AI-augmented screening improves hereditary hemolytic anemia detection with 92.8% pooled sensitivityAI tools improve accuracy for screening hereditary blood disorders

Human mutation Published June 29, 2026 Study authors: Ali Naif Taleb, Abdullah Radfan Saleh, Mehdi Mansour Abdulnabi H, Ali Gamila Saleh, Ali Hana Mohsen,… PubMed ↗ DOI ↗ Editorial oversight: Dr. Amelia Tan, PhD · Internal Medicine & Chronic Disease

AI-generated summary of the cited source, checked by automated accuracy review. How we work

Key Takeaway

Consider AI-augmented tests as a means to improve HHA detection and reduce costs, but note potential regional biases.

This systematic review and meta-analysis evaluated the diagnostic accuracy of artificial intelligence (AI) augmented tests for identifying carriers of hereditary hemolytic anemia (HHA), including Thalassemia and Sickle Cell Disease, during premarital screening. The analysis included a large global population of 133,498 individuals across 23 countries to assess how AI-enhanced interpretation of complete blood count (CBC), blood smear, and erythrocyte sedimentation rate (ESR) compares to conventional methods.

The primary outcome was the diagnostic accuracy of AI-augmented tests for HHA carrier identification. The meta-analysis reported a pooled sensitivity of 92.8% for AI-augmented screening, which represents a 12.3% improvement over conventional interpretation (95% CI: 91.3%-94.1%; p < 0.001). Additionally, the pooled specificity for AI-augmented screening was reported at 91.5% (95% CI: 89.7%-93.0%).

Secondary outcomes highlighted specific technological and regional variations. Deep learning models demonstrated a sensitivity of 95.1%, while Explainable AI (XAI) showed a specificity of 94.3%. The integration of both CBC and blood smear data was associated with a 5.5% increase in specificity. From an economic and logistical standpoint, the use of AI-augmented tests was associated with a 23.7% reduction in confirmatory testing and cost savings of $8.50 per individual.

Geographic disparities were noted in the data. Sensitivity in Sub-Saharan Africa was reported at 86.5%, which was significantly lower than the sensitivity observed in the Middle East, which was recorded at 94.8% (p < 0.001). These findings suggest that while AI provides a robust framework for screening, regional variations may impact diagnostic performance.

Safety and tolerability data were not reported in this meta-analysis. However, the study noted significant methodological limitations. Specifically, 68% of the validation studies included in the analysis utilized research-grade samples rather than routine clinical samples. Furthermore, there was a lack of prospective clinic-to-algorithm validation to confirm real-world performance.

These results suggest that AI can enhance HHA screening and reduce costs. However, clinicians should note that algorithmic bias exists against certain genotypes in specific regions. To address these disparities and ensure equity, the study suggests that federated learning and XAI compliance are necessary for future implementation.

Several questions remain regarding the long-term integration of these tools into standard clinical workflows. Specifically, the impact of using non-research-grade samples on accuracy remains to be fully quantified in a prospective setting. While the data indicates significant potential for AI to improve screening and reduce costs, real-world deployment requires further validation to ensure consistency across diverse populations.

How this fits prior evidence

How this fits prior evidence: This meta-analysis addresses a gap in screening technology for conditions mentioned in previous reports, such as sickle cell disease. While previous coverage highlighted the high prevalence of sickle cell disease in sub-Saharan Africa and the development of gene therapies like reni-cel, this study focuses on the diagnostic accuracy of AI-augmented tools to identify carriers. The finding of lower sensitivity in Sub-Saharan Africa (86.5%) compared to the Middle East (94.8%) highlights specific regional challenges in screening for these conditions.

For many families, knowing if they carry a gene for a blood disorder is a vital step in planning for the future. Conditions like Sickle Cell Disease and Thalassemia are inherited, meaning parents need to know their status before starting a family. Currently, identifying these carriers involves looking at blood counts and cell shapes. However, human error or subtle variations can sometimes make it hard to catch every case during routine screening.

To address this, researchers looked at how artificial intelligence (AI) can help doctors read these tests more accurately. They analyzed data from over 133,000 people across 23 countries who were being screened for hereditary hemolytic anemia (HHA). This group included people looking for signs of conditions like sickle cell disease and thalassemia. The study compared traditional ways of reading blood tests against methods that use AI to help analyze the results.

