AI-Powered Model Enhances Early Screening for Rare Genetic Disease
đ§ Introduction
Early detection of rare diseases like Mucopolysaccharidosis (MPS) can significantly improve treatment outcomesâbut diagnosis is often delayed. Researchers supported by APMAD have developed a machine learningâbased screening approach using UAE health records to accelerate MPS identification.
đ Research Summary
The study evaluated various machine learning (ML) models for early MPS diagnosis using electronic health records (EHR) from SEHA, spanning 2004â2022. The dataset included 115 patient records aged 19 years or younger. Using nested cross-validation and multiple feature selection strategies, the team identified the most accurate model: Naive Bayes trained on features selected by domain experts.
This model achieved impressive metrics, including 93% accuracy, 96% AUC, and 91% F1-score. Interpretability techniquesâSHAP and LIMEâhighlighted key diagnostic features such as acute gingivitis, dental caries, and respiratory infections, offering insights into clinical relevance.
đ Broader Impact
This research introduces a cost-effective, non-invasive diagnostic aid for MPS, leveraging real-world EHR data and advanced AI techniques. The model supports early identification of a rare disease often missed in standard diagnostics, aligning with APMADâs mission to integrate digital tools into precision medicine.
It also showcases the potential of UAEâs healthcare data to inform innovative screening strategies and improve clinical decision-making in rare genetic disorders.
đ Reference
@article{AlShehhi2025,
title = {Comparison of machine learning models for mucopolysaccharidosis early diagnosis using UAE medical records},
volume = {15},
ISSN = {2045-2322},
url = {http://dx.doi.org/10.1038/s41598-025-13879-3},
DOI = {10.1038/s41598-025-13879-3},
number = {1},
journal = {Scientific Reports},
publisher = {Springer Science and Business Media LLC},
author = {AlShehhi, Aamna and Alblooshi, Hiba and Fadul, Ruba and Tumzghi, Natnael and Tenaiji, Amal Al and Harbi, Mariam Al and Al-Jasmi, Fatma},
year = {2025},
month = aug
}



Leave a Reply