Clinically validated machine learning algorithm for detecting residual diseases with multicolor flow cytometry analysis in acute myeloid leukemia and myelodysplastic syndrome

Prediction of inhibitor development in previously untreated and minimally treated children with severe and moderately severe hemophilia A using a machine-learning network

BACKGROUND

Prediction of inhibitor development in patients with hemophilia A (HA) remains a challenge.

OBJECTIVES

To construct a predictive model for inhibitor development in HA using a network of clinical variables and biomarkers based on the individual similarity network.

METHODS

Previously untreated and minimally treated children with severe/moderately severe HA, participants of the HEMFIL Cohort Study, were followed up until reaching 75 exposure days (EDs) without inhibitor (INH-) or upon inhibitor development (INH+). Clinical data and biological samples were collected before the start of factor (F)VIII replacement (T0). A predictive model (HemfilNET) was built to compare the networks and potential global topological differences between INH- and INH+ at T0, considering the network robustness. For validation, the "leave-one-out" cross-validation technique was employed. Accuracy, precision, recall, and F1-score were used as evaluation metrics for the machine-learning model.

RESULTS

We included 95 children with HA (CHA), of whom 31 (33%) developed inhibitors. The algorithm, featuring 37 variables, identified distinct patterns of networks at T0 for INH+ and INH-. The accuracy of the model was 74.2% for CHA INH+ and 98.4% for INH-. By focusing the analysis on CHA with high-risk F8 mutations for inhibitor development, the accuracy in identifying CHA INH+ increased to 82.1%.

CONCLUSION

Our machine-learning algorithm demonstrated an overall accuracy of 90.5% for predicting inhibitor development in CHA, which further improved when restricting the analysis to CHA with a high-risk F8 genotype. However, our model requires validation in other cohorts. Yet, missing data for some variables hindered more precise predictions.

Comparison of three machine learning algorithms for classification of B-cell neoplasms using clinical flow cytometry data

Multiparameter flow cytometry data is visually inspected by expert personnel as part of standard clinical disease diagnosis practice. This is a demanding and costly process, and recent research has demonstrated that it is possible to utilize artificial intelligence (AI) algorithms to assist in the interpretive process. Here we report our examination of three previously published machine learning methods for classification of flow cytometry data and apply these to a B-cell neoplasm dataset to obtain predicted disease subtypes. Each of the examined methods classifies samples according to specific disease categories using ungated flow cytometry data. We compare and contrast the three algorithms with respect to their architectures, and we report the multiclass classification accuracies and relative required computation times. Despite different architectures, two of the methods, flowCat and EnsembleCNN, had similarly good accuracies with relatively fast computational times. We note a speed advantage for EnsembleCNN, particularly in the case of addition of training data and retraining of the classifier.

Measurable (Minimal) Residual Disease in Myelodysplastic Neoplasms (MDS): Current State and Perspectives

Myelodysplastic Neoplasms (MDS) have been traditionally studied through the assessment of blood counts, cytogenetics, and morphology. In recent years, the introduction of molecular assays has improved our ability to diagnose MDS. The role of Measurable (minimal) Residual Disease (MRD) in MDS is evolving, and molecular and flow cytometry techniques have been used in several studies. In this review, we will highlight the evolving concept of MRD in MDS, outline the various techniques utilized, and provide an overview of the studies reporting MRD and the correlation with outcomes.

Validation of Artificial Intelligence (AI)-Assisted Flow Cytometry Analysis for Immunological Disorders

Flow cytometry is a vital diagnostic tool for hematologic and immunologic disorders, but manual analysis is prone to variation and time-consuming. Over the last decade, artificial intelligence (AI) has advanced significantly. In this study, we developed and validated an AI-assisted flow cytometry workflow using 379 clinical cases from 2021, employing a 3-tube, 10-color flow panel with 21 antibodies for primary immunodeficiency diseases and related immunological disorders. The AI software (DeepFlow™, version 2.1.1) is fully automated, reducing analysis time to under 5 min per case. It interacts with hematopatholoists for manual gating adjustments when necessary. Using proprietary multidimensional density-phenotype coupling algorithm, the AI model accurately classifies and enumerates T, B, and NK cells, along with important immune cell subsets, including CD4+ helper T cells, CD8+ cytotoxic T cells, CD3+/CD4-/CD8- double-negative T cells, and class-switched or non-switched B cells. Compared to manual analysis with hematopathologist-determined lymphocyte subset percentages as the gold standard, the AI model exhibited a strong correlation (r > 0.9) across lymphocyte subsets. This study highlights the accuracy and efficiency of AI-assisted flow cytometry in diagnosing immunological disorders in a clinical setting, providing a transformative approach within a concise timeframe.

Using Artificial Intelligence to Interpret Clinical Flow Cytometry Datasets for Automated Disease Diagnosis and/or Monitoring

Flow cytometry (FC) is routinely used for hematological disease diagnosis and monitoring. Advancement in this technology allows us to measure an increasing number of markers simultaneously, generating complex high-dimensional datasets. However, current analytic software and methods rely on experienced analysts to perform labor-intensive manual inspection and interpretation on a series of 2-dimensional plots via a complex, sequential gating process. With an aggravating shortage of professionals and growing demands, it is very challenging to provide the FC analysis results in a fast, accurate, and reproducible way. Artificial intelligence has been widely used in many sectors to develop automated detection or classification tools. Here we describe a type of machine learning method for developing automated disease classification and residual disease monitoring on clinical flow datasets.

A Path of Novelty from Nanoparticles to Nanobots: Theragnostic Approach for Targeting Cancer Therapy

Pharmaceutical development of cancer therapeutics is a dynamic area of research. Even after decades of intensive work, cancer continues to be a dreadful disease with an ever-increasing global incidence. The progress of nanotechnology in cancer research has overcome inherent limitations in conventional cancer chemotherapy and fulfilled the need for target-specific drug carriers. Nanotechnology uses the altered patho-physiological microenvironment of malignant cells and offers various advantages like improved solubility, reduced toxicity, prolonged drug circulation with controlled release, circumventing multidrug resistance, and enhanced biodistribution. Early cancer detection has a crucial role in selecting the best drug regime, thus, diagnosis and therapeutics go hand in hand. Furthermore, nanobots are an amazing possibility and promising innovation with numerous significant applications, particularly in fighting cancer and cleaning out blood vessels. Nanobots are tiny robots, ranging in size from 1 to 100 nm. Moreover, the nanobots would work similarly to white blood cells, watching the bloodstream and searching for indications of distress. This review articulates the evolution of various organic and inorganic nanoparticles and nanobots used as therapeutics, along with their pros and cons. It also highlights the shift in diagnostics from conventional methods to more advanced techniques. This rapidly growing domain is providing more space for engineering desired nanoparticles that can show miraculous results in therapeutic and diagnostic trials.

A cell-level discriminative neural network model for diagnosis of blood cancers

MOTIVATION

Precise identification of cancer cells in patient samples is essential for accurate diagnosis and clinical monitoring but has been a significant challenge in machine learning approaches for cancer precision medicine. In most scenarios, training data are only available with disease annotation at the subject or sample level. Traditional approaches separate the classification process into multiple steps that are optimized independently. Recent methods either focus on predicting sample-level diagnosis without identifying individual pathologic cells or are less effective for identifying heterogeneous cancer cell phenotypes.

