Noninvasive molecular subtyping of pediatric low-grade glioma with self-supervised transfer learning.

Autor:	Tak D; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, MA, USA.; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Ye Z; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, MA, USA.; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Zapaishchykova A; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, MA, USA.; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Zha Y; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, MA, USA.; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Boyd A; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, MA, USA.; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Vajapeyam S; Department of Radiology, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Chopra R; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, MA, USA.; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Hayat H; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, MA, USA.; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Prabhu S; Department of Radiology, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Liu KX; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Elhalawani H; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Nabavizadeh A; Center for Data-Driven Discovery in Biomedicine (D3b), Children's Hospital of Philadelphia, Philadelphia, PA, USA.; Department of Radiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA., Familiar A; Center for Data-Driven Discovery in Biomedicine (D3b), Children's Hospital of Philadelphia, Philadelphia, PA, USA.; Department of Neurosurgery, Children's Hospital of Philadelphia, Philadelphia, PA, USA., Resnick A; Department of Neurosurgery, Children's Hospital of Philadelphia, Philadelphia, PA, USA., Mueller S; Department of Neurology, University of California San Francisco, San Francisco, CA. USA.; Department of Pediatrics, University of California San Francisco, San Francisco, CA, USA.; Department of Neurological Surgery, University of California San Francisco, San Francisco, CA, USA., Aerts HJWL; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, MA, USA.; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA.; Department of Radiology, Brigham and Women's Hospital, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA, USA.; Radiology and Nuclear Medicine, CARIM & GROW, Maastricht University, Maastricht, the Netherlands., Bandopadhayay P; Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Ligon K; Department of Pathology, Dana-Farber Cancer Institute, Boston Children's Hospital, Harvard Medical School, Boston, A, USA., Haas-Kogan D; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Poussaint T; Department of Radiology, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA., Kann BH; Artificial Intelligence in Medicine (AIM) Program, Mass General Brigham, Harvard Medical School, Boston, MA, USA.; Department of Radiation Oncology, Dana-Farber Cancer Institute | Brigham and Women's Hospital | Boston Children's Hospital, Harvard Medical School, Boston, MA, USA.
Jazyk:	angličtina
Zdroj:	MedRxiv : the preprint server for health sciences [medRxiv] 2023 Nov 22. Date of Electronic Publication: 2023 Nov 22.
DOI:	10.1101/2023.08.04.23293673
Abstrakt:	Purpose: To develop and externally validate a scan-to-prediction deep-learning pipeline for noninvasive, MRI-based BRAF mutational status classification for pLGG. Materials and Methods: We conducted a retrospective study of two pLGG datasets with linked genomic and diagnostic T2-weighted MRI of patients: BCH (development dataset, n=214 [60 (28%) BRAF fusion, 50 (23%) BRAF V600E, 104 (49%) wild-type), and Child Brain Tumor Network (CBTN) (external validation, n=112 [60 (53%) BRAF-Fusion, 17 (15%) BRAF-V600E, 35 (32%) wild-type]). We developed a deep learning pipeline to classify BRAF mutational status (V600E vs. fusion vs. wildtype) via a two-stage process: 1) 3D tumor segmentation and extraction of axial tumor images, and 2) slice-wise, deep learning-based classification of mutational status. We investigated knowledge-transfer and self-supervised approaches to prevent model overfitting with a primary endpoint of the area under the receiver operating characteristic curve (AUC). To enhance model interpretability, we developed a novel metric, COMDist, that quantifies the accuracy of model attention around the tumor. Results: A combination of transfer learning from a pretrained medical imaging-specific network and self-supervised label cross-training (TransferX) coupled with consensus logic yielded the highest macro-average AUC (0.82 [95% CI: 0.70-0.90]) and accuracy (77%) on internal validation, with an AUC improvement of +17.7% and a COMDist improvement of +6.4% versus training from scratch. On external validation, the TransferX model yielded AUC (0.73 [95% CI 0.68-0.88]) and accuracy (75%). Conclusion: Transfer learning and self-supervised cross-training improved classification performance and generalizability for noninvasive pLGG mutational status prediction in a limited data scenario. Competing Interests: Competing Interests All the authors declare no competing interests.
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu