Cargando…

MLTI-17. DIFFERENTIATION OF RADIATION INJURY FROM RECURRENT BRAIN METASTASIS USING COMBINED FET PET/MRI RADIOMICS

BACKGROUND: The aim of this study was to investigate the potential of combined radiomics textural feature analysis of contrast-enhanced MRI (CE MRI) and static O-(2-[(18)F]fluoroethyl)-L-tyrosine(FET) PET for the differentiation of recurrent brain metastasis from radiation injury. PATIENTS AND METHO...

Descripción completa

Detalles Bibliográficos
Autores principales: Lohmann, Philipp, Kocher, Martin, Ceccon, Garry, Bauer, Elena, Stoffels, Gabriele, Viswanathan, Shivakumar, Ruge, Maximilian, Neumaier, Bernd, Shah, Nadim, Fink, Gereon, Langen, Karl-Josef, Galldiks, Norbert
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Oxford University Press 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7213311/
http://dx.doi.org/10.1093/noajnl/vdz014.076
Descripción
Sumario:BACKGROUND: The aim of this study was to investigate the potential of combined radiomics textural feature analysis of contrast-enhanced MRI (CE MRI) and static O-(2-[(18)F]fluoroethyl)-L-tyrosine(FET) PET for the differentiation of recurrent brain metastasis from radiation injury. PATIENTS AND METHODS: Fifty-two patients with newly diagnosed or progressive contrast-enhancing brain lesions on MRI after radiotherapy (predominantly radiosurgery, 84% of patients) of brain metastases were additionally investigated using FET PET. Based on histology (n=19) or clinicoradiological follow-up (n=33), local recurrent brain metastases were diagnosed in 21 patients (40%) and radiation injury in 31 patients (60%). Forty-two features (shape-based, first and second order features) were calculated on both unfiltered and filtered CE MRI and summed FET PET images (20–40 min p.i). After feature selection, logistic regression models using a maximum of five features to avoid overfitting were calculated for each imaging modality separately and for the combined FET PET/MRI features. The resulting models were validated using cross-validation. Diagnostic accuracies were calculated for each imaging modality separately as well as for the combined model. RESULTS: For differentiation between radiation injury and brain metastasis recurrence, textural features extracted from CE MRI had a diagnostic accuracy of 81%. FET PET textural features revealed a slightly higher diagnostic accuracy of 83%. However, the highest diagnostic accuracy was obtained when combining CE MRI and FET PET features (accuracy, 89%). CONCLUSION: Our findings suggest that combined FET PET/MRI radiomics using textural feature analysis offers a great potential to contribute significantly to the management of patients with brain metastases. SUPPORT: This work was supported by the Wilhelm-Sander Stiftung, Germany