RESULTS

We developed a generalized end-to-end differentiable model, the Cell Scoring Neural Network (CSNN), which takes sample-level training data and predicts the diagnosis of the testing samples and the identity of the diagnostic cells in the sample, simultaneously. The cell-level density differences between samples are linked to the sample diagnosis, which allows the probabilities of individual cells being diagnostic to be calculated using backpropagation. We applied CSNN to two independent clinical flow cytometry datasets for leukemia diagnosis. In both qualitative and quantitative assessments, CSNN outperformed preexisting neural network modeling approaches for both cancer diagnosis and cell-level classification. Post hoc decision trees and 2D dot plots were generated for interpretation of the identified cancer cells, showing that the identified cell phenotypes match the cancer endotypes observed clinically in patient cohorts. Independent data clustering analysis confirmed the identified cancer cell populations.

AVAILABILITY AND IMPLEMENTATION

The source code of CSNN and datasets used in the experiments are publicly available on GitHub (http://github.com/erobl/csnn). Raw FCS files can be downloaded from FlowRepository (ID: FR-FCM-Z6YK).

Exploring alternative approaches to precision medicine through genomics and artificial intelligence - a systematic review

The core idea behind precision medicine is to pinpoint the subpopulations that differ from one another in terms of disease risk, drug responsiveness, and treatment outcomes due to differences in biology and other traits. Biomarkers are found through genomic sequencing. Multi-dimensional clinical and biological data are created using these biomarkers. Better analytic methods are needed for these multidimensional data, which can be accomplished by using artificial intelligence (AI). An updated review of 80 latest original publications is presented on four main fronts-preventive medicine, medication development, treatment outcomes, and diagnostic medicine-All these studies effectively illustrated the significance of AI in precision medicine. Artificial intelligence (AI) has revolutionized precision medicine by swiftly analyzing vast amounts of data to provide tailored treatments and predictive diagnostics. Through machine learning algorithms and high-resolution imaging, AI assists in precise diagnoses and early disease detection. AI's ability to decode complex biological factors aids in identifying novel therapeutic targets, allowing personalized interventions and optimizing treatment outcomes. Furthermore, AI accelerates drug discovery by navigating chemical structures and predicting drug-target interactions, expediting the development of life-saving medications. With its unrivaled capacity to comprehend and interpret data, AI stands as an invaluable tool in the pursuit of enhanced patient care and improved health outcomes. It's evident that AI can open a new horizon for precision medicine by translating complex data into actionable information. To get better results in this regard and to fully exploit the great potential of AI, further research is required on this pressing subject.

Artificial Intelligence for Clinical Flow Cytometry

In this review, the authors discuss the fundamental principles of machine learning. They explore recent studies and approaches in implementing machine learning into flow cytometry workflows. These applications are promising but not without their shortcomings. Explainability may be the biggest barrier to adoption, as they contain "black boxes" in which a complex network of mathematical processes learns features of data that are not translatable into real language. The authors discuss the current limitations of machine learning models and the possibility that, without a multiinstitutional development process, these applications could have poor generalizability. They also discuss widespread deployment of augmented decision-making.

Highly sensitive single tube B-lymphoblastic leukemia/lymphoma minimal/measurable residual disease test robust to surface antigen directed therapy

BACKGROUND

Measurement of minimal/measurable residual disease (MRD) in B-lymphoblastic leukemia/lymphoma (B-ALL) has become a routine clinical evaluation tool and remains the strongest predictor of treatment outcome. In recent years, new targeted anti-CD19 and anti-CD22 antibody-based and cellular therapies have revolutionized the treatment of the high-risk B-ALL. The new treatments raise challenges for diagnostic flow cytometry, which relies on the presence of specific surface antigens to identify the population of interest. So far, reported flow cytometry-based assays are developed to either achieve a deeper MRD level or to accommodate the loss of surface antigens post-target therapies, but not both.

METHODS

We developed a single tube flow cytometry assay (14-color-16-parameters). The method was validated using 94 clinical samples as well as spike-in and replicate experiments.

RESULTS

The assay was well suited for monitoring response to targeted therapies and reached a sensitivity below 10-5 with acceptable precision (coefficient of variation < 20%), accuracy, and interobserver variability (κ = 1).

CONCLUSIONS

The assay allows for sensitive disease detection of B-ALL MRD independent of CD19 and CD22 expression and allows uniform analysis of samples regardless of anti-CD19 and CD22 therapy.

Liquid biopsies and minimal residual disease in lymphoid malignancies

Minimal residual disease (MRD) assessment using peripheral blood instead of bone marrow aspirate/biopsy specimen or the biopsy of the cancerous infiltrated by lymphoid malignancies is an emerging technique with enormous interest of research and technological innovation at the current time. In some lymphoid malignancies (particularly ALL), Studies have shown that MRD monitoring of the peripheral blood may be an adequate alternative to frequent BM aspirations. However, additional studies investigating the biology of liquid biopsies in ALL and its potential as an MRD marker in larger patient cohorts in treatment protocols are warranted. Despite the promising data, there are still limitations in liquid biopsies in lymphoid malignancies, such as standardization of the sample collection and processing, determination of timing and duration for liquid biopsy analysis, and definition of the biological characteristics and specificity of the techniques evaluated such as flow cytometry, molecular techniques, and next generation sequencies. The use of liquid biopsy for detection of minimal residual disease in T-cell lymphoma is still experimental but it has made significant progress in multiple myeloma for example. Recent attempt to use artificial intelligence may help simplify the algorithm for testing and may help avoid inter-observer variation and operator dependency in these highly technically demanding testing process.

Challenging gold standard hematology diagnostics through the introduction of whole genome sequencing and artificial intelligence

The diagnosis of hematological malignancies is rather complex and requires the application of a plethora of different assays, techniques and methodologies. Some of the methods, like cytomorphology, have been in use for decades, while other methods, such as next-generation sequencing or even whole genome sequencing (WGS), are relatively new. The application of the methods and the evaluation of the results require distinct skills and knowledge and place different demands on the practitioner. However, even with experienced hematologists, diagnostic ambiguity remains a regular occurrence and the comprehensive analysis of high-dimensional WGS data soon exceeds any human's capacity. Hence, in order to reduce inter-observer variability and to improve the timeliness and accuracy of diagnoses, machine learning based approaches have been developed to assist in the decision making process. Moreover, to achieve the goal of precision oncology, comprehensive genomic profiling is increasingly being incorporated into routine standard of care.

Artificial intelligence in clinical multiparameter flow cytometry and mass cytometry-key tools and progress

There are many research studies and emerging tools using artificial intelligence (AI) and machine learning to augment flow and mass cytometry workflows. Emerging AI tools can quickly identify common cell populations with continuous improvement of accuracy, uncover patterns in high-dimensional cytometric data that are undetectable by human analysis, facilitate the discovery of cell subpopulations, perform semi-automated immune cell profiling, and demonstrate potential to automate aspects of clinical multiparameter flow cytometric (MFC) diagnostic workflow. Utilizing AI in the analysis of cytometry samples can reduce subjective variability and assist in breakthroughs in understanding diseases. Here we review the diverse types of AI that are being applied to clinical cytometry data and how AI is driving advances in data analysis to improve diagnostic sensitivity and accuracy. We review supervised and unsupervised clustering algorithms for cell population identification, various dimensionality reduction techniques, and their utilities in visualization and machine learning pipelines, and supervised learning approaches for classifying entire cytometry samples.Understanding the AI landscape will enable pathologists to better utilize open source and commercially available tools, plan exploratory research projects to characterize diseases, and work with machine learning and data scientists to implement clinical data analysis pipelines.

Artificial intelligence in hematological diagnostics: Game changer or gadget?

The future of clinical diagnosis and treatment of hematologic diseases will inevitably involve the integration of artificial intelligence (AI)-based systems into routine practice to support the hematologists' decision making. Several studies have shown that AI-based models can already be used to automatically differentiate cells, reliably detect malignant cell populations, support chromosome banding analysis, and interpret clinical variants, contributing to early disease detection and prognosis. However, even the best tool can become useless if it is misapplied or the results are misinterpreted. Therefore, in order to comprehensively judge and correctly apply newly developed AI-based systems, the hematologist must have a basic understanding of the general concepts of machine learning. In this review, we provide the hematologist with a comprehensive overview of various machine learning techniques, their current implementations and approaches in different diagnostic subfields (e.g., cytogenetics, molecular genetics), and the limitations and unresolved challenges of the systems.

A cell-level discriminative neural network model for diagnosis of blood cancers

Motivation

Precise identification of cancer cells in patient samples is essential for accurate diagnosis and clinical monitoring but has been a significant challenge in machine learning approaches for cancer precision medicine. In most scenarios, training data are only available with disease annotation at the subject or sample level. Traditional approaches separate the classification process into multiple steps that are optimized independently. Recent methods either focus on predicting sample-level diagnosis without identifying individual pathologic cells or are less effective for identifying heterogeneous cancer cell phenotypes.

Results

We developed a generalized end-to-end differentiable model, the Cell Scoring Neural Network (CSNN), which takes the available sample-level training data and predicts both the diagnosis of the testing samples and the identity of the diagnostic cells in the sample, simultaneously. The cell-level density differences between samples are linked to the sample diagnosis, which allows the probabilities of individual cells being diagnostic to be calculated using backpropagation. We applied CSNN to two independent clinical flow cytometry datasets for leukemia diagnosis. In both qualitative and quantitative assessments, CSNN outperformed preexisting neural network modeling approaches for both cancer diagnosis and cell-level classification. Post hoc decision trees and 2D dot plots were generated for interpretation of the identified cancer cells, showing that the identified cell phenotypes match the cancer endotypes observed clinically in patient cohorts. Independent data clustering analysis confirmed the identified cancer cell populations.

Availability

The source code of CSNN and datasets used in the experiments are publicly available on GitHub and FlowRepository.

Contact

Edgar E. Robles: roblesee@uci.edu and Yu Qian: mqian@jcvi.org.

Supplementary information

Supplementary data are available on GitHub and at Bioinformatics online.

Hematology and Machine Learning

BACKGROUND

Substantial improvements in computational power and machine learning (ML) algorithm development have vastly increased the limits of what autonomous machines are capable of. Since its beginnings in the 19th century, laboratory hematology has absorbed waves of progress yielding improvements in both of accuracy and efficiency. The next wave of change in laboratory hematology will be the result of the ML revolution that has already touched many corners of healthcare and society at large.

CONTENT

This review will describe the manifestations of ML and artificial intelligence (AI) already utilized in the clinical hematology laboratory. This will be followed by a topical summary of the innovative and investigational applications of this technology in each of the major subdomains within laboratory hematology.

SUMMARY

Application of this technology to laboratory hematology will increase standardization and efficiency by reducing laboratory staff involvement in automatable activities. This will unleash time and resources for focus on more meaningful activities such as the complexities of patient care, research and development, and process improvement.

Multiparameter flow cytometry in the evaluation of myelodysplasia: Analytical issues: Recommendations from the European LeukemiaNet/International Myelodysplastic Syndrome Flow Cytometry Working Group

Multiparameter flow cytometry (MFC) is one of the essential ancillary methods in bone marrow (BM) investigation of patients with cytopenia and suspected myelodysplastic syndrome (MDS). MFC can also be applied in the follow-up of MDS patients undergoing treatment. This document summarizes recommendations from the International/European Leukemia Net Working Group for Flow Cytometry in Myelodysplastic Syndromes (ELN iMDS Flow) on the analytical issues in MFC for the diagnostic work-up of MDS. Recommendations for the analysis of several BM cell subsets such as myeloid precursors, maturing granulocytic and monocytic components and erythropoiesis are given. A core set of 17 markers identified as independently related to a cytomorphologic diagnosis of myelodysplasia is suggested as mandatory for MFC evaluation of BM in a patient with cytopenia. A myeloid precursor cell (CD34⁺ CD19^- ) count >3% should be considered immunophenotypically indicative of myelodysplasia. However, MFC results should always be evaluated as part of an integrated hematopathology work-up. Looking forward, several machine-learning-based analytical tools of interest should be applied in parallel to conventional analytical methods to investigate their usefulness in integrated diagnostics, risk stratification, and potentially even in the evaluation of response to therapy, based on MFC data. In addition, compiling large uniform datasets is desirable, as most of the machine-learning-based methods tend to perform better with larger numbers of investigated samples, especially in such a heterogeneous disease as MDS.

Liquid biopsies and minimal residual disease in myeloid malignancies

Minimal residual disease (MRD) assessment through blood component sampling by liquid biopsies (LBs) is increasingly being investigated in myeloid malignancies. Blood components then undergo molecular analysis by flow cytometry or sequencing techniques and can be used as a powerful tool for prognostic and predictive purposes in myeloid malignancies. There is evidence and more is evolving about the quantification and identification of cell-based and gene-based biomarkers in myeloid malignancies to monitor treatment response. MRD based acute myeloid leukemia protocol and clinical trials are currently incorporating LB testing and preliminary results are encouraging for potential widespread use in clinic in the near future. MRD monitoring using LBs are not standard in myelodysplastic syndrome (MDS) but this is an area of active investigation. In the future, LBs can replace more invasive techniques such as bone marrow biopsies. However, the routine clinical application of these markers continues to be an issue due to lack of standardization and limited number of studies investigating their specificities. Integrating artificial intelligence (AI) could help simplify the complex interpretation of molecular testing and reduce errors related to operator dependency. Though the field is rapidly evolving, the applicability of MRD testing using LB is mostly limited to research setting at this time due to the need for validation, regulatory approval, payer coverage, and cost issues. This review focuses on the types of biomarkers, most recent research exploring MRD and LB in myeloid malignancies, ongoing clinical trials, and the future of LB in the setting of AI.

Reproducible measurable residual disease detection by multiparametric flow cytometry in acute myeloid leukemia

Measurable residual disease (MRD) detected by multiparametric flow cytometry (MFC) is associated with unfavorable outcome in patients with AML. A simple, broadly applicable eight-color panel was implemented and analyzed utilizing a hierarchical gating strategy with fixed gates to develop a clear-cut LAIP-based DfN approach. In total, 32 subpopulations with aberrant phenotypes with/without expression of markers of immaturity were monitored in 246 AML patients after completion of induction chemotherapy. Reference values were established utilizing 90 leukemia-free controls. Overall, 73% of patients achieved a response by cytomorphology. In responders, the overall survival was shorter for MRDpos patients (HR 3.8, p = 0.006). Overall survival of MRDneg non-responders was comparable to MRDneg responders. The inter-rater-reliability for MRD detection was high with a Krippendorffs α of 0.860. The mean time requirement for MRD analyses at follow-up was very short with 04:31 minutes. The proposed one-tube MFC approach for detection of MRD allows a high level of standardization leading to a promising inter-observer-reliability with a fast turnover. MRD defined by this strategy provides relevant prognostic information and establishes aberrancies outside of cell populations with markers of immaturity as an independent risk feature. Our results imply that this strategy may provide the base for multicentric immunophenotypic MRD assessment.

