Radiomics in the evaluation of ovarian masses - a systematic review

Area of Science:

• Medical Imaging
• Radiology
• Oncology

Background:

• Radiomics extracts quantitative features from medical images.
• Ovarian lesions require accurate diagnostic tools for effective patient management.

Purpose of the Study:

• To systematically review the literature on radiomics applied to ovarian lesion imaging.
• To assess the current state and potential of radiomics in ovarian lesion diagnosis and management.

Main Methods:

• Systematic literature search across multiple databases (MEDLINE/PubMed, Web of Science, Scopus, EMBASE, Ovid, ClinicalTrials.gov).
• Data extraction using PRISMA criteria.
• Quality and bias assessment using the Quality in Prognosis Studies (QUIPS) tool.
• Radiomic methodology evaluation using the Radiomic Quality Score (RQS).

Main Results:

• 33 eligible studies were identified, focusing on lesion classification, treatment outcomes, and predictions.
• CT and MRI were the most common imaging modalities; 70% of studies used 3D segmentation.
• Radiomic models outperformed radiologist interpretation in 5 studies, but only 18% included external validation.
• A high risk of bias (58%) was noted, with the highest RQS achieved being 61.1%.

Conclusions:

• Radiomics holds potential as a diagnostic tool for ovarian masses, aiding in lesion stratification and personalized care.
• Standardization of methods, larger cohorts, and real-world evaluation are crucial for clinical translation.
• Radiomics demonstrates promise in enhancing clinical decisions for ovarian lesions, but requires further development.