Classification of fetal and adult red blood cells based on hydrodynamic deformation and deep video recognition.
Ontology highlight
ABSTRACT:
SUBMITTER: Kampen PJT
PROVIDER: S-EPMC10721708 | biostudies-literature | 2023 Dec
REPOSITORIES: biostudies-literature
Publications
Classification of fetal and adult red blood cells based on hydrodynamic deformation and deep video recognition.
Kampen Peter Johannes Tejlgaard PJT Støttrup-Als Gustav Ragnar GR Bruun-Andersen Nicklas N Secher Joachim J Høier Freja F Hansen Anne Todsen AT Dziegiel Morten Hanefeld MH Christensen Anders Nymark AN Berg-Sørensen Kirstine K
Biomedical microdevices 20231214 1
Flow based deformation cytometry has shown potential for cell classification. We demonstrate the principle with an injection moulded microfluidic chip from which we capture videos of adult and fetal red blood cells, as they are being deformed in a microfluidic chip. Using a deep neural network - SlowFast - that takes the temporal behavior into account, we are able to discriminate between the cells with high accuracy. The accuracy was larger for adult blood cells than for fetal blood cells. Howev ...[more]
Similar Datasets
Project description:Seizure semiology is a well-established method to classify epileptic seizure types, but requires a significant amount of resources as long-term Video-EEG monitoring needs to be visually analyzed. Therefore, computer vision based diagnosis support tools are a promising approach. In this article, we utilize infrared (IR) and depth (3D) videos to show the feasibility of a 24/7 novel object and action recognition based deep learning (DL) monitoring system to differentiate between epileptic seizures in frontal lobe epilepsy (FLE), temporal lobe epilepsy (TLE) and non-epileptic events. Based on the largest 3Dvideo-EEG database in the world (115 seizures/+680,000 video-frames/427GB), we achieved a promising cross-subject validation f1-score of 0.833±0.061 for the 2 class (FLE vs. TLE) and 0.763 ± 0.083 for the 3 class (FLE vs. TLE vs. non-epileptic) case, from 2 s samples, with an automated semi-specialized depth (Acc.95.65%) and Mask R-CNN (Acc.96.52%) based cropping pipeline to pre-process the videos, enabling a near-real-time seizure type detection and classification tool. Our results demonstrate the feasibility of our novel DL approach to support 24/7 epilepsy monitoring, outperforming all previously published methods.
| S-EPMC9666544 | biostudies-literature
Project description:Changes in red blood cell (RBC) morphology distribution have emerged as a quantitative biomarker for the degradation of RBC functional properties during hypothermic storage. Previously published automated methods for classifying the morphology of stored RBCs often had insufficient accuracy and relied on proprietary code and datasets, making them difficult to use in many research and clinical applications. Here we describe the development and validation of a highly accurate open-source RBC morphology classification pipeline based on ensemble deep learning (DL). The DL-enabled pipeline utilized adaptive thresholding or semantic segmentation for RBC identification, a deep ensemble of four convolutional neural networks (CNNs) to classify RBC morphology, and Kalman filtering with Hungarian assignment for tracking changes in the morphology of individual RBCs over time. The ensembled CNNs were trained and evaluated on thousands of individual RBCs from two open-access datasets previously collected to quantify the morphological heterogeneity and washing-induced shape recovery of stored RBCs. Confusion matrices and reliability diagrams demonstrated under-confidence of the constituent models and an accuracy of about 98% for the deep ensemble. Such a high accuracy allowed the CNN ensemble to uncover new insights over our previously published studies. Re-analysis of the datasets yielded much more accurate distributions of the effective diameters of stored RBCs at each stage of morphological degradation (discocyte: 7.821 ± 0.429 µm, echinocyte 1: 7.800 ± 0.581 µm, echinocyte 2: 7.304 ± 0.567 µm, echinocyte 3: 6.433 ± 0.490 µm, sphero-echinocyte: 5.963 ± 0.348 µm, spherocyte: 5.904 ± 0.292 µm, stomatocyte: 7.080 ± 0.522 µm). The effective diameter distributions were significantly different across all morphologies, with considerable effect sizes for non-neighboring classes. A combination of morphology classification with cell tracking enabled the discovery of a relatively rare and previously overlooked shape recovery of some sphero-echinocytes to early-stage echinocytes after washing with 1% human serum albumin solution. Finally, the datasets and code have been made freely available online to enable replication, further improvement, and adaptation of our work for other applications.
