EUVIMEDEUVIMED entdecken
Multizentrische StudieOpen Access

Identification of Multiclass Pediatric Low-Grade Neuroepithelial Tumor Molecular Subtype with ADC MR Imaging and Machine Learning

Matheus D Soldatelli iD, Khashayar Namdar, Uri Tabori iD, Cynthia Hawkins iD, Kristen Yeom iD, Farzad Khalvati iD, Birgit B Ertl-Wagner iD, Matthias W Wagner iD

American Journal of Neuroradiology · 2024

DOI: 10.3174/ajnr.a8199 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift American Journal of Neuroradiology, 2024. Verfasst von Matheus D Soldatelli und 7 weiteren Autor:innen. Beteiligte Einrichtung: From the Department Diagnostic Imaging (M.D.S., B.B.E.-W., M.W.W.), Division of Neuroradiology, The Hospital for Sick Children, Toronto, Ontario, Can…. Hauptthemen laut MEDLINE: Machine Learning, Neoplasms, Neuroepithelial, Brain Neoplasms, Diffusion Magnetic Resonance Imaging. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: Canadian Institutes for Health Research, Canadian Cancer Society Research Institute.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

BACKGROUND AND PURPOSE: Molecular biomarker identification increasingly influences the treatment planning of pediatric low-grade neuroepithelial tumors (PLGNTs). We aimed to develop and validate a radiomics-based ADC signature predictive of the molecular status of PLGNTs. MATERIALS AND METHODS: In this retrospective bi-institutional study, we searched the PACS for baseline brain MRIs from children with PLGNTs. Semiautomated tumor segmentation on ADC maps was performed using the semiautomated level tracing effect tool with 3D Slicer. Clinical variables, including age, sex, and tumor location, were collected from chart review. The molecular status of tumors was derived from biopsy. Multiclass random forests were used to predict the molecular status and fine-tuned using a grid search on the validation sets. Models were evaluated using independent and unseen test sets based on the combined data, and the area under the receiver operating characteristic curve (AUC) was calculated for the prediction of 3 classes: KIAA1549-BRAF fusion, BRAF V600E mutation, and non-BRAF cohorts. Experiments were repeated 100 times using different random data splits and model initializations to ensure reproducible results. RESULTS: Two hundred ninety-nine children from the first institution and 23 children from the second institution were included (53.6% male; mean, age 8.01 years; 51.8% supratentorial; 52.2% with KIAA1549-BRAF fusion). For the 3-class prediction using radiomics features only, the average test AUC was 0.74 (95% CI, 0.73-0.75), and using clinical features only, the average test AUC was 0.67 (95% CI, 0.66-0.68). The combination of both radiomics and clinical features improved the AUC to 0.77 (95% CI, 0.75-0.77). The diagnostic performance of the per-class test AUC was higher in identifying KIAA1549-BRAF fusion tumors among the other subgroups (AUC = 0.81 for the combined radiomics and clinical features versus 0.75 and 0.74 for BRAF V600E mutation and non-BRAF, respectively). CONCLUSIONS: ADC values of tumor segmentations have differentiative signals that can be used for training machine learning classifiers for molecular biomarker identification of PLGNTs. ADC-based pretherapeutic differentiation of the BRAF status of PLGNTs has the potential to avoid invasive tumor biopsy and enable earlier initiation of targeted therapy.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Machine LearningNeoplasms, NeuroepithelialBrain NeoplasmsDiffusion Magnetic Resonance ImagingHumansChildFemaleMaleRetrospective StudiesChild, PreschoolAdolescentProto-Oncogene Proteins B-rafInfantNeoplasm GradingBiomarkers, Tumor

Substanzen

Proto-Oncogene Proteins B-rafBiomarkers, TumorBRAF protein, human

Autor:innen und Institutionen

  • From the Department Diagnostic Imaging (M.D.S., B.B.E.-W., M.W.W.), Division of Neuroradiology, The Hospital for Sick Children, Toronto, Ontario, Canada md.soldatelli@gmail.com.
  • Department of Medical Imaging (M.D.S., K.N., F.K., B.B.E.-W., M.W.W.), University of Toronto, Toronto, Ontario, Canada.
  • Institute of Medical Science (M.D.S., K.N., U.T., F.K., B.B.E.-W.), University of Toronto, Toronto, Ontario, Canada.
  • Vector Institute (K.N., F.K.), Toronto, Ontario, Canada.
  • The Arthur and Sonia Labatt Brain Tumour Research Centre (U.T., C.H.), The Hospital for Sick Children, Toronto, Ontario, Canada.
  • Program in Genetics and Genome Biology (U.T.) The Hospital for Sick Children, Toronto, Ontario, Canada.

ORCID-Profile: Matheus D Soldatelli, Uri Tabori, Cynthia Hawkins, Kristen Yeom, Farzad Khalvati, Birgit B Ertl-Wagner, Matthias W Wagner

Förderung

  • Canadian Institutes for Health Research
  • Canadian Cancer Society Research Institute

Zitieren

APA

Soldatelli, M. D., Namdar, K., Tabori, U., Hawkins, C., Yeom, K., Khalvati, F., Ertl-Wagner, B. B., & Wagner, M. W. (2024). Identification of Multiclass Pediatric Low-Grade Neuroepithelial Tumor Molecular Subtype with ADC MR Imaging and Machine Learning. American Journal of Neuroradiology. https://doi.org/10.3174/ajnr.a8199

Vancouver

Soldatelli MD, Namdar K, Tabori U, Hawkins C, Yeom K, Khalvati F, et al. Identification of Multiclass Pediatric Low-Grade Neuroepithelial Tumor Molecular Subtype with ADC MR Imaging and Machine Learning. American Journal of Neuroradiology. 2024. doi:10.3174/ajnr.a8199

BibTeX
@article{Soldatelli2024,
  author = {Soldatelli, Matheus D and Namdar, Khashayar and Tabori, Uri and Hawkins, Cynthia and Yeom, Kristen and Khalvati, Farzad and Ertl-Wagner, Birgit B and Wagner, Matthias W},
  title = {Identification of Multiclass Pediatric Low-Grade Neuroepithelial Tumor Molecular Subtype with ADC MR Imaging and Machine Learning},
  journal = {American Journal of Neuroradiology},
  year = {2024},
  doi = {10.3174/ajnr.a8199}
}
Kostenlos starten →