Automatic detection and characterization of quantitative phase images of thalassemic red blood cells using a mask region-based convolutional neural network

Advancing Laboratory Medicine Practice With Machine Learning: Swift yet Exact

Machine learning (ML) is currently being widely studied and applied in data analysis and prediction in various fields, including laboratory medicine. To comprehensively evaluate the application of ML in laboratory medicine, we reviewed the literature on ML applications in laboratory medicine published between February 2014 and March 2024. A PubMed search using a search string yielded 779 articles on the topic, among which 144 articles were selected for this review. These articles were analyzed to extract and categorize related fields within laboratory medicine, research objectives, specimen types, data types, ML models, evaluation metrics, and sample sizes. Sankey diagrams and pie charts were used to illustrate the relationships between categories and the proportions within each category. We found that most studies involving the application of ML in laboratory medicine were designed to improve efficiency through automation or expand the roles of clinical laboratories. The most common ML models used are convolutional neural networks, multilayer perceptrons, and tree-based models, which are primarily selected based on the type of input data. Our findings suggest that, as the technology evolves, ML will rise in prominence in laboratory medicine as a tool for expanding research activities. Nonetheless, expertise in ML applications should be improved to effectively utilize this technology.

Biophysical profiling of red blood cells from thin-film blood smears using deep learning

Microscopic inspection of thin-film blood smears is widely used to identify red blood cell (RBC) pathologies, including malaria parasitism and hemoglobinopathies, such as sickle cell disease and thalassemia. Emerging research indicates that non-pathologic changes in RBCs can also be detected in images, such as deformability and morphological changes resulting from the storage lesion. In transfusion medicine, cell deformability is a potential biomarker for the quality of donated RBCs. However, a major impediment to the clinical translation of this biomarker is the difficulty associated with performing this measurement. To address this challenge, we developed an approach for biophysical profiling of RBCs based on cell images in thin-film blood smears. We hypothesize that subtle cellular changes are evident in blood smear images, but this information is inaccessible to human expert labellers. To test this hypothesis, we developed a deep learning strategy to analyze Giemsa-stained blood smears to assess the subtle morphologies indicative of RBC deformability and storage-based degradation. Specifically, we prepared thin-film blood smears from 27 RBC samples (9 donors evaluated at 3 storage time points) and imaged them using high-resolution microscopy. Using this dataset, we trained a convolutional neural network to evaluate image-based morphological features related to cell deformability. The prediction of donor deformability is strongly correlated to the microfluidic scores and can be used to categorize images into specific deformability groups with high accuracy. We also used this model to evaluate differences in RBC morphology resulting from cold storage. Together, our results demonstrate that deep learning models can detect subtle cellular morphology differences resulting from deformability and cold storage. This result suggests the potential to assess donor blood quality from thin-film blood smears, which can be acquired ubiquitously in clinical workflows.

[Chinese expert consensus on the technical and clinical practice specifications of artificial intelligence assisted morphology examination of blood cells (2024)]

Blood cell morphological examination is a crucial method for the diagnosis of blood diseases, but traditional manual microscopy is characterized by low efficiency and susceptibility to subjective biases. The application of artificial intelligence (AI) technology has improved the efficiency and quality of blood cell examinations and facilitated the standardization of test results. Currently, a variety of AI devices are either in clinical use or under research, with diverse technical requirements and configurations. The Experimental Diagnostic Study Group of the Hematology Branch of the Chinese Medical Association has organized a panel of experts to formulate this consensus. The consensus covers term definitions, scope of application, technical requirements, clinical application, data management, and information security. It emphasizes the importance of specimen preparation, image acquisition, image segmentation algorithms, and cell feature extraction and classification, and sets forth basic requirements for the cell recognition spectrum. Moreover, it provides detailed explanations regarding the fine classification of pathological cells, requirements for cell training and testing, quality control standards, and assistance in issuing diagnostic reports by humans. Additionally, the consensus underscores the significance of data management and information security to ensure the safety of patient information and the accuracy of data.