| S-EPMC9950070 | biostudies-literature
Project description:BackgroundDNA methylation is an epigenetic modification that plays an important role during mammalian development. Around birth in humans, the main site of red blood cell production moves from the fetal liver to the bone marrow. DNA methylation changes at the β-globin locus and a switch from fetal to adult hemoglobin production characterize this transition. Understanding this globin switch may improve the treatment of patients with sickle cell disease and β-thalassemia, two of the most common Mendelian diseases in the world. The goal of our study was to describe and compare the genome-wide patterns of DNA methylation in fetal and adult human erythroblasts.MethodsWe used the Illumina HumanMethylation 450 k BeadChip to measure DNA methylation at 402,819 CpGs in ex vivo-differentiated erythroblasts from 12 fetal liver and 12 bone marrow CD34+ donors.ResultsWe identified 5,937 differentially methylated CpGs that overlap with erythroid enhancers and binding sites for erythropoiesis-related transcription factors. Combining this information with genome-wide association study results, we show that erythroid enhancers define particularly promising genomic regions to identify new genetic variants associated with fetal hemoglobin (HbF) levels in humans. Many differentially methylated CpGs are located near genes with unanticipated roles in red blood cell differentiation and proliferation. For some of these new candidate genes, we confirm the correlation between DNA methylation and gene expression levels in red blood cell progenitors. We also provide evidence that DNA methylation and genetic variation at the β-globin locus independently control globin gene expression in adult erythroblasts.ConclusionsOur DNA methylome maps confirm the widespread dynamic changes in DNA methylation that occur during human erythropoiesis. These changes tend to happen near erythroid enhancers, further highlighting their importance in erythroid regulation and HbF production. Finally, DNA methylation may act independently of the transcription factor BCL11A to repress fetal hemoglobin production. This provides cues on strategies to more efficiently re-activate HbF production in sickle cell disease and β-thalassemia patients.
| S-EPMC4298057 | biostudies-literature
Project description:Sickle cell disease (SCD) is a hematological disorder leading to blood vessel occlusion accompanied by painful episodes and even death. Red blood cells (RBCs) of SCD patients have diverse shapes that reveal important biomechanical and bio-rheological characteristics, e.g. their density, fragility, adhesive properties, etc. Hence, having an objective and effective way of RBC shape quantification and classification will lead to better insights and eventual better prognosis of the disease. To this end, we have developed an automated, high-throughput, ex-vivo RBC shape classification framework that consists of three stages. First, we present an automatic hierarchical RBC extraction method to detect the RBC region (ROI) from the background, and then separate touching RBCs in the ROI images by applying an improved random walk method based on automatic seed generation. Second, we apply a mask-based RBC patch-size normalization method to normalize the variant size of segmented single RBC patches into uniform size. Third, we employ deep convolutional neural networks (CNNs) to realize RBC classification; the alternating convolution and pooling operations can deal with non-linear and complex patterns. Furthermore, we investigate the specific shape factor quantification for the classified RBC image data in order to develop a general multiscale shape analysis. We perform several experiments on raw microscopy image datasets from 8 SCD patients (over 7,000 single RBC images) through a 5-fold cross validation method both for oxygenated and deoxygenated RBCs. We demonstrate that the proposed framework can successfully classify sickle shape RBCs in an automated manner with high accuracy, and we also provide the corresponding shape factor analysis, which can be used synergistically with the CNN analysis for more robust predictions. Moreover, the trained deep CNN exhibits good performance even for a deoxygenated dataset and distinguishes the subtle differences in texture alteration inside the oxygenated and deoxygenated RBCs.
