AI in Cervical Cancer Cytology Diagnostics: A Narrative Review of Cutting-Edge Studies
<b>Background:</b> The integration of artificial intelligence (AI) into cervical cancer diagnostics has shown promising advancements in recent years. AI technologies, particularly in the analysis of cytological images, offer potential improvements in diagnostic accuracy and screening eff...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-07-01
|
| Series: | Bioengineering |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2306-5354/12/7/769 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | <b>Background:</b> The integration of artificial intelligence (AI) into cervical cancer diagnostics has shown promising advancements in recent years. AI technologies, particularly in the analysis of cytological images, offer potential improvements in diagnostic accuracy and screening efficiency. However, challenges regarding model generalizability, explainability, and operational integration into clinical workflows persist, impeding widespread adoption. <b>Aim:</b> This narrative review aims to critically evaluate the current state of AI in cervical cancer diagnostic cytology, identifying trends, key developments, and areas requiring further research. It also explores the potential for AI to improve diagnostic processes, alongside examining international guidelines and consensus on its adoption. <b>Methods:</b> A narrative review was conducted through a comprehensive search of PubMed and Scopus databases. Thirty studies published between 2020 and 2025 were selected based on their relevance. <b>Results:</b> The literature review reveals a growing interest in the application of AI for cervical cancer diagnostics, particularly in the automated interpretation. However, large-scale clinical adoption remains limited. Most studies are experimental or application-based in controlled settings. Consensus efforts and specific recommendations for this domain are still limited and not specific. Key barriers include limited model generalizability, lack of explainability, challenges in integration into clinical workflows, and regulatory and infrastructural constraints. <b>Conclusions:</b> A sustainable and meaningful integration of AI in cervical cancer diagnostics requires a unified framework that addresses both technical challenges and operational needs, supported by context-specific strategies and broader consensus-building efforts. |
|---|---|
| ISSN: | 2306-5354 |