What type of study is this?

This is a Meta-Analysis study (also classified as: Systematic Review).

What question did this study set out to answer?

This research evaluates the diagnostic accuracy of AI/ML models for detecting sickle cell disease (SCD) in blood smears.

December 8, 2025BloodOpen Access

View Full Paper

Ask AI

Bookmark

Authors

Meta-analysis reveals high diagnostic accuracy for AI/ML in sickle cell disease detection, suggesting robust models with needs for validation.

This research evaluates the diagnostic accuracy of AI/ML models for detecting sickle cell disease (SCD) in blood smears.
Conducted a systematic search of MEDLINE, EMBASE, and CENTRAL for ML-based SCD diagnosis studies.
Performed meta-analysis using Stata 18.0 and assessed risk of bias with QUADAS-2.
Reviewed 800 studies with 21 meeting the inclusion criteria, analyzing 25,428 images.
AI/ML models showed high accuracy for SCD detection, exceeding 98% in many cases.
A deep convolutional neural network achieved over 99% accuracy with a dataset of 30 images.
Reported sensitivity and specificity were 96.4% and 98.3%, respectively.

Aneesh Sai Reddy Gurajala (2025) studied this question.

View Full Paper

Ask AI

Bookmark

Meta-analysis reveals high diagnostic accuracy for AI/ML in sickle cell disease detection, suggesting robust models with needs for validation.