| S-EPMC5654260 | biostudies-literature
Project description:The capacity to undergo substantial deformation is a defining characteristic of the red blood cell (RBC), facilitating transit through the splenic interendothelial slits and microvasculature. Establishment of this remarkable property occurs during a process of reticulocyte maturation that begins with egress through micron-wide pores in the bone marrow and is completed within the circulation. The requirement to undertake repeated cycles of deformation necessitates that both reticulocytes and erythrocytes regulate membrane-cytoskeletal protein interactions in order to maintain cellular stability. In the absence of transcriptional activity, modulation of these interactions in RBCs is likely to be achieved primarily through specific protein posttranslational modifications, which at present remain undefined. In this study, we use high-throughput methods to define the processes that underlie the response to deformation and shear stress in both reticulocytes and erythrocytes. Through combination of a bead-based microsphiltration assay with phosphoproteomics we describe posttranslational modification of RBC proteins associated with deformation. Using microsphiltration and microfluidic biochip-based assays, we explore the effect of inhibiting kinases identified using this dataset. We demonstrate roles for GSK3 and Lyn in capillary transit and maintenance of membrane stability following deformation and show that combined inhibition of these kinases significantly decreases reticulocyte capacity to undergo repeated deformation. Finally, we derive a comprehensive and integrative phosphoproteomic dataset that provides a valuable resource for further mechanistic dissection of the molecular pathways that underlie the RBC's response to mechanical stimuli and for the study of reticulocyte maturation.
| S-EPMC6737423 | biostudies-literature
Project description:Hepatic sinusoids present complex anatomical structures such as the endothelial sieve pores and the Disse space, which govern the microscopic blood flow in the sinusoids and are associated with structural variations in liver fibrosis and cirrhosis. However, the contributions of the permeability of endothelial and collagen layers and the roughness of hepatocyte microvilli to the features of this microflow remain largely unknown. Here, an immersed boundary method coupled with a lattice Boltzmann method was adopted in an in vitro hepatic sinusoidal model, and flow field and erythrocyte deformation analyses were conducted by introducing three new source terms including permeability of the endothelial layer, resistance of hepatocyte microvilli and collagen layers, and deformation of red blood cells (RBCs). Numerical calculations indicated that alterations in endothelial permeability could significantly affect the flow velocity and flow rate distributions in hepatic sinusoids. Interestingly, a biphasic regulating pattern of shear stress occurred simultaneously on the surface of hepatocytes and the lower side of endothelium, i.e., the shear stress increased with increased thickness of hepatocyte microvilli and collagen layer when the endothelial permeability was high but decreased with the increase of the thickness at low endothelial permeability. Additionally, this specified microflow manipulates typical RBC deformation inside the sinusoid, yielding one-third of the variation of deformable index with varied endothelial permeability. These simulations not only are consistent with experimental measurements using in vitro liver sinusoidal chip but also elaborate the contributions of endothelial and collagen layer permeability and wall roughness. Thus, our results provide a basis for further characterizing this microflow and understanding its effects on cellular migration and deformation in the hepatic sinusoids.
| S-EPMC8595567 | biostudies-literature
Project description:Sickle cell disease, a genetic disorder affecting a sizeable global demographic, manifests in sickle red blood cells (sRBCs) with altered shape and biomechanics. sRBCs show heightened adhesive interactions with inflamed endothelium, triggering painful vascular occlusion events. Numerous studies employ microfluidic-assay-based monitoring tools to quantify characteristics of adhered sRBCs from high resolution channel images. The current image analysis workflow relies on detailed morphological characterization and cell counting by a specially trained worker. This is time and labor intensive, and prone to user bias artifacts. Here we establish a morphology based classification scheme to identify two naturally arising sRBC subpopulations-deformable and non-deformable sRBCs-utilizing novel visual markers that link to underlying cell biomechanical properties and hold promise for clinically relevant insights. We then set up a standardized, reproducible, and fully automated image analysis workflow designed to carry out this classification. This relies on a two part deep neural network architecture that works in tandem for segmentation of channel images and classification of adhered cells into subtypes. Network training utilized an extensive data set of images generated by the SCD BioChip, a microfluidic assay which injects clinical whole blood samples into protein-functionalized microchannels, mimicking physiological conditions in the microvasculature. Here we carried out the assay with the sub-endothelial protein laminin. The machine learning approach segmented the resulting channel images with 99.1±0.3% mean IoU on the validation set across 5 k-folds, classified detected sRBCs with 96.0±0.3% mean accuracy on the validation set across 5 k-folds, and matched trained personnel in overall characterization of whole channel images with R2 = 0.992, 0.987 and 0.834 for total, deformable and non-deformable sRBC counts respectively. Average analysis time per channel image was also improved by two orders of magnitude (∼ 2 minutes vs ∼ 2-3 hours) over manual characterization. Finally, the network results show an order of magnitude less variance in counts on repeat trials than humans. This kind of standardization is a prerequisite for the viability of any diagnostic technology, making our system suitable for affordable and high throughput disease monitoring.
