# Role of Imaging Techniques in Ovarian Cancer Diagnosis: Current Approaches and Future Directions

## TL;DR

This paper reviews how advanced imaging techniques and AI can improve the accuracy of diagnosing ovarian cancer, especially when traditional methods are unclear.

## Contribution

The paper synthesizes recent advances in contrast-enhanced MRI, O-RADS scoring, and AI applications for ovarian cancer diagnosis.

## Key findings

- Contrast-enhanced MRI provides high accuracy (83–93%) in characterizing indeterminate ovarian masses.
- The O-RADS MRI Score achieves 93% sensitivity and 91% specificity for malignancy risk assessment.
- AI-based methods show promise in improving diagnostic precision but require further clinical validation.

## Abstract

Ovarian cancer is a leading cause of death among gynecological malignancies. Standard ultrasound scans may not be conclusive, especially when ovarian masses are difficult to classify. This review highlights recent advances aimed at reducing diagnostic uncertainty. Contrast-enhanced MRI has demonstrated high accuracy in differentiating benign from malignant lesions, and the O-RADS MRI scoring system provides structured risk assessment with strong sensitivity and specificity. New classification methods are also being developed to further support clinical decision-making. In addition, artificial intelligence (AI) approaches, including machine learning and deep learning, are being tested to improve diagnostic precision by analyzing complex imaging data. Overall, the integration of advanced imaging with AI has the potential to substantially improve the evaluation and management of women with suspected ovarian cancer.

Background: Ovarian cancer is a leading gynecological malignancy with high global mortality. Early and accurate diagnosis is critical for optimal management; however, a considerable portion of ovarian masses remain indeterminate after initial evaluation. Although transvaginal ultrasound is the first-line imaging tool, up to 30% of cases yield inconclusive findings, complicating treatment decisions. Methods: This review summarizes current diagnostic strategies for ovarian masses, with an emphasis on advanced imaging and emerging technologies. Topics include the diagnostic utility of contrast-enhanced MRI, the application of the O-RADS MRI scoring system, and the integration of Artificial Intelligence (AI) into imaging workflows. Results: Contrast-enhanced MRI offers high diagnostic accuracy (83–93%) for characterizing indeterminate ovarian masses. The O-RADS MRI Score offers a reported sensitivity of 93% and specificity of 91% for malignancy risk assessment. Additionally, new classification systems have been proposed to further improve diagnostic performance. AI-based approaches, particularly machine learning and deep learning applied to imaging data, show potential in improving diagnostic precision; however, most techniques require further clinical validation before widespread adoption. Conclusions: Advanced imaging techniques and AI-driven methods are reshaping the diagnostic landscape of ovarian cancer. While current tools like MRI and O-RADS enhance accuracy, ongoing research into novel models and AI applications suggests further improvements are possible. Clinical validation and expert oversight remain essential for their integration into routine practice.