Related Experiment Video
Updated: Dec 23, 2025
06:41
In Vivo Functional Study of Disease-associated Rare Human Variants Using Drosophila
Published on: August 20, 2019
14.1K
Diagnosis support systems for rare diseases: a scoping review
Carole Faviez1, Xiaoyi Chen2, Nicolas Garcelon2,3
1Centre de Recherche des Cordeliers, INSERM, Université de Paris, Sorbonne Université, F-75006, Paris, France. carole.faviez@inserm.fr.
Orphanet Journal of Rare Diseases
|April 18, 2020
Summary
Computerized systems aid rare disease diagnosis, leveraging health data and machine learning. While promising, varied approaches hinder direct comparison, necessitating standardization for validation and reproducibility.
Area of Science:
- Medical Informatics
- Genomics
- Clinical Decision Support
Background:
- Rare diseases impact 350 million globally, often facing diagnostic delays due to limited clinician knowledge and specialized centers.
- Computerized diagnosis support systems (CDSS) are crucial for rare diseases, utilizing health data and expertise.
- Existing CDSS often rely on phenotype concepts, images, or fluids for diagnosis.
Purpose of the Study:
- To review and analyze initiatives focused on developing computerized systems for rare disease diagnosis.
- To understand the methodologies and data sources employed in these diagnostic support systems.
Main Methods:
- A scoping review was conducted following Arksey and O'Malley's methodology.
- A data charting form was developed to systematically analyze and categorize 68 relevant studies.
Main Results:
- Sixty-eight studies were analyzed, targeting single or multiple rare diseases.
- Machine learning algorithms (66%) and expert knowledge (57%) were common, with some using simple similarities or manual methods.
- Most systems showed satisfactory performance, with 14 offering online tools, often for broad rare disease diagnosis using phenotype data.
Conclusions:
- Emerging CDSS show promising preliminary results for rare disease diagnosis.
- Variability in methodologies and evaluation complicates result comparison.
- Standardization in validation, reproducibility, and explainability is essential for advancing these tools.
