Radiomics in the CT diagnosis of ovarian cystic malignancies - a pilot study.

Autor: Mărginean L; Radiology and Medical Imaging, Clinical Sciences Department, 'George Emil Palade' University of Medicine, Pharmacy, Science, and Technology, Târgu Mureş, Romania.; Interventional Radiology Department, Târgu Mureş County Emergency Clinical Hospital, Târgu Mureş, Romania., Ştefan PA; Interventional Radiology Department, Târgu Mureş County Emergency Clinical Hospital, Târgu Mureş, Romania.; Department of Biomedical Imaging and Image-Guided Therapy, General Hospital of Vienna (AKH), Medical University of Vienna, Austria.; Department of Anatomy and Embriology, Morphological Sciences, Iuliu Hatieganu University of Medicine and Pharmacy, Cluj-Napoca, Romania.; Department of Radiology and Imaging, Cluj County Emergency Clinical Hospital, Cluj-Napoca, Romania., Filep RC; Radiology and Medical Imaging, Clinical Sciences Department, 'George Emil Palade' University of Medicine, Pharmacy, Science, and Technology, Târgu Mureş, Romania.; Interventional Radiology Department, Târgu Mureş County Emergency Clinical Hospital, Târgu Mureş, Romania., Csutak C; Department of Radiology and Imaging, Cluj County Emergency Clinical Hospital, Cluj-Napoca, Romania.; Department of Radiology and Imaging, Surgical Specialties, Iuliu Hatieganu University of Medicine and Pharmacy, Cluj-Napoca, Romania., Lebovici A; Department of Radiology and Imaging, Cluj County Emergency Clinical Hospital, Cluj-Napoca, Romania.; Department of Radiology and Imaging, Surgical Specialties, Iuliu Hatieganu University of Medicine and Pharmacy, Cluj-Napoca, Romania., Gherman D; Department of Radiology and Imaging, Cluj County Emergency Clinical Hospital, Cluj-Napoca, Romania.; Department of Radiology and Imaging, Surgical Specialties, Iuliu Hatieganu University of Medicine and Pharmacy, Cluj-Napoca, Romania., Lupean RA; Department of Histology, Morphological Sciences, Iuliu Hatieganu University of Medicine and Pharmacy, Cluj-Napoca, Romania.; 'Dominic Stanca' Obstetrics and Gynecology Clinic, Cluj County Emergency Clinical Hospital, Cluj-Napoca, Romania., Suciu BA; First Surgical Clinic, Târgu Mureş County Emergency Clinical Hospital, Târgu Mureş, Romania.; Department of Anatomy, Morphological Sciences, 'George Emil Palade' University of Medicine, Pharmacy, Science, and Technology, Târgu Mureş, Romania.
Jazyk: angličtina
Zdroj: Medicine and pharmacy reports [Med Pharm Rep] 2024 Apr; Vol. 97 (2), pp. 169-177. Date of Electronic Publication: 2024 Apr 25.
DOI: 10.15386/mpr-2594
Abstrakt: Background and Aims: The conventional computed tomography (CT) appearance of ovarian cystic masses is often insufficient to adequately differentiate between benign and malignant entities. This study aims to investigate whether texture analysis of the fluid component can augment the CT diagnosis of ovarian cystic tumors.
Methods: Eighty-four patients with adnexal cystic lesions who underwent CT examinations were retrospectively included. All patients had a final diagnosis that was established by histological analysis in forty four cases. The texture features of the lesions content were extracted using dedicated software and further used for comparing benign and malignant lesions, primary tumors and metastases, malignant and borderline lesions, and benign and borderline lesions. Texture features' discriminatory ability was evaluated through univariate and receiver operating characteristics analysis and also by the use of the k-nearest-neighbor classifier.
Results: The univariate analysis showed statistically significant results when comparing benign and malignant lesions (the Difference Variance parameter, p=0.0074) and malignant and borderline tumors (the Correlation parameter, p=0.488). The highest accuracy (83.33%) was achieved by the classifier when discriminating primary tumors from ovarian metastases.
Conclusion: Texture parameters were able to successfully discriminate between different types of ovarian cystic lesions based on their content, but it is not entirely clear whether these differences are a result of the physical properties of the fluids or their appartenance to a particular histopathological group. If further validated, radiomics can offer a rapid and non-invasive alternative in the diagnosis of ovarian cystic tumors.
Databáze: MEDLINE
Externí odkaz: