JMIR Publications

ABSTRACT

Background:

It is controversial that artificial intelligence(AI) can improve the accuracy of LVSI detection.

Objective:

This meta-analysis aimed to assess the diagnostic performance of artificial intelligence based on images for predicting lymph vascular space invasion(LVSI) in cervical cancer.

Methods:

We conducted a comprehensive literature search across multiple databases including PubMed, Embase, and Web of Science, identifying relevant studies published up to the specified date. The selected research focused on evaluating artificial intelligence's diagnostic performance in detecting LVSI in cervical cancer. We employed a bivariate random-effects model to calculate pooled sensitivity and specificity with corresponding 95% confidence intervals. Study heterogeneity was assessed using the I2 statistic.

Results:

Of 403 studies identified, 16 studies (2,514 patients) were included. For interval validation set, the pooled sensitivity, specificity, and area under the curve(AUC) for detecting LVSI was 0.84 (95% CI: 0.79-0.87), 0.78 (95% CI: 0.75-0.81), and 0.87 (95% CI: 0.84-0.90). For external validation set, the pooled sensitivity, specificity, and AUC for detecting LVSI was 0.79 (95% CI: 0.70-0.86),0.76 (95% CI: 0.67-0.83), and 0.84 (95% CI: 0.81-0.87). Using likelihood ratio test for subgroup analysis, deep learning demonstrated significantly higher sensitivity compared to machine learning (P = 0.01). Moreover, AI models based on PET/CT exhibited superior sensitivity relative to those based on MRI (P = 0.01).

Conclusions:

Imaging-based AI, particularly deep learning algorithms, shows promising diagnostic performance in predicting LVSI in cervical cancer. These findings underscore AI's potential as an auxiliary diagnostic tool, necessitating further large-scale prospective validation.