Label-Free White Blood Cell Classification Using Refractive Index Tomography and Deep Learning

Objective and Impact Statement.We propose a rapid and accurate blood cell identification method exploiting deep learning and label-free refractive index(RI)tomography.Our computational approach that fully utilizes tomographic information of bone marrow(BM)white blood cell(WBC)enables us to not only classify the blood cells with deep learning but also quantitatively study their morphological and biochemical properties for hematology research.Introduction.Conventional methods for examining blood cells,such as blood smear analysis by medical professionals and fluorescence-activated cell sorting,require significant time,costs,and domain knowledge that could affect test results.While label-free imaging techniques that use a specimen’s intrinsic contrast(e.g.,multiphoton and Raman microscopy)have been used to characterize blood cells,their imaging procedures and instrumentations are relatively time-consuming and complex.Methods.The RI tomograms of the BM WBCs are acquired via Mach-Zehnder interferometer-based tomographic microscope and classified by a 3D convolutional neural network.We test our deep learning classifier for the four types of bone marrow WBC collected from healthy donors(n=10):monocyte,myelocyte,B lymphocyte,and T lymphocyte.The quantitative parameters of WBC are directly obtained from the tomograms.Results.Our results show>99%accuracy for the binary classification of myeloids and lymphoids and>96%accuracy for the four-type classification of B and T lymphocytes,monocyte,and myelocytes.The feature learning capability of our approach is visualized via an unsupervised dimension reduction technique.Conclusion.We envision that the proposed cell classification framework can be easily integrated into existing blood cell investigation workflows,providing cost-effective and rapid diagnosis for hematologic malignancy.

Quantum dots-labeled polymeric scaffolds for in vivo tracking of degradation and tissue formation

The inevitable gap between in vitro and in vivo degradation rate of biomaterials has been a challenging factor in the optimal designing of scaffold’s degradation to be balanced with new tissue formation.To enable non-/minimum-invasive tracking of in vivo scaffold degradation,chemical modifications have been applied to label polymers with fluorescent dyes.However,the previous approaches may have limited expandability due to complicated synthesis processes.Here,we introduce a simple and efficient method to fluorescence labeling of polymeric scaffolds via blending with near-infrared(NIR)quantum dots(QDs),semiconductor nanocrystals with superior optical properties.QDs-labeled,3D-printed PCL scaffolds showed promising efficiency and reliability in quantitative measurement of degradation using a custom-built fiber-optic imaging modality.Furthermore,QDs-PCL scaffolds showed neither cytotoxicity nor secondary labeling of adjacent cells.QDs-PCL scaffolds also supported the engineering of fibrous,cartilaginous,and osteogenic tissues from mesenchymal stem/progenitor cells(MSCs).In addition,QDs-PCL enabled a distinction between newly forming tissue and the remaining mass of scaffolds through multi-channel imaging.Thus,our findings suggest a simple and efficient QDs-labeling of PCL scaffolds and minimally invasive imaging modality that shows significant potential to enable in vivo tracking of scaffold degradation as well as new tissue formation.

Preface

It is our great pleasure to edit this special section of the Journal of Computer Science and Technology （JCST）. The database field has experienced a rapid growth with increasing of data. Therefore, novel technology for covering emerging databases such as network or graph analysis, spatial or temporal data analysis, search, recommendation, and data mining is required. The goal of the section is to provide state-of-the-art research issues, challenges, new technologies, and solutions of emerging databases. This section publishes seven interesting articles related to query processing, trajectory data reduction, botnet evolution, recommendation system, bielustering, and protein structure alignment. The articles are summarized as follows.