Deep learning applications in visual data for benign and malignant hematologic conditions: a systematic review and...

Area of Science:

• Hematology
• Artificial Intelligence
• Medical Imaging

Background:

• Deep learning (DL) algorithms excel at analyzing complex patterns in imaging data.
• DL has demonstrated significant potential in medical diagnostics and prognostics.
• Emerging DL applications in hematology offer novel research and clinical opportunities.

Approach:

• This narrative review provides a visual glossary of DL principles.
• It systematically reviews DL applications in malignant and non-malignant hematologic conditions.
• The review is organized by clinical care phases to aid understanding.

Key Points:

• DL's ability to interpret subtle graphical features enables accurate predictions.
• Understanding DL concepts and potential pitfalls is crucial for practicing hematologists.
• The review highlights key literature and considerations for DL implementation in clinical practice.

Conclusions:

• DL is a rapidly advancing field with substantial potential in hematology.
• Familiarity with DL is becoming essential for hematologists.
• Critical understanding of DL development and implementation is necessary for its effective clinical use.