Biosensors and machine learning for enhanced detection, stratification, and classification of cells: a review

Biological cells, by definition, are the basic units which contain the fundamental molecules of life of which all living things are composed. Understanding how they function and differentiating cells from one another, therefore, is of paramount importance for disease diagnostics as well as therapeutics. Sensors focusing on the detection and stratification of cells have gained popularity as technological advancements have allowed for the miniaturization of various components inching us closer to Point-of-Care (POC) solutions with each passing day. Furthermore, Machine Learning has allowed for enhancement in the analytical capabilities of these various biosensing modalities, especially the challenging task of classification of cells into various categories using a data-driven approach rather than physics-driven. In this review, we provide an account of how Machine Learning has been applied explicitly to sensors that detect and classify cells. We also provide a comparison of how different sensing modalities and algorithms affect the classifier accuracy and the dataset size required.

A Review of Artificial Intelligence Applications in Hematology Management: Current Practices and Future Prospects

Background

Machine learning (ML) and deep learning (DL) methods have recently garnered a great deal of attention in the field of cancer research by making a noticeable contribution to the growth of predictive medicine and modern oncological practices. Considerable focus has been particularly directed toward hematologic malignancies because of the complexity in detecting early symptoms. Many patients with blood cancer do not get properly diagnosed until their cancer has reached an advanced stage with limited treatment prospects. Hence, the state-of-the-art revolves around the latest artificial intelligence (AI) applications in hematology management.

Objective

This comprehensive review provides an in-depth analysis of the current AI practices in the field of hematology. Our objective is to explore the ML and DL applications in blood cancer research, with a special focus on the type of hematologic malignancies and the patient’s cancer stage to determine future research directions in blood cancer.

Methods

We searched a set of recognized databases (Scopus, Springer, and Web of Science) using a selected number of keywords. We included studies written in English and published between 2015 and 2021. For each study, we identified the ML and DL techniques used and highlighted the performance of each model.

Results

Using the aforementioned inclusion criteria, the search resulted in 567 papers, of which 144 were selected for review.

Conclusions

The current literature suggests that the application of AI in the field of hematology has generated impressive results in the screening, diagnosis, and treatment stages. Nevertheless, optimizing the patient’s pathway to treatment requires a prior prediction of the malignancy based on the patient’s symptoms or blood records, which is an area that has still not been properly investigated.

A Chunking-for-Pooling Strategy for Cytometric Representation Learning for Automatic Hematologic Malignancy Classification

Differentiating types of hematologic malignancies is vital to determine therapeutic strategies for the newly diagnosed patients. Flow cytometry (FC) can be used as diagnostic indicator by measuring the multi-parameter fluorescent markers on thousands of antibody-bound cells, but the manual interpretation of large scale flow cytometry data has long been a time-consuming and complicated task for hematologists and laboratory professionals. Past studies have led to the development of representation learning algorithms to perform sample-level automatic classification. In this work, we propose a chunking-for-pooling strategy to include large-scale FC data into a supervised deep representation learning procedure for automatic hematologic malignancy classification. The use of discriminatively-trained representation learning strategy and the fixed-size chunking and pooling design are key components of this framework. It improves the discriminative power of the FC sample-level embedding and simultaneously addresses the robustness issue due to an inevitable use of down-sampling in conventional distribution based approaches for deriving FC representation. We evaluated our framework on two datasets. Our framework outperformed other baseline methods and achieved 92.3% unweighted average recall (UAR) for four-class recognition on the UPMC dataset and 85.0% UAR for five-class recognition on the hema.to dataset. We further compared the robustness of our proposed framework with that of the traditional downsampling approach. Analysis of the effects of the chunk size and the error cases revealed further insights about different hematologic malignancy characteristics in the FC data.

Circulating lnc-LOC as a novel noninvasive biomarker in the treatment surveillance of acute promyelocytic leukaemia

BACKGROUND

Acute promyelocytic leukaemia (APL) is a unique subtype of acute myeloid leukaemia (AML) characterized by haematopoietic failure caused by the accumulation of abnormal promyelocytic cells in bone marrow (BM). However, indispensable BM biopsy frequently afflicts patients in leukaemia surveillance, which increases the burden on patients and reduces compliance. This study aimed to explore whether the novel circulating long noncoding RNA LOC100506453 (lnc-LOC) could be a target in diagnosis, assess the treatment response and supervise the minimal residual disease (MRD) of APL, thereby blazing a trail in noninvasive lncRNA biomarkers of APL.

METHODS

Our study comprised 100 patients (40 with APL and 60 with non-APL AML) and 60 healthy donors. BM and peripheral blood (PB) sample collection was accomplished from APL patients at diagnosis and postinduction. Quantitative real-time PCR (qRT-PCR) was conducted to evaluate lnc-LOC expression. A receiver operating characteristic (ROC) analysis was implemented to analyse the value of lnc-LOC in the diagnosis of APL and treatment monitoring. For statistical analysis, the Mann-Whitney U test, a t test, and Spearman's rank correlation test were utilized.

RESULTS

Our results showed that BM lnc-LOC expression was significantly different between APL and healthy donors and non-APL AML. lnc-LOC was drastically downregulated in APL patients' BM after undergoing induction therapy. Lnc-LOC was upregulated in APL cell lines and downregulated after all-trans retinoic acid (ATRA)-induced myeloid differentiation, preliminarily verifying that lnc-LOC has the potential to be considered a treatment monitoring biomarker. PB lnc-LOC was positively correlated with BM lnc-LOC in APL patients, non-APL AML patients and healthy donors and decreased sharply after complete remission (CR). However, upregulated lnc-LOC was manifested in relapsed-refractory patients. A positive correlation was revealed between PB lnc-LOC and PML-RARα transcript levels in BM samples. Furthermore, we observed a positive correlation between PB lnc-LOC and BM lnc-LOC expression in APL patients, suggesting that lnc-LOC can be utilized as a noninvasive biomarker for MRD surveillance.

CONCLUSIONS

Our study demonstrated that PB lnc-LOC might serve as a novel noninvasive biomarker in the treatment surveillance of APL, and it innovated the investigation and application of newly found lncRNAs in APL noninvasive biomarkers used in diagnosis and detection.

A Machine Learning Approach to the Classification of Acute Leukemias and Distinction From Nonneoplastic Cytopenias Using Flow Cytometry Data

OBJECTIVES

Flow cytometry (FC) is critical for the diagnosis and monitoring of hematologic malignancies. Machine learning (ML) methods rapidly classify multidimensional data and should dramatically improve the efficiency of FC data analysis. We aimed to build a model to classify acute leukemias, including acute promyelocytic leukemia (APL), and distinguish them from nonneoplastic cytopenias. We also sought to illustrate a method to identify key FC parameters that contribute to the model's performance.

METHODS

Using data from 531 patients who underwent evaluation for cytopenias and/or acute leukemia, we developed an ML model to rapidly distinguish among APL, acute myeloid leukemia/not APL, acute lymphoblastic leukemia, and nonneoplastic cytopenias. Unsupervised learning using gaussian mixture model and Fisher kernel methods were applied to FC listmode data, followed by supervised support vector machine classification.

