Related Experiment Video For Artificial intelligence
Updated: Jan 16, 2026
Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
Artificial intelligence-assisted enhancement of diagnostic accuracy and efficiency in detecting cervical lymph node
Chih-Huang Tseng1,2,3, Chang-Wei Su1,4, Yu-Min Lin1,4
1School of Dentistry, College of Dental Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan.
Background/Purpose:
Cervical lymph node metastasis represents a critical prognostic factor in oral squamous cell carcinoma (SCC); however, early-stage or subtle metastases often pose diagnostic challenges under conventional microscopy. Recent advancements in artificial intelligence (AI) offer promising solutions to enhance diagnostic accuracy and efficiency. This study aimed to evaluate the extent to which AI-assisted tools can improve diagnostic performance and efficiency in the detection of cervical lymph node metastases.
Materials And Methods:
Sixty-six hematoxylin-eosin-stained slides containing 621 lymph nodes from oral SCC cases were digitized. Four participants (two oral pathologists, one postgraduate year (PGY) resident, one fourth-year dental student) reviewed slides with and without the AI-assistant tool. Diagnostic accuracy and interpretation time were compared.
Results:
AI assistance significantly improved diagnostic accuracy and efficiency across different participants. False positives and false negatives decreased notably, especially for junior participants. Review time was also significantly shortened for negative and macrometastatic slides (P < 0.0001 and P < 0.05, respectively), with the greatest benefit seen among less-experienced participants.
Conclusion:
The AI-assisted tool improved diagnostic accuracy and efficiency in detecting cervical lymph node metastases in oral SCC. It may serve as a preliminary screening tool and a valuable educational aid for training junior pathologists, underscoring its potential for broader application in digital pathology.
