Aimof this study is to assess the role of radiomics in breast cancer, focusing on its ability to predict histological phenotypes.Methods: It was carried out a retrospective study including all breast cancer patients who referred to the Diagnostic Imaging 2 Institute to perform a breast MRI prior to surgery. Histologic confirmation on surgical specimen was obtained in all patients.Results: All patients underwent a high-field (1,5 T) breast MRI (T2-weight sequences with fat suppression, T1-weight sequences and T1 after IV contrast administration). MRI enhanced sequences were contoured and first-order radiomics features were extracted through dedicated software, called 3DSlicer. Taking into account the current literature, three features were selected: entropy, 90° percentile and skewness. The association between the selected features and the histological patterns were explored.Conclusion: A possible integration of quantitative data, provided by a radiomics analysis, with clinical and histological data could enable clinicians to provide an early and accurate diagnosis as well as personalized treatments for breast cancer patients. Current limitations of a routinely application of radiomics seem both the limited knowledge among physicians of this quantitative radiological tool and the lack of standardized systems of feature extraction and data sharing.
Aspetti radiomici del tumore mammario: studio di associazione trasignatureradiomica ed istotipo / Crivelli, Paola. - (2019 Feb 25).
Aspetti radiomici del tumore mammario: studio di associazione trasignatureradiomica ed istotipo
CRIVELLI, Paola
2019-02-25
Abstract
Aimof this study is to assess the role of radiomics in breast cancer, focusing on its ability to predict histological phenotypes.Methods: It was carried out a retrospective study including all breast cancer patients who referred to the Diagnostic Imaging 2 Institute to perform a breast MRI prior to surgery. Histologic confirmation on surgical specimen was obtained in all patients.Results: All patients underwent a high-field (1,5 T) breast MRI (T2-weight sequences with fat suppression, T1-weight sequences and T1 after IV contrast administration). MRI enhanced sequences were contoured and first-order radiomics features were extracted through dedicated software, called 3DSlicer. Taking into account the current literature, three features were selected: entropy, 90° percentile and skewness. The association between the selected features and the histological patterns were explored.Conclusion: A possible integration of quantitative data, provided by a radiomics analysis, with clinical and histological data could enable clinicians to provide an early and accurate diagnosis as well as personalized treatments for breast cancer patients. Current limitations of a routinely application of radiomics seem both the limited knowledge among physicians of this quantitative radiological tool and the lack of standardized systems of feature extraction and data sharing.File | Dimensione | Formato | |
---|---|---|---|
Crivelli_P_Aspetti_radiomici_del_tumore.pdf
accesso aperto
Tipologia:
Altro materiale allegato
Licenza:
Non specificato
Dimensione
2.1 MB
Formato
Adobe PDF
|
2.1 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.