Imaging-Based AI for Predicting Lymphovascular Space Invasion in Cervical Cancer: Systematic Review and Meta-Analysis
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Artificial intelligence (AI) shows promise in detecting lymphovascular space invasion (LVSI) in cervical cancer, with deep learning models outperforming traditional machine learning. Further validation is needed to confirm AI
Area Of Science
- Oncology
- Medical Imaging
- Artificial Intelligence
Background
- The accuracy of artificial intelligence (AI) in detecting lymphovascular space invasion (LVSI) in cervical cancer is under debate.
- Accurate LVSI detection is critical for cervical cancer staging and treatment planning.
Purpose Of The Study
- To evaluate the diagnostic accuracy of imaging-based AI for predicting LVSI in cervical cancer through a meta-analysis.
- To synthesize existing evidence on AI's performance in identifying LVSI in cervical cancer patients.
Main Methods
- A comprehensive literature search was conducted across PubMed, Embase, and Web of Science up to November 9, 2024.
- Included studies evaluated the diagnostic performance of imaging-based AI models for LVSI detection in cervical cancer.
- A bivariate random-effects model was used to calculate pooled sensitivity, specificity, and AUC, with heterogeneity assessed by the I2 statistic.
Main Results
- The meta-analysis included 16 studies with 2514 patients.
- Pooled sensitivity, specificity, and AUC for detecting LVSI were 0.84, 0.78, and 0.87 (interval validation), and 0.79, 0.76, and 0.84 (external validation), respectively.
- Deep learning models showed significantly higher sensitivity than machine learning (P=.01), and AI models using PET/CT were superior to MRI-based models (P=.01).
Conclusions
- Imaging-based AI, especially deep learning, demonstrates strong diagnostic performance for predicting cervical cancer LVSI.
- Limitations include potential biases from limited external validation and retrospective study designs.
- AI holds potential as an auxiliary diagnostic tool, warranting large-scale prospective validation.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.