Related Experiment Video
Updated: May 20, 2025
06:38
In Vitro Modeling of Down Syndrome Neurogenesis Using Human-Induced Pluripotent Stem Cells
Published on: March 7, 2025
287
A Review of Artificial Intelligence-Based Down Syndrome Detection Techniques.
Mujeeb Ahmed Shaikh1,2, Hazim Saleh Al-Rawashdeh3, Abdul Rahaman Wahab Sait4
1Department of Basic Medical Science, College of Medicine, AlMaarefa University, Diriyah 13713, Riyadh, Saudi Arabia.
Life (Basel, Switzerland)
|March 27, 2025
Summary
Artificial intelligence (AI) and machine learning (ML) show promise for diagnosing Down syndrome (DS) using various data types. However, dataset limitations impact AI model generalizability, necessitating strategies for clinical integration.
Area of Science:
- Medical Diagnostics
- Artificial Intelligence
- Genetics
Background:
- Down syndrome (DS) is a common chromosomal abnormality impacting global health.
- AI and ML are increasingly used to improve DS diagnostic accuracy.
- A comprehensive evaluation of AI's impact on DS diagnostics is lacking.
Purpose of the Study:
- To identify AI methodologies and technologies for DS diagnostics.
- To evaluate AI model performance using standard metrics.
- To highlight the strengths and limitations of AI in DS diagnosis.
Main Methods:
- Systematic review adhering to PRISMA guidelines.
- Extensive literature search across major academic databases.
- Selection of 25 relevant articles based on inclusion/exclusion criteria.
Main Results:
- AI demonstrates significant advancements in DS diagnostics across facial images, ultrasound, and genetic data.
- AI approaches show strong potential for early Down syndrome detection.
- Limitations include small, imbalanced datasets affecting AI model generalizability.
Conclusions:
- AI offers promising tools for early Down syndrome diagnosis.
- Addressing dataset limitations is crucial for improving AI model performance.
- Actionable strategies are proposed to facilitate clinical adoption of AI in DS diagnostics.
