ScholarDeck

AI / informaticsNPJ digital medicine · 2026

AI-driven diagnostic algorithm enhances early detection of paroxysmal nocturnal hemoglobinuria in real-world settings.

Robert Dewor, Michal J Dabrowski, Łukasz Więcek … Grzegorz Basak

Read paper

HematologyAI / informatics

AI screening identified PNH in 10.92% of high-risk patients referred for testing.

AI-driven diagnostic algorithm enhances early detection of paroxysmal nocturnal hemoglobinuria in real-world settings.

Robert Dewor et al. · NPJ digital medicine · 2026

Methods

Retrospective analysis revealed potentially preventable diagnostic delays of 74-1337 days.

n = 1,307,140 patients

Results

PNH was diagnosed in high-risk patients referred for testing by AI screening

positive predictive va

10.92%

conventional screening

6.9%

More results

Paroxysmal nocturnal haemoglobinuria (PNH) is a rare, life-threatening hematologic disease with diagnostic delays exceeding 5 years in 24% of cases.

We developed and deployed an artificial intelligence algorithm analyzing structured and unstructured electronic health record data across 14 healthcare organizations in Poland.

More results

High-risk patients were significantly older (median 69.5 years) with elevated rates of fatigue (76.4% vs 29.19%), anaemia (72.2% vs 7.61%), and myelodysplastic syndrome (49.2% vs 0.24%; all p < 0.001).

Only 2.25% presented with haemoglobinuria versus 45-62% in registry cohorts.

“

Conclusion

Monte Carlo feature selection identified Coombs-negative haemolysis and visit frequency as strongest predictors, supporting the potential utility of AI-assisted screening for identifying atypical PNH presentations.

Post

AI-driven diagnostic algorithm enhances early detection of paroxysmal nocturnal hemoglobinuria in real-world settings.

AI screening identified PNH in 10.92% of high-risk patients referred for testing.

Related papers