RESULTS

High accuracy (ACC, 94.2%; area under the curve [AUC], 99.5%) was achieved based on the 37-parameter FC panel. Using only 3 parameters, however, yielded similar performance (ACC, 91.7%; AUC, 98.3%) and highlighted the significant contribution of light scatter properties.

CONCLUSIONS

Our findings underscore the potential for ML to automatically identify and prioritize FC specimens that have critical results, including APL and other acute leukemias.

Diagnosis of Acute Leukemia by Multiparameter Flow Cytometry with the Assistance of Artificial Intelligence

We developed an artificial intelligence (AI) model that evaluates the feasibility of AI-assisted multiparameter flow cytometry (MFC) diagnosis of acute leukemia. Two hundred acute leukemia patients and 94 patients with cytopenia(s) or hematocytosis were selected to study the AI application in MFC diagnosis of acute leukemia. The kappa test analyzed the consistency of the diagnostic results and the immunophenotype of acute leukemia. Bland–Altman and Pearson analyses evaluated the consistency and correlation of the abnormal cell proportion between the AI and manual methods. The AI analysis time for each case (83.72 ± 23.90 s, mean ± SD) was significantly shorter than the average time for manual analysis (15.64 ± 7.16 min, mean ± SD). The total consistency of diagnostic results was 0.976 (kappa (κ) = 0.963). The Bland–Altman evaluation of the abnormal cell proportion between the AI analysis and manual analysis showed that the bias ± SD was 0.752 ± 6.646, and the 95% limit of agreement was from −12.775 to 13.779 (p = 0.1225). The total consistency of the AI immunophenotypic diagnosis and the manual results was 0.889 (kappa, 0.775). The consistency and speedup of the AI-assisted workflow indicate its promising clinical application.

Potential for Process Improvement of Clinical Flow Cytometry by Incorporating Real-Time Automated Screening of Data to Expedite Addition of Antibody Panels

OBJECTIVES

We desired an automated approach to expedite ordering additional antibody panels in our clinical flow cytometry lab. This addition could improve turnaround times, decrease time spent revisiting cases, and improve consistency.

METHODS

We trained a machine learning classifier to use our screening B-cell panel to predict whether we should order an additional panel to distinguish chronic lymphocytic lymphoma from mantle cell lymphoma. We used data from 2016 to 2018 for training and validation, and cases were restricted to the first case per patient (9,635 cases, 887 with the additional panel). We applied the model in real time over approximately 2.5 months in 2020 to 376 sequential cases, with automated email notifications for positive predictions.

RESULTS

Using 80% of the data from 2016 to 2018 to train and 20% for validation, we achieved 95% area under the receiving operating characteristic curve (AUROC) and 94% accuracy in the validation set. Applying the classifier in real time achieved 89% AUROC and 94% real-time prediction accuracy (precision [positive predictive value] = 51%, recall [sensitivity] = 78%, and F1 score = 0.62). Fourteen of the 17 false positives had prior diagnoses to which the algorithm was not privy.

CONCLUSIONS

As an observational, not interventional study, our system performed well on testing within our laboratory for identifying cases to be flagged but cannot be used without laboratory-specific modifications.

Light sheet based volume flow cytometry (VFC) for rapid volume reconstruction and parameter estimation on the go

Optical imaging is paramount for disease diagnosis and to access its progression over time. The proposed optical flow imaging (VFC/iLIFE) is a powerful technique that adds new capabilities (3D volume visualization, organelle-level resolution, and multi-organelle screening) to the existing system. Unlike state-of-the-art point-illumination-based biomedical imaging techniques, the sheet-based VFC technique is capable of single-shot sectional visualization, high throughput interrogation, real-time parameter estimation, and instant volume reconstruction with organelle-level resolution of live specimens. The specimen flow system was realized on a multichannel (Y-type) microfluidic chip that enables visualization of organelle distribution in several cells in-parallel at a relatively high flow-rate (2000 nl/min). The calibration of VFC system requires the study of point emitters (fluorescent beads) at physiologically relevant flow-rates (500-2000 nl/min) for determining flow-induced optical aberration in the system point spread function (PSF). Subsequently, the recorded raw images and volumes were computationally deconvolved with flow-variant PSF to reconstruct the cell volume. High throughput investigation of the mitochondrial network in HeLa cancer cell was carried out at sub-cellular resolution in real-time and critical parameters (mitochondria count and size distribution, morphology, entropy, and cell strain statistics) were determined on-the-go. These parameters determine the physiological state of cells, and the changes over-time, revealing the metastatic progression of diseases. Overall, the developed VFC system enables real-time monitoring of sub-cellular organelle organization at a high-throughput with high-content capacity.

A Clinical Prognostic Model Based on Machine Learning from the Fondazione Italiana Linfomi (FIL) MCL0208 Phase III Trial

BACKGROUND

Multicenter clinical trials are producing growing amounts of clinical data. Machine Learning (ML) might facilitate the discovery of novel tools for prognostication and disease-stratification. Taking advantage of a systematic collection of multiple variables, we developed a model derived from data collected on 300 patients with mantle cell lymphoma (MCL) from the Fondazione Italiana Linfomi-MCL0208 phase III trial (NCT02354313).

METHODS

We developed a score with a clustering algorithm applied to clinical variables. The candidate score was correlated to overall survival (OS) and validated in two independent data series from the European MCL Network (NCT00209222, NCT00209209); Results: Three groups of patients were significantly discriminated: Low, Intermediate (Int), and High risk (High). Seven discriminants were identified by a feature reduction approach: albumin, Ki-67, lactate dehydrogenase, lymphocytes, platelets, bone marrow infiltration, and B-symptoms. Accordingly, patients in the Int and High groups had shorter OS rates than those in the Low and Int groups, respectively (Int→Low, HR: 3.1, 95% CI: 1.0-9.6; High→Int, HR: 2.3, 95% CI: 1.5-4.7). Based on the 7 markers, we defined the engineered MCL international prognostic index (eMIPI), which was validated and confirmed in two independent cohorts; Conclusions: We developed and validated a ML-based prognostic model for MCL. Even when currently limited to baseline predictors, our approach has high scalability potential.

De Novo Identification and Visualization of Important Cell Populations for Classic Hodgkin Lymphoma Using Flow Cytometry and Machine Learning

OBJECTIVES

Automated classification of flow cytometry data has the potential to reduce errors and accelerate flow cytometry interpretation. We desired a machine learning approach that is accurate, is intuitively easy to understand, and highlights the cells that are most important in the algorithm's prediction for a given case.

METHODS

We developed an ensemble of convolutional neural networks for classification and visualization of impactful cell populations in detecting classic Hodgkin lymphoma using two-dimensional (2D) histograms. Data from 977 and 245 clinical flow cytometry cases were used for training and testing, respectively. Seventy-eight nongated 2D histograms were created per flow cytometry file. Shapley additive explanation (SHAP) values were calculated to determine the most impactful 2D histograms and regions within histograms. SHAP values from all 78 histograms were then projected back to the original cell data for gating and visualization using standard flow cytometry software.

RESULTS

The algorithm achieved 67.7% recall (sensitivity), 82.4% precision, and 0.92 area under the receiver operating characteristic. Visualization of the important cell populations for individual predictions demonstrated correlations with known biology.