Label-free cell classification in holographic flow cytometry through an unbiased learning strategy

Nowadays, label-free imaging flow cytometry at the single-cell level is considered the stepforward lab-on-a-chip technology to address challenges in clinical diagnostics, biology, life sciences and healthcare. In this framework, digital holography in microscopy promises to be a powerful imaging modality thanks to its multi-refocusing and label-free quantitative phase imaging capabilities, along with the encoding of the highest information content within the imaged samples. Moreover, the recent achievements of new data analysis tools for cell classification based on deep/machine learning, combined with holographic imaging, are urging these systems toward the effective implementation of point of care devices. However, the generalization capabilities of learning-based models may be limited from biases caused by data obtained from other holographic imaging settings and/or different processing approaches. In this paper, we propose a combination of a Mask R-CNN to detect the cells, a convolutional auto-encoder, used to the image feature extraction and operating on unlabelled data, thus overcoming the bias due to data coming from different experimental settings, and a feedforward neural network for single cell classification, that operates on the above extracted features. We demonstrate the proposed approach in the challenging classification task related to the identification of drug-resistant endometrial cancer cells.

PyCoCa:A quantifying tool of carbon content in airway macrophage for assessment the internal dose of particles

Given the lack of a comprehensive understanding of the complex metabolism and variable exposure environment, carbon particles in macrophages have become a potentially valuable biomarker to assess the exposure level of atmospheric particles, such as black carbon. However, the tedious and subjective quantification method limits the application of carbon particles as a valid biomarker. Aiming to obtain an accurate carbon particles quantification method, the deep learning and binarization algorithm were implemented to develop a quantitative tool for carbon content in airway macrophage (CCAM), named PyCoCa. Two types of macrophages, normal and foamy appearance, were applied for the development of PyCoCa. In comparison with the traditional methods, PyCoCa significantly improves the identification efficiency for over 100 times. Consistency assessment with the gold standard revealed that PyCoCa exhibits outstanding prediction ability with the Interclass Correlation Coefficient (ICC) values of over 0.80. And a proper fresh dye will enhance the performance of PyCoCa (ICC = 0.89). Subsequent sensitivity analysis confirmed an excellent performance regarding accuracy and robustness of PyCoCa under high/low exposure environments (sensitivity > 0.80). Furthermore, a successful application of our quantitative tool in cohort studies indicates that carbon particles induce macrophage foaming and the foaming decrease the carbon particles internalization in reverse. Our present study provides a robust and efficient tool to accurately quantify the carbon particles loading in macrophage for exposure assessment.

Specific refraction-index increments of oxygenated hemoglobin from thalassemia-minor patients are not significantly different than those from healthy individuals

The mass and concentration of hemoglobin per erythrocyte are important hematological parameters. Measuring these parameters from intact erythrocytes requires the value of specific refraction-index increment (RII) of oxygenated hemoglobin, which diverges in the literature. Refractive indices of hemoglobin solutions are measured directly by digital holographic microscopy on a microfluidic channel filled with hemoglobin solutions prepared by hemolysis of fresh human erythrocytes and refractive-index standards sequentially. Hemoglobin extracted from thalassemic patients shows 3-4% higher RII than that from healthy volunteers, but the difference is not significant in comparison to inter-subject variations within each group. The quantified RIIs are applied to quantify mean corpuscular hemoglobin mass of blood from 37 human subjects, and results are in accord with standard clinical test results.

Method to analyze effects of low-level laser therapy on biological cells with a digital holographic microscope

Low-level laser therapy (LLLT) is a therapeutic tool that uses the photobiochemical interaction between light and tissue. Its effectiveness is controversial due to a strong dependence on dosimetric parameters. In this work, we demonstrate that digital holographic microscopy is an effective label-free imaging technique to analyze the effects of LLLT on biological cells, and we propose the full methodology to create correct synthetic aperture phase maps for further extensive, highly accurate statistical analysis. The proposed methodology has been designed to provide a basis for many other biological experiments using quantitative phase imaging. We use SHSY-5Y and HaCaT cells irradiated with different doses of red light for the experiment. The analysis shows quantitative changes in cell dry mass density and the projected cell surface in response to different radiation doses.

