Systematic Review: MRI AI Segmentation for Cervical Cancer Shows Promise but Low Evidence Certainty

Across thirty studies, deep learning MRI segmentation of cervical cancer reached Dice coefficients 0.60–0.93; GRADE certainty low due to heterogeneity and limited external validation — promising but not yet routine.

• Systematic review of thirty studies (PROSPERO CRD420251247441) evaluating machine learning and deep learning for cervical cancer MRI segmentation; most used U-Net variants, convolutional neural networks, and transformer-based architectures.
• Reported Dice similarity coefficients ranged from 0.60 to 0.93, varying by target structure, tumor characteristics, and imaging protocol.
• QUADAS-2 risk of bias was 'some concerns' mainly due to patient selection, reference standards, and limited external validation; GRADE certainty of evidence was low.

Evidence grade: low

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.