Available online 15 February 2023

Author links open overlay panel, , , , , , Abstract

Digital pathology has a crucial role in diagnostic pathology and is increasingly a technological requirement in the field. Integration of digital slides into the pathology workflow, advanced algorithms, and computer-aided diagnostic techniques extend the frontiers of the pathologist's view beyond the microscopic slide and enable true integration of knowledge and expertise. There is clear potential for artificial intelligence (AI) breakthroughs in pathology and hematopathology.

In this review article, we discuss the approach of using machine learning in the diagnosis, classification, and treatment guidelines of hematolymphoid disease, as well as recent progress of artificial intelligence in flow cytometric analysis of hematolymphoid diseases. We review these topics specifically through the potential clinical applications of CellaVision, an automated digital image analyzer of peripheral blood, and Morphogo, a novel artificial intelligence-based bone marrow analyzing system. Adoption of these new technologies will allow pathologists to streamline workflow and achieve faster turnaround time in diagnosing hematological disease.

Section snippetsBackground

Recent advances in artificial intelligence (AI) and data science are having a profound impact on many industries, including medicine1. Currently, AI does not have a large presence in hematopathology. However, clinical hematology and hematopathology are well-positioned to take advantage of this technological revolution to improve routine patient care1. There are multiple benefits associated with the adoption of digital pathology. First, pathologists can review cases from different sites at a

Machine learning in the diagnosis, classification, and treatment guidelines of hematolymphoid disease

ML, which is a subdomain of AI, attempts to extract meaningful patterns and associations within complex data12. ML can perform a task without receiving explicit instructions, often with little human intervention and surpassing human capability3,13. In diagnosing hematolymphoid disease, ML has impacted cytomorphology, cytogenetics, immunophenotyping, and molecular genetics12,13. ML algorithms have been used to generate a differential diagnosis, rank therapeutic options, and predict outcomes13.

CellaVision

Cell-locating devices that automate the WBC differential include the CellaVision systems and the EasyCell assistant. The Bloodhound™ Integrated Hematology System, manufactured by Constitution Medical, Inc., is both an automated differential cell counter and a cell-locating device. The DI‐60 Integrated Slide Processing System (Sysmex, Kobe, Japan) was the first fully integrated cell image analyzer on the market. It consists of two Sysmex XN Hematology Analyzers, the Sysmex slide-making/staining

Morphogo

The examination of bone marrow aspirate smear is a key diagnostic procedure in the workup of hematological diseases such as AML, MDS, and multiple myeloma, among others. However, manual microscopy, the gold standard for generating a manual bone marrow differential count has many disadvantages previously discussed, such as its time-consuming nature and intraobserver variability. In addition, to share a cell of interest with a colleague, a pathologist must use a micro-slide field finder to

AI in flow cytometric analysis of hematolymphoid diseases

Cells are generally classified either by analysis of microscopic images of cells in cytomorphology or by analysis of cell populations in bidimensional plots obtained by flow cytometry32. Multiparameter flow cytometry is a key method used in the diagnoses of leukemias and lymphoma32,33. Despite increasing awareness of the importance of genetic context, diagnoses in hematology are still mainly based on phenotypic evaluation. To identify and classify cell populations of interest by flow

Discussion

In the hematology lab today, uses of AI and ML are limited. Approved devices are mostly restricted to the morphologic analysis of blood smears. Digital imaging has allowed for faster, more efficient, and standardized ways of performing morphological analysis of peripheral blood smears and classifying hematological cells17,35. For a subset of leukocytes, digital system classification correlates well with the gold standard, manual microscopy. Yet, the technology falls short of the gold standard

Conclusion

In the future, we envision a scenario where AI-based algorithms can help integrate complex data in efficient and novel ways with practical applications in patient care. Such a model is a long way off, but it is certainly worth discussion and further study. Flow cytometry, which has often relied much on user capability and therefore is prone to error, can be standardized and made into a powerful tool for the diagnosis of hematopoietic malignancies. Systems such as CellaVision and Morphogo have

Declaration of Competing Interest

Dr. M Chen received research grant support from ALAB and served on the advisory board as a consultant of Zhiwei Tech.

References (35)W Walter et al.How artificial intelligence might disrupt diagnostics in hematology in the near future

Oncogene

(2021)

K Bera et al.Artificial intelligence in digital pathology - new tools for diagnosis and precision oncology

Nat Rev Clin Oncol

(2019)

N Radakovich et al.Artificial Intelligence in Hematology: Current Challenges and Opportunities

Curr Hematol Malig Rep

(2020)

A Echle et al.Deep learning in cancer pathology: a new generation of clinical biomarkers

Br J Cancer

(2021)

J Krause et al.Deep learning detects genetic alterations in cancer histology generated by adversarial networks

J Pathol

(2021)

KH Yu et al.Artificial intelligence in healthcare

Nat Biomed Eng

(2018)

EJ. TopolHigh-performance medicine: the convergence of human and artificial intelligence

Nat Med

(2019)

View full text

© 2023 Elsevier Inc. All rights reserved.