Deep neural networks allow expert-level brain meningioma segmentation and present potential for improvement of clinical practice.

Autor: Boaro A; Computational Neuroscience Outcomes Center, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA. alessandro.boaro@univr.it.; Section of Neurosurgery, Department of Neurosciences, Biomedicine and Movement Sciences, University of Verona, Verona, Italy. alessandro.boaro@univr.it., Kaczmarzyk JR; McGovern Institute for Brain Research, Massachusetts Institute of Technology, Cambridge, MA, USA.; Medical Scientist Training Program, Stony Brook University School of Medicine, Stony Brook, NY, USA., Kavouridis VK; Computational Neuroscience Outcomes Center, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA., Harary M; Computational Neuroscience Outcomes Center, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA.; Department of Neurosurgery, University of California Los Angeles, Los Angeles, CA, USA., Mammi M; Computational Neuroscience Outcomes Center, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA., Dawood H; Computational Neuroscience Outcomes Center, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA.; Department of Neurosurgery, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA., Shea A; Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA., Cho EY; Computational Neuroscience Outcomes Center, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA., Juvekar P; Department of Neurosurgery, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA., Noh T; Department of Neurosurgery, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA., Rana A; Computational Neuroscience Outcomes Center, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA.; McGovern Institute for Brain Research, Massachusetts Institute of Technology, Cambridge, MA, USA., Ghosh S; McGovern Institute for Brain Research, Massachusetts Institute of Technology, Cambridge, MA, USA. satra@mit.edu., Arnaout O; Computational Neuroscience Outcomes Center, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA.; Department of Neurosurgery, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA.
Jazyk: angličtina
Zdroj: Scientific reports [Sci Rep] 2022 Sep 14; Vol. 12 (1), pp. 15462. Date of Electronic Publication: 2022 Sep 14.
DOI: 10.1038/s41598-022-19356-5
Abstrakt: Accurate brain meningioma segmentation and volumetric assessment are critical for serial patient follow-up, surgical planning and monitoring response to treatment. Current gold standard of manual labeling is a time-consuming process, subject to inter-user variability. Fully-automated algorithms for meningioma segmentation have the potential to bring volumetric analysis into clinical and research workflows by increasing accuracy and efficiency, reducing inter-user variability and saving time. Previous research has focused solely on segmentation tasks without assessment of impact and usability of deep learning solutions in clinical practice. Herein, we demonstrate a three-dimensional convolutional neural network (3D-CNN) that performs expert-level, automated meningioma segmentation and volume estimation on MRI scans. A 3D-CNN was initially trained by segmenting entire brain volumes using a dataset of 10,099 healthy brain MRIs. Using transfer learning, the network was then specifically trained on meningioma segmentation using 806 expert-labeled MRIs. The final model achieved a median performance of 88.2% reaching the spectrum of current inter-expert variability (82.6-91.6%). We demonstrate in a simulated clinical scenario that a deep learning approach to meningioma segmentation is feasible, highly accurate and has the potential to improve current clinical practice.
(© 2022. The Author(s).)
Databáze: MEDLINE
Externí odkaz:
Nepřihlášeným uživatelům se plný text nezobrazuje