012 .

[15]  Erwan Scornet,et al.  A random forest guided tour , 2015, TEST.

[16]  A. Schechter,et al.  Hemoglobin S polymerization: primary determinant of the hemolytic and clinical severity of the sickling syndromes. , 1985, Blood.

[17]  A K Ray,et al.  Quantitative microscopy approach for shape‐based erythrocytes characterization in anaemia , 2013, Journal of microscopy.