CONCLUSIONS

The method presented enables model explainability while highlighting important cell populations in individual flow cytometry specimens, with potential applications in both diagnosis and discovery of previously overlooked key cell populations.

Machine Learning Applications in the Diagnosis of Benign and Malignant Hematological Diseases

The use of machine learning (ML) and deep learning (DL) methods in hematology includes diagnostic, prognostic, and therapeutic applications. This increase is due to the improved access to ML and DL tools and the expansion of medical data. The utilization of ML remains limited in clinical practice, with some disciplines further along in their adoption, such as radiology and histopathology. In this review, we discuss the current uses of ML in diagnosis in the field of hematology, including image-recognition, laboratory, and genomics-based diagnosis. Additionally, we provide an introduction to the fields of ML and DL, highlighting current trends, limitations, and possible areas of improvement.

The Prognostic Value of Early Detection of Minimal Residual Disease as Defined by Flow Cytometry and Gene Mutation Clearance for Myelodysplastic Syndrome Patients After Myeloablative Allogeneic Hematopoietic Stem-Cell Transplantation

High relapse incidence remains a major problem for myelodysplastic syndrome (MDS) patients who have received an allogeneic hematopoietic stem-cell transplantation (allo-HSCT). We retrospectively analyzed the correlations between clinical outcomes and minimal residual disease (MRD) by using mutations (MUT) and flow cytometry (FCM) analysis of 115 MDS patients with allo-HSCT. We divided 115 MDS patients into four groups based on molecular genetics and FCM MRD results at day 30 post-HSCT. There were significant differences in the 2-year progression-free survival (PFS) between the FCM^high MUT^pos and FCM^low MUT^neg groups (20% vs 79%, P < 0.001). In addition, by univariate analysis, we found that an IPSS-R score ≥4 pre-HSCT (HR, 5.061; P=0.007), DNMT3A mutations (HR, 2.291; P=0.052), TP53 mutations (HR, 3.946; P=0.011), and poor and very poor revised International Prognostic Scoring System (IPSS-R) cytogenetic risk (HR, 4.906; P < 0.001) were poor risk factors for PFS. In multivariate analysis, we found that an IPSS-R score ≥ 4 pre-HSCT (HR, 4.488; P=0.015), DNMT3A mutations (HR, 2.385; P=0.049), positive FCM MRD combined with persistence gene mutations at day 30 (HR, 5.198; P=0.013) were independent risk factors for disease progression. In conclusion, our data indicated that monitoring MRD by FCM combined with gene mutation clearance at day 30 could help in the prediction of disease progression for MDS patients after transplantation.

miR-638 in circulating leukaemia cells as a non-invasive biomarker in diagnosis, treatment response and MRD surveillance of acute promyelocytic leukaemia

BACKGROUND

MicroRNA (miRNA) expression has been implicated in leukaemia. In recent years, miRNAs have been under investigation for their potential as non-invasive biomarkers in acute promyelocytic leukaemia (APL). We investigated whether miR-638 in circulating leukaemia cells is a non-invasive biomarker in diagnosis, assessment of the treatment response and minimal residual disease (MRD) surveillance of APL.

METHODS

Sixty cases of acute myeloid leukaemia (AML), including 30 cases of APL and 30 cases of non-APL AML, were selected. Thirty healthy controls were also selected. Bone marrow (BM) and peripheral blood (PB) samples were collected from APL patients at diagnosis and post-induction. Microarray analysis and quantitative real-time PCR (qRT-PCR) were performed for miRNA profiling and miR-638 expression analysis, respectively. For statistical analysis, Mann-Whitney U test, Wilcoxon Signed Rank test, receiver operating characteristic (ROC) curve analysis and Spearman's rho correlation test were used.

RESULTS

Both microarray and qRT-PCR data showed that miR-638 was significantly upregulated in BM after APL patients received induction therapy. Moreover, miR-638, which is specifically downregulated in APL cell lines, was upregulated after all-trans retinoic acid (ATRA)-induced myeloid differentiation. Receiver operating characteristic (ROC) curve analyses revealed that miR-638 could serve as a valuable biomarker for differentiating APL from controls or non-APL AML. Furthermore, miR-638 expression was sharply increased after induction therapy and complete remission (CR). An inverse correlation was observed between miR-638 and PML-RARα transcripts levels in BM samples, while a positive correlation was revealed between PB miR-638 and BM miR-638 levels in APL patients after induction therapy.

CONCLUSIONS

Our study suggested that miR-638 may serve as a potential APL biomarker for diagnosis and assessment of the response to targeted therapy, and PB miR-638 could be used for non-invasive MRD surveillance in APL.

How artificial intelligence might disrupt diagnostics in hematology in the near future

Artificial intelligence (AI) is about to make itself indispensable in the health care sector. Examples of successful applications or promising approaches range from the application of pattern recognition software to pre-process and analyze digital medical images, to deep learning algorithms for subtype or disease classification, and digital twin technology and in silico clinical trials. Moreover, machine-learning techniques are used to identify patterns and anomalies in electronic health records and to perform ad-hoc evaluations of gathered data from wearable health tracking devices for deep longitudinal phenotyping. In the last years, substantial progress has been made in automated image classification, reaching even superhuman level in some instances. Despite the increasing awareness of the importance of the genetic context, the diagnosis in hematology is still mainly based on the evaluation of the phenotype. Either by the analysis of microscopic images of cells in cytomorphology or by the analysis of cell populations in bidimensional plots obtained by flow cytometry. Here, AI algorithms not only spot details that might escape the human eye, but might also identify entirely new ways of interpreting these images. With the introduction of high-throughput next-generation sequencing in molecular genetics, the amount of available information is increasing exponentially, priming the field for the application of machine learning approaches. The goal of all the approaches is to allow personalized and informed interventions, to enhance treatment success, to improve the timeliness and accuracy of diagnoses, and to minimize technically induced misclassifications. The potential of AI-based applications is virtually endless but where do we stand in hematology and how far can we go?

Clinical implications of sequential MRD monitoring by NGS at 2 time points after chemotherapy in patients with AML

Next-generation sequencing (NGS) has been applied to measurable/minimal residual disease (MRD) monitoring after induction chemotherapy in patients with acute myeloid leukemia (AML), but the optimal time point for the test remains unclear. In this study, we aimed to investigate the clinical significance of NGS MRD at 2 different time points. We performed targeted NGS of 54 genes in bone marrow cells serially obtained at diagnosis, first complete remission (first time point), and after the first consolidation chemotherapy (second time point) from 335 de novo AML patients. Excluding DNMT3A, TET2, and ASXL1 mutations, which are commonly present in individuals with clonal hematopoiesis of indeterminate potential, MRD could be detected in 46.4% of patients at the first time point (MRD1st), and 28.9% at the second time point (MRD2nd). The patients with detectable NGS MRD at either time point had a significantly higher cumulative incidence of relapse and shorter relapse-free survival and overall survival. In multivariate analysis, MRD1st and MRD2nd were both independent poor prognostic factors. However, the patients with positive MRD1st but negative MRD2nd had a similar good prognosis as those with negative MRD at both time points. The incorporation of multiparameter flow cytometry and NGS MRD revealed that the presence of NGS MRD predicted poorer prognosis among the patients without detectable MRD by multiparameter flow cytometry at the second time point but not the first time point. In conclusion, the presence of NGS MRD, especially after the first consolidation therapy, can help predict the clinical outcome of AML patients.