Machine learning-based approaches for identifying human blood cells harboring CRISPR-mediated fetal chromatin domain ablations

Two common hemoglobinopathies, sickle cell disease (SCD) and β-thalassemia, arise from genetic mutations within the β-globin gene. In this work, we identified a 500-bp motif (Fetal Chromatin Domain, FCD) upstream of human ϒ-globin locus and showed that the removal of this motif using CRISPR technology reactivates the expression of ϒ-globin. Next, we present two different cell morphology-based machine learning approaches that can be used identify human blood cells (KU-812) that harbor CRISPR-mediated FCD genetic modifications. Three candidate models from the first approach, which uses multilayer perceptron algorithm (MLP 20-26, MLP26-18, and MLP 30-26) and flow cytometry-derived cellular data, yielded 0.83 precision, 0.80 recall, 0.82 accuracy, and 0.90 area under the ROC (receiver operating characteristic) curve when predicting the edited cells. In comparison, the candidate model from the second approach, which uses deep learning (T2D5) and DIC microscopy-derived imaging data, performed with less accuracy (0.80) and ROC AUC (0.87). We envision that equivalent machine learning-based models can complement currently available genotyping protocols for specific genetic modifications which result in morphological changes in human cells.

Red Blood Cell Classification Based on Attention Residual Feature Pyramid Network

Clinically, red blood cell abnormalities are closely related to tumor diseases, red blood cell diseases, internal medicine, and other diseases. Red blood cell classification is the key to detecting red blood cell abnormalities. Traditional red blood cell classification is done manually by doctors, which requires a lot of manpower produces subjective results. This paper proposes an Attention-based Residual Feature Pyramid Network (ARFPN) to classify 14 types of red blood cells to assist the diagnosis of related diseases. The model performs classification directly on the entire red blood cell image. Meanwhile, a spatial attention mechanism and channel attention mechanism are combined with residual units to improve the expression of category-related features and achieve accurate extraction of features. Besides, the RoI align method is used to reduce the loss of spatial symmetry and improve classification accuracy. Five hundred and eighty eight red blood cell images are used to train and verify the effectiveness of the proposed method. The Channel Attention Residual Feature Pyramid Network (C-ARFPN) model achieves an mAP of 86%; the Channel and Spatial Attention Residual Feature Pyramid Network (CS-ARFPN) model achieves an mAP of 86.9%. The experimental results indicate that our method can classify more red blood cell types and better adapt to the needs of doctors, thus reducing the doctor's time and improving the diagnosis efficiency.

Assessing red blood cell deformability from microscopy images using deep learning

Red blood cells (RBCs) must be highly deformable to transit through the microvasculature to deliver oxygen to tissues. The loss of RBC deformability resulting from pathology, natural aging, or storage in blood bags can impede the proper function of these cells. A variety of methods have been developed to measure RBC deformability, but these methods require specialized equipment, long measurement time, and highly skilled personnel. To address this challenge, we investigated whether a machine learning approach could be used to predict donor RBC deformability based on morphological features from single cell microscope images. We used the microfluidic ratchet device to sort RBCs based on deformability. Sorted cells are then imaged and used to train a deep learning model to classify RBC based image features related to cell deformability. This model correctly predicted deformability of individual RBCs with 81 ± 11% accuracy averaged across ten donors. Using this model to score the deformability of RBC samples was accurate to within 10.4 ± 6.8% of the value obtained using the microfluidic ratchet device. While machine learning methods are frequently developed to automate human image analysis, our study is remarkable in showing that deep learning of single cell microscopy images could be used to assess RBC deformability, a property not normally measurable by imaging. Measuring RBC deformability by imaging is also desirable because it can be performed rapidly using a standard microscopy system, potentially enabling RBC deformability studies to be performed as part of routine clinical assessments.

Self-supervised pretraining for transferable quantitative phase image cell segmentation

In this paper, a novel U-Net-based method for robust adherent cell segmentation for quantitative phase microscopy image is designed and optimised. We designed and evaluated four specific post-processing pipelines. To increase the transferability to different cell types, non-deep learning transfer with adjustable parameters is used in the post-processing step. Additionally, we proposed a self-supervised pretraining technique using nonlabelled data, which is trained to reconstruct multiple image distortions and improved the segmentation performance from 0.67 to 0.70 of object-wise intersection over union. Moreover, we publish a new dataset of manually labelled images suitable for this task together with the unlabelled data for self-supervised pretraining.