Machine Learning and Artificial Intelligence-Based Clinical Decision Support for Modern Hematology.

Area of Science:

• Medical Informatics
• Clinical Pathology
• Computational Biology

Background:

• Hematology is a data-rich medical field.
• Technological innovation is rapidly advancing hematology.
• Machine learning (ML) integration into diagnostics is increasing.

Purpose of the Study:

• To review the current landscape of ML research and clinical applications in hematology.
• To inform healthcare professionals about ML's impact on workflows.
• To explore successes and limitations of ML in hematology.

Main Methods:

• Literature review of ML in hematology.
• Analysis of research across various hematology subdomains.
• Examination of clinical applications and deployments.

Main Results:

• ML shows significant potential in diverse hematology areas.
• Successes are noted in hematopathology, hemoglobinopathies, and coagulopathy.
• Limitations and challenges in ML research and deployment are identified.

Conclusions:

• Understanding ML is crucial for hematology care teams.
• ML tools require careful integration into clinical workflows.
• Further research is needed to address ML limitations in hematology.