Artificial Intelligence in Hematology: Current Challenges and Opportunities

PURPOSE OF REVIEW

Artificial intelligence (AI), and in particular its subcategory machine learning, is finding an increasing number of applications in medicine, driven in large part by an abundance of data and powerful, accessible tools that have made AI accessible to a larger circle of investigators.

RECENT FINDINGS

AI has been employed in the analysis of hematopathological, radiographic, laboratory, genomic, pharmacological, and chemical data to better inform diagnosis, prognosis, treatment planning, and foundational knowledge related to benign and malignant hematology. As more widespread implementation of clinical AI nears, attention has also turned to the effects this will have on other areas in medicine. AI offers many promising tools to clinicians broadly, and specifically in the practice of hematology. Ongoing research into its various applications will likely result in an increasing utilization of AI by a broader swath of clinicians.

Application of machine learning in the management of acute myeloid leukemia: current practice and future prospects

Machine learning (ML) is rapidly emerging in several fields of cancer research. ML algorithms can deal with vast amounts of medical data and provide a better understanding of malignant disease. Its ability to process information from different diagnostic modalities and functions to predict prognosis and suggest therapeutic strategies indicates that ML is a promising tool for the future management of hematologic malignancies; acute myeloid leukemia (AML) is a model disease of various recent studies. An integration of these ML techniques into various applications in AML management can assure fast and accurate diagnosis as well as precise risk stratification and optimal therapy. Nevertheless, these techniques come with various pitfalls and need a strict regulatory framework to ensure safe use of ML. This comprehensive review highlights and discusses recent advances in ML techniques in the management of AML as a model disease of hematologic neoplasms, enabling researchers and clinicians alike to critically evaluate this upcoming, potentially practice-changing technology.

Computational analysis of flow cytometry data in hematological malignancies: future clinical practice?

PURPOSE OF REVIEW

This review outlines the advancements that have been made in computational analysis for clinical flow cytometry data in hematological malignancies.

RECENT FINDINGS

In recent years, computational analysis methods have been applied to clinical flow cytometry data of hematological malignancies with promising results. Most studies combined dimension reduction (principle component analysis) or clustering methods (FlowSOM, generalized mixture models) with machine learning classifiers (support vector machines, random forest). For diagnosis and classification of hematological malignancies, many studies have reported results concordant with manual expert analysis, including B-cell chronic lymphoid leukemia detection and acute leukemia classification. Other studies, e.g. concerning diagnosis of myelodysplastic syndromes and classification of lymphoma, have shown to be able to increase diagnostic accuracy. With respect to treatment response monitoring, studies have focused on, for example, computational minimal residual disease detection in multiple myeloma and posttreatment classification of healthy or diseased in acute myeloid leukemia. The results of these studies are encouraging, although accurate relapse prediction remains challenging. To facilitate clinical implementation, collaboration and (prospective) validation in multicenter setting are necessary.

SUMMARY

Computational analysis methods for clinical flow cytometry data hold the potential to increase ease of use, objectivity and accuracy in the clinical work-up of hematological malignancies.

Liquid biopsy technologies for hematological diseases

Since the discovery of circulating tumor cells in 1869, technological advances in studying circulating biomarkers from patients' blood have made the diagnosis of nonhematologic cancers less invasive. Technological advances in the detection and analysis of biomarkers provide new opportunities for the characterization of other disease types. When compared with traditional biopsies, liquid biopsy markers, such as exfoliated bladder cancer cells, circulating cell-free DNA (cfDNA), and extracellular vesicles (EV), are considered more convenient than conventional biopsies. Liquid biopsy markers undoubtedly have the potential to influence disease management and treatment dynamics. Our main focuses of this review will be the cell-based, gene-based, and protein-based key liquid biopsy markers (including EV and cfDNA) in disease detection, and discuss the research progress of these biomarkers used in conjunction with liquid biopsy. First, we highlighted the key technologies that have been broadly adopted used in hematological diseases. Second, we introduced the latest technological developments for the specific detection of cardiovascular disease, leukemia, and coronavirus disease. Finally, we concluded with perspectives on these research areas, focusing on the role of microfluidic technology and artificial intelligence in point-of-care medical applications. We believe that the noninvasive capabilities of these technologies have great potential in the development of diagnostics and can influence treatment options, thereby advancing precision disease management.

Machine learning in haematological malignancies

Machine learning is a branch of computer science and statistics that generates predictive or descriptive models by learning from training data rather than by being rigidly programmed. It has attracted substantial attention for its many applications in medicine, both as a catalyst for research and as a means of improving clinical care across the cycle of diagnosis, prognosis, and treatment of disease. These applications include the management of haematological malignancy, in which machine learning has created inroads in pathology, radiology, genomics, and the analysis of electronic health record data. As computational power becomes cheaper and the tools for implementing machine learning become increasingly democratised, it is likely to become increasingly integrated into the research and practice landscape of haematology. As such, machine learning merits understanding and attention from researchers and clinicians alike. This narrative Review describes important concepts in machine learning for unfamiliar readers, details machine learning's current applications in haematological malignancy, and summarises important concepts for clinicians to be aware of when appraising research that uses machine learning.

A Knowledge-Reserved Distillation with Complementary Transfer for Automated FC-based Classification Across Hematological Malignancies

Acute leukemia often comes with life-threatening prognosis outcome and remains a critical clinical issue today. The implementation of measurable residual disease (MRD) using flow cytometry (FC) is highly effective but the interpretation is time-consuming and suffers from physician idiosyncrasy. Recent machine learning algorithms have been proposed to automatically classify acute leukemia samples with and without MRD to address this clinical need. However, most prior works either validate only on a small data cohort or focus on one specific type of leukemia which lacks generalization. In this work, we propose a transfer learning approach in performing automatic MRD classification that takes advantage of a large scale acute myeloid leukemia (AML) database to facilitate better learning on a small cohort of acute lymphoblastic leukemia (ALL). Specifically, we develop a knowledge-reserved distilled AML pre-trained network with ALL complementary learning to enhance the ALL MRD classification. Our framework achieves 84.5% averaged AUC which shows its transferability across acute leukemia, and our further analysis reveals that younger and elder ALL patient samples benefit more from using the pre-trained AML model.

Machine learning and artificial intelligence in haematology

Digitalization of the medical record and integration of genomic methods into clinical practice have resulted in an unprecedented wealth of data. Machine learning is a subdomain of artificial intelligence that attempts to computationally extract meaningful insights from complex data structures. Applications of machine learning in haematological scenarios are steadily increasing. However, basic concepts are often unfamiliar to clinicians and investigators. The purpose of this review is to provide readers with tools to interpret and critically appraise machine learning literature. We begin with the elucidation of standard terminology and then review examples in haematology. Guidelines for designing and evaluating machine-learning studies are provided. Finally, we discuss limitations of the machine-learning approach.

A Clinically Applicable Approach to the Classification of B-Cell Non-Hodgkin Lymphomas with Flow Cytometry and Machine Learning

The immunophenotype is a key element to classify B-cell Non-Hodgkin Lymphomas (B-NHL); while it is routinely obtained through immunohistochemistry, the use of flow cytometry (FC) could bear several advantages. However, few FC laboratories can rely on a long-standing practical experience, and the literature in support is still limited; as a result, the use of FC is generally restricted to the analysis of lymphomas with bone marrow or peripheral blood involvement. In this work, we applied machine learning to our database of 1465 B-NHL samples from different sources, building four artificial predictive systems which could classify B-NHL in up to nine of the most common clinico-pathological entities. Our best model shows an overall accuracy of 92.68%, a mean sensitivity of 88.54% and a mean specificity of 98.77%. Beyond the clinical applicability, our models demonstrate (i) the strong discriminatory power of MIB1 and Bcl2, whose integration in the predictive model significantly increased the performance of the algorithm; (ii) the potential usefulness of some non-canonical markers in categorizing B-NHL; and (iii) that FC markers should not be described as strictly positive or negative according to fixed thresholds, but they rather correlate with different B-NHL depending on their level of expression.

