Review of AI models shows high accuracy for ovarian cancer diagnosis and outcome prediction

Ovarian cancer (OC) is a predominant cause of fatality amongst gynecological malignancies, frequently identified at its later stages owing to its asymptomatic characteristics and the absence of adequate screening techniques. Imaging techniques such as ultrasound (US), magnetic resonance imaging (MRI), and computed tomography (CT) are crucial for diagnosis, but traditional methods rely heavily on subjective evaluations by radiologists. AI and radiomics offer a data-driven approach to extract quantitative features from medical images, enabling more accurate and personalized diagnosis and prognosis. This review highlights the role of AI in improving the analysis of biomarkers like CA-125, HE4, and microRNAs, and discusses the potential of integrating multiomics data (genomics, transcriptomics, epigenomics, etc.) with imaging data to enhance predictive models. Radiomics, which involves extracting high-dimensional features from medical images, has shown promise in differentiating between benign and malignant tumors, predicting genetic mutations (e.g., BRCA), and assessing tumor heterogeneity. Artificial intelligence (AI) models, particularly deep learning (DL) algorithms, have demonstrated high accuracy in diagnosing OC and predicting patient outcomes, often outperforming traditional methods.