The findings show that using AI-augmented tests significantly improved the ability to correctly identify carriers. Specifically, the sensitivity—the ability of a test to correctly identify those with the condition—rose by about 12% compared to traditional methods. The study also found that combining different types of blood data into one AI system helped make the results even more reliable. Beyond just accuracy, the researchers found that using these tools could reduce the need for expensive follow-up tests by nearly 24% and save about $8.50 in costs for each person screened.

While these results are promising, there are important things to keep in mind. Most of the data came from high-quality research samples rather than everyday clinic samples. Additionally, some regions, like Sub-Saharan Africa, showed lower accuracy rates compared to the Middle East, suggesting that AI tools might need more tuning to work equally well for everyone regardless of where they live. There is also a lack of long-term testing in real-world clinics.

For patients right now, this means that while AI isn't replacing doctors today, it shows a clear path toward making screening more reliable and affordable. It could eventually help ensure that fewer people miss a diagnosis due to human error or technical limitations. However, because the technology still needs more testing in everyday clinics, these tools are not yet the standard of care for every patient everywhere.

What this means for you:

AI can improve the accuracy of blood tests for hereditary disorders and lower costs, but more real-world testing is needed.

Study Details

Study typeMeta analysis

Sample sizen = 133,498

EvidenceLevel 1

PublishedJan 2026

PMID42369207

View Original Abstract ↓

Artificial intelligence (AI) augmentation of routine hematological tests offers a promising strategy to improve hereditary hemolytic anemia (HHA) carrier detection in premarital screening, especially in resource-limited settings. HHA in this review specifically encompasses -thalassemia, -thalassemia, sickle cell disease (including HbS and HbC variants), and other hemoglobinopathies with autosomal recessive inheritance patterns requiring carrier detection for prevention. These conditions share hemolytic phenotypes but differ in hematological signatures, necessitating separate subgroup analyses. This global systematic review and meta-analysis evaluated the diagnostic accuracy, equity implications, and implementation challenges of AI-augmented complete blood count (CBC), blood smear, and erythrocyte sedimentation rate (ESR) for HHA carrier identification. We systematically searched seven databases and included 85 studies ( = 133,498 participants, 23 countries). AI-augmented screening achieved a pooled sensitivity of 92.8% (95% CI: 91.3%-94.1%) and specificity of 91.5% (89.7%-93.0%), representing a 12.3% sensitivity improvement over conventional interpretation ( < 0.001). However, significant geographic disparities were observed: sensitivity in Sub-Saharan Africa was 86.5% compared with 94.8% in the Middle East ( < 0.001), partly due to algorithmic bias against African HbS/HbC variants and infrastructural barriers. Deep learning models achieved the highest sensitivity (95.1%), whereas explainable artificial intelligence (XAI) provided optimal specificity (94.3%). Integrating CBC with blood smear increased specificity by 5.5% at minimal additional cost. AI triage reduced confirmatory testing by 23.7%, saving $8.50 per individual. For equitable implementation, we recommend the following: (1) federated learning to include underrepresented genotypes, (2) WHO/CDC certification of affordable, offline-capable edge AI devices, and (3) mandatory XAI compliance with bias audits. AI can transform HHA screening, but deliberate efforts are needed to avoid exacerbating global health inequities. Importantly, 68% of validation studies used research-grade rather than routine clinical samples, and prospective clinic-to-algorithm validation remains a critical gap requiring urgent attention before real-world deployment.

AI-augmented screening improves hereditary hemolytic anemia detection with 92.8% pooled sensitivityAI tools improve accuracy for screening hereditary blood disorders

How this fits prior evidence

Study Details

Biomarkers and clinical factors show statistical associations with sepsis-induced coagulopathy in 37,459 patients

Your Blood May Signal a Deadly Clotting Crisis Before It Strikes

Clinical research that matters. Delivered to your inbox.

AI-augmented screening improves hereditary hemolytic anemia detection with 92.8% pooled sensitivityAI tools improve accuracy for screening hereditary blood disorders

How this fits prior evidence

More on Sickle Cell Disease

Study Details

Biomarkers and clinical factors show statistical associations with sepsis-induced coagulopathy in 37,459 patients

Your Blood May Signal a Deadly Clotting Crisis Before It Strikes

Clinical research that matters. Delivered to your inbox.

Related in Hematology

From Other Specialties