Measurable Residual Disease in Acute Myeloid Leukemia Using Flow Cytometry: A Review of Where We Are and Where We Are Going

The detection of measurable residual disease (MRD) has become a key investigation that plays a role in the prognostication and management of several hematologic malignancies. Acute myeloid leukemia (AML) is the most common acute leukemia in adults and the role of MRD in AML is still emerging. Prognostic markers are complex, largely based upon genetic and cytogenetic aberrations. MRD is now being incorporated into prognostic models and is a powerful predictor of relapse. While PCR-based MRD methods are sensitive and specific, many patients do not have an identifiable molecular marker. Immunophenotypic MRD methods using multiparametric flow cytometry (MFC) are widely applicable, and are based on the identification of surface marker combinations that are present on leukemic cells but not normal hematopoietic cells. Current techniques include a "different from normal" and/or a "leukemia-associated immunophenotype" approach. Limitations of MFC-based MRD analyses include the lack of standardization, the reliance on a high-quality marrow aspirate, and variable sensitivity. Emerging techniques that look to improve the detection of leukemic cells use dimensional reduction analysis, incorporating more leukemia specific markers and identifying leukemic stem cells. This review will discuss current methods together with new and emerging techniques to determine the role of MFC MRD analysis.

Automated quantitative analysis of Ki-67 staining and HE images recognition and registration based on whole tissue sections in breast carcinoma

Background

The scoring of Ki-67 is highly relevant for the diagnosis, classification, prognosis, and treatment in breast invasive ductal carcinoma (IDC). Traditional scoring method of Ki-67 staining followed by manual counting, is time-consumption and inter−/intra observer variability, which may limit its clinical value. Although more and more algorithms and individual platforms have been developed for the assessment of Ki-67 stained images to improve its accuracy level, most of them lack of accurate registration of immunohistochemical (IHC) images and their matched hematoxylin-eosin (HE) images, or did not accurately labelled each positive and negative cell with Ki-67 staining based on whole tissue sections (WTS). In view of this, we introduce an accurate image registration method and an automatic identification and counting software of Ki-67 based on WTS by deep learning.

Methods

We marked 1017 breast IDC whole slide imaging (WSI), established a research workflow based on the (i) identification of IDC area, (ii) registration of HE and IHC slides from the same anatomical region, and (iii) counting of positive Ki-67 staining.

Results

The accuracy, sensitivity, and specificity levels of identifying breast IDC regions were 89.44, 85.05, and 95.23%, respectively, and the contiguous HE and Ki-67 stained slides perfectly registered. We counted and labelled each cell of 10 Ki-67 slides as standard for testing on WTS, the accuracy by automatic calculation of Ki-67 positive rate in attained IDC was 90.2%. In the human-machine competition of Ki-67 scoring, the average time of 1 slide was 2.3 min with 1 GPU by using this software, and the accuracy was 99.4%, which was over 90% of the results provided by participating doctors.

Conclusions

Our study demonstrates the enormous potential of automated quantitative analysis of Ki-67 staining and HE images recognition and registration based on WTS, and the automated scoring of Ki67 can thus successfully address issues of consistency, reproducibility and accuracy. We will provide those labelled images as an open-free platform for researchers to assess the performance of computer algorithms for automated Ki-67 scoring on IHC stained slides.

From Bivariate to Multivariate Analysis of Cytometric Data: Overview of Computational Methods and Their Application in Vaccination Studies

Flow and mass cytometry are used to quantify the expression of multiple extracellular or intracellular molecules on single cells, allowing the phenotypic and functional characterization of complex cell populations. Multiparametric flow cytometry is particularly suitable for deep analysis of immune responses after vaccination, as it allows to measure the frequency, the phenotype, and the functional features of antigen-specific cells. When many parameters are investigated simultaneously, it is not feasible to analyze all the possible bi-dimensional combinations of marker expression with classical manual analysis and the adoption of advanced automated tools to process and analyze high-dimensional data sets becomes necessary. In recent years, the development of many tools for the automated analysis of multiparametric cytometry data has been reported, with an increasing record of publications starting from 2014. However, the use of these tools has been preferentially restricted to bioinformaticians, while few of them are routinely employed by the biomedical community. Filling the gap between algorithms developers and final users is fundamental for exploiting the advantages of computational tools in the analysis of cytometry data. The potentialities of automated analyses range from the improvement of the data quality in the pre-processing steps up to the unbiased, data-driven examination of complex datasets using a variety of algorithms based on different approaches. In this review, an overview of the automated analysis pipeline is provided, spanning from the pre-processing phase to the automated population analysis. Analysis based on computational tools might overcame both the subjectivity of manual gating and the operator-biased exploration of expected populations. Examples of applications of automated tools that have successfully improved the characterization of different cell populations in vaccination studies are also presented.

Machine Learning of Discriminative Gate Locations for Clinical Diagnosis

High-throughput single-cell cytometry technologies have significantly improved our understanding of cellular phenotypes to support translational research and the clinical diagnosis of hematological and immunological diseases. However, subjective and ad hoc manual gating analysis does not adequately handle the increasing volume and heterogeneity of cytometry data for optimal diagnosis. Prior work has shown that machine learning can be applied to classify cytometry samples effectively. However, many of the machine learning classification results are either difficult to interpret without using characteristics of cell populations to make the classification, or suboptimal due to the use of inaccurate cell population characteristics derived from gating boundaries. To date, little has been done to optimize both the gating boundaries and the diagnostic accuracy simultaneously. In this work, we describe a fully discriminative machine learning approach that can simultaneously learn feature representations (e.g., combinations of coordinates of gating boundaries) and classifier parameters for optimizing clinical diagnosis from cytometry measurements. The approach starts from an initial gating position and then refines the position of the gating boundaries by gradient descent until a set of globally-optimized gates across different samples are achieved. The learning procedure is constrained by regularization terms encoding domain knowledge that encourage the algorithm to seek interpretable results. We evaluate the proposed approach using both simulated and real data, producing classification results on par with those generated via human expertise, in terms of both the positions of the gating boundaries and the diagnostic accuracy. © 2019 The Authors. Cytometry Part A published by Wiley Periodicals, Inc. on behalf of International Society for Advancement of Cytometry.

Guidelines for the use of flow cytometry and cell sorting in immunological studies (second edition)

These guidelines are a consensus work of a considerable number of members of the immunology and flow cytometry community. They provide the theory and key practical aspects of flow cytometry enabling immunologists to avoid the common errors that often undermine immunological data. Notably, there are comprehensive sections of all major immune cell types with helpful Tables detailing phenotypes in murine and human cells. The latest flow cytometry techniques and applications are also described, featuring examples of the data that can be generated and, importantly, how the data can be analysed. Furthermore, there are sections detailing tips, tricks and pitfalls to avoid, all written and peer-reviewed by leading experts in the field, making this an essential research companion.