| S-EPMC8659663 | biostudies-literature
Project description:The deformability of a red blood cell (RBC) is one of the most important biological parameters affecting blood flow, both in large arteries and in the microcirculation, and hence it can be used to quantify the cell state. Despite numerous studies on the mechanical properties of RBCs, including cell rigidity, much is still unknown about the relationship between deformability and the configuration of flowing cells, especially in a confined rectangular channel. Recent computer simulation techniques have successfully been used to investigate the detailed behavior of RBCs in a channel, but the dynamics of a translating RBC in a narrow rectangular microchannel have not yet been fully understood. In this study, we numerically investigated the behavior of RBCs flowing at different velocities in a narrow rectangular microchannel that mimicked a microfluidic device. The problem is characterized by the capillary number C a , which is the ratio between the fluid viscous force and the membrane elastic force. We found that confined RBCs in a narrow rectangular microchannel maintained a nearly unchanged biconcave shape at low C a , then assumed an asymmetrical slipper shape at moderate C a , and finally attained a symmetrical parachute shape at high C a . Once a RBC deformed into one of these shapes, it was maintained as the final stable configurations. Since the slipper shape was only found at moderate C a , measuring configurations of flowing cells will be helpful to quantify the cell state.
| S-EPMC6470855 | biostudies-literature
Project description:Small RNAs isolated from RBCs were size fractionated by gel electrophoresis and used for the creation of 6 libraries. For the library from healthy children (library 1), RNA from 4 individuals was pooled. Libraries were multiplexed and analyzed on a 454 sequencing platform 26 yielding 569,621 sequence reads after demultiplexing into the 6 libraries
2012-07-01 | GSE27250 | GEO
Project description:Vaso-occlusive crisis, a common painful complication of sickle cell disease, is a complex process triggered by intercellular adhesive interactions among blood cells and the endothelium in all human organs (e.g., the oxygen-rich lung as well as hypoxic systems such as liver and kidneys). We present a combined experimental-computational study to quantify the adhesive characteristics of sickle mature erythrocytes (SMEs) and irreversibly sickled cells (ISCs) under flow conditions mimicking those in postcapillary venules. We employed an in vitro microfluidic cell adherence assay, which is coated uniformly with fibronectin. We investigated the adhesion dynamics of SMEs and ISCs in pulsatile flow under well-controlled hypoxic conditions, inferring the cell adhesion strength by increasing the flow rate (or wall shear stress (WSS)) until the onset of cell detachment. In parallel, we performed simulations of individual SMEs and ISCs under shear. We introduced two metrics to quantify the adhesion process, the cell aspect ratio (AR) as a function of WSS and its rate of change (the dynamic deformability index). We found that the AR of SMEs decreases significantly with the increase of WSS, consistent between the experiments and simulations. In contrast, the AR of ISCs remains constant in time and independent of the flow rate. The critical WSS value for detaching a single SME in oxygenated state is in the range of 3.9-5.5 Pa depending on the number of adhesion sites; the critical WSS value for ISCs is lower than that of SMEs. Our simulations show that the critical WSS value for SMEs in deoxygenated state is above 6.2 Pa (multiple adhesion sites), which is greater than their oxygenated counterparts. We investigated the effect of cell shear modulus on the detachment process; we found that for the same cell adhesion spring constant, the higher shear modulus leads to an earlier cell detachment from the functionalized surface. These findings may aid in the understanding of individual roles of sickle cell types in sickle cell disease vaso-occlusion.
| S-EPMC6350013 | biostudies-literature