TY - JOUR
T1 - Tumor segmentation analysis at different post-contrast time points
T2 - A possible source of variability of quantitative DCE-MRI parameters in locally advanced breast cancer
AU - Romeo, Valeria
AU - Cavaliere, Carlo
AU - Imbriaco, Massimo
AU - Verde, Francesco
AU - Petretta, Mario
AU - Franzese, Monica
AU - Stanzione, Arnaldo
AU - Cuocolo, Renato
AU - Aiello, Marco
AU - Basso, Luca
AU - Amitrano, Michele
AU - Lauria, Rossella
AU - Accurso, Antonello
AU - Brunetti, Arturo
AU - Salvatore, Marco
N1 - Funding Information:
This work was partially supported by the Italian Ministry of Education, University and Research (project title MOLIM ONCOBRAIN LAB, CUP B26G18000410005) and by the Italian Ministry of Health (RICERCA CORRENTE grant) for IRCCS SDN.
Publisher Copyright:
© 2020 Elsevier B.V.
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2020/5
Y1 - 2020/5
N2 - Purpose: to assess if tumor segmentation analysis performed at different post-contrast time points (TPs) on dynamic images could influence the extraction of dynamic contrast enhanced (DCE)-MRI parameters in locally advanced breast cancer (LABC), and potentially represent a source of variability. Method: forty patients with forty-two LABC lesions were prospectively enrolled and underwent breast DCE-MRI examination at 3 T. On post-processed dynamic images, enhancing tumor lesions were manually segmented at four different TPs: at the first post-contrast dynamic image in which the lesion was appreciable (TP 1) and at 1, 5 and 10 min after contrast-agent administration (TPs 2, 3 and 4, respectively) and corresponding DCE-MRI parameters were extracted. Friedman's test followed by Bonferroni-adjusted Wilcoxon signed rank test for post-hoc analysis was used to compare DCE-MRI parameters. Intra- and inter-observer reliability of DCE-MRI parameters measurements was assessed using the Intraclass Correlation Coefficient (ICC) analysis. Results: Ktrans, Kep and iAUC were significantly higher when extracted from ROIs placed at TP1 and progressively decreased from TP 2–4. The intra-observer reliability ranged from good to excellent (ICC's: 0.894 to 0.990). The inter-observer reliability varied from moderate to excellent (0.770 to 0.942). The inter-observer reliability was significantly higher for Ktrans and Kep extracted at TPs1 and 2 as compared to TPs 3 and 4. Conclusions: A significant variability of DCE-MRI quantitative parameters occurs when tumor segmentation is performed at different TPs. We suggest to performing tumor delineation at an established TP, preferably the earliest, in order to extract reliable and comparable DCE-MRI data.
AB - Purpose: to assess if tumor segmentation analysis performed at different post-contrast time points (TPs) on dynamic images could influence the extraction of dynamic contrast enhanced (DCE)-MRI parameters in locally advanced breast cancer (LABC), and potentially represent a source of variability. Method: forty patients with forty-two LABC lesions were prospectively enrolled and underwent breast DCE-MRI examination at 3 T. On post-processed dynamic images, enhancing tumor lesions were manually segmented at four different TPs: at the first post-contrast dynamic image in which the lesion was appreciable (TP 1) and at 1, 5 and 10 min after contrast-agent administration (TPs 2, 3 and 4, respectively) and corresponding DCE-MRI parameters were extracted. Friedman's test followed by Bonferroni-adjusted Wilcoxon signed rank test for post-hoc analysis was used to compare DCE-MRI parameters. Intra- and inter-observer reliability of DCE-MRI parameters measurements was assessed using the Intraclass Correlation Coefficient (ICC) analysis. Results: Ktrans, Kep and iAUC were significantly higher when extracted from ROIs placed at TP1 and progressively decreased from TP 2–4. The intra-observer reliability ranged from good to excellent (ICC's: 0.894 to 0.990). The inter-observer reliability varied from moderate to excellent (0.770 to 0.942). The inter-observer reliability was significantly higher for Ktrans and Kep extracted at TPs1 and 2 as compared to TPs 3 and 4. Conclusions: A significant variability of DCE-MRI quantitative parameters occurs when tumor segmentation is performed at different TPs. We suggest to performing tumor delineation at an established TP, preferably the earliest, in order to extract reliable and comparable DCE-MRI data.
KW - Dynamic contrast enhanced MRI
KW - Ktrans
KW - Locally advanced breast cancer
KW - Magnetic resonance imaging
KW - Perfusion weighted MRI
UR - http://www.scopus.com/inward/record.url?scp=85080930756&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85080930756&partnerID=8YFLogxK
U2 - 10.1016/j.ejrad.2020.108907
DO - 10.1016/j.ejrad.2020.108907
M3 - Article
C2 - 32145597
AN - SCOPUS:85080930756
VL - 126
JO - European Journal of Radiology
JF - European Journal of Radiology
SN - 0720-048X
M1 - 108907
ER -