Science and Technology Production
Assessing robustness and generalization of a deep neural network for brain MS lesion segmentation on real-world data
Article
Authorship:
Hernán Chaves ; Maria M Serra ; Diego Shalom ; Fernanda Rueda ; Emilia Osa Sanz ; Nadia Stefanoff ; Sofia Rodriguez Murua ; Martin E Acosta ; Felipe Kitamura ; Paulina Yañez ; Claudia Cejas ; Jorge Correale ; FERRANTE, ENZO ; Diego Fernandez Slezak ; Mauricio FarezDate:
2023Publishing House and Editing Place:
Springer Science and Business Media Deutschland GmbHMagazine:
European Radiology, vol. 34 (pp. 2024-2035) Springer Science and Business Media Deutschland GmbHSummary *
This ambispective, multicenter study assessed the performance of a DL-based model for MS lesion segmentation and compared it to alternative DL- and non-DL-based methods. Models were tested on internal (n = 20) and external (n = 18) datasets from Latin America, and on an external dataset from Europe (n = 49). We also examined robustness by rescanning six patients (n = 6) from our MS clinical cohort. Moreover, we studied inter-human annotator agreement and discussed our findings in light of these results. Performance and robustness were assessed using intraclass correlation coefficient (ICC), Dice coefficient (DC), and coefficient of variation (CV). Information provided by the agent in SIGEVAKey Words
deep learningmagnetic resonance imagingwhite mattermultiple sclerosis