Academic Journal
Differentiating MYCN-amplified RB1 wild-type retinoblastoma from biallelic RB1 mutant retinoblastoma using MR-based radiomics: a retrospective multicenter case-control study.
| العنوان: | Differentiating MYCN-amplified RB1 wild-type retinoblastoma from biallelic RB1 mutant retinoblastoma using MR-based radiomics: a retrospective multicenter case-control study. |
|---|---|
| المؤلفون: | de Bloeme, C.M., Jansen, R.W., Cardoen, L., Göricke, S., van Elst, S., Jessen, J.L., Ramasubramanian, A., Skalet, A.H., Miller, A.K., Maeder, P., Uner, O.E., Hubbard, G.B., Grossniklaus, H., Boldt, H.C., Nichols, K.E., Brennan, R.C., Sen, S., Koob, M., Sirin, S., Brisse, H.J., Galluzzi, P., Dommering, C.J., Cysouw, M., Boellaard, R., Dorsman, J.C., Moll, A.C., de Jong, M.C., de Graaf, P. |
| المساهمون: | European Retinoblastoma Imaging Collaboration |
| المصدر: | Scientific reports, vol. 14, no. 1, pp. 25103 |
| سنة النشر: | 2024 |
| المجموعة: | Université de Lausanne (UNIL): Serval - Serveur académique lausannois |
| مصطلحات موضوعية: | Humans, Retinoblastoma/genetics, Retinoblastoma/diagnostic imaging, Retinoblastoma/pathology, N-Myc Proto-Oncogene Protein/genetics, Female, Magnetic Resonance Imaging/methods, Case-Control Studies, Male, Retrospective Studies, Retinoblastoma Binding Proteins/genetics, Ubiquitin-Protein Ligases/genetics, Child, Preschool, Infant, Retinal Neoplasms/genetics, Retinal Neoplasms/diagnostic imaging, Retinal Neoplasms/pathology, Machine Learning, Mutation, Diagnosis, Differential, Radiomics, MYCN-amplification, MRI, Retinoblastoma |
| الوصف: | MYCN-amplified RB1 wild-type (MYCN amp RB1 +/+ ) retinoblastoma is a rare and aggressive subtype, often resistant to standard therapies. Identifying unique MRI features is crucial for diagnosing this subtype, as biopsy is not recommended. This study aimed to differentiate MYCN amp RB1 +/+ from the most prevalent RB1 -/- retinoblastoma using pretreatment MRI and radiomics. Ninety-eight unilateral retinoblastoma patients (19 MYCN cases and 79 matched controls) were included. Tumors on T2-weighted MR images were manually delineated and validated by experienced radiologists. Radiomics analysis extracted 120 features per tumor. Several combinations of feature selection methods, oversampling techniques and machine learning (ML) classifiers were evaluated in a repeated fivefold cross-validation machine learning pipeline to yield the best-performing prediction model for MYCN. The best model used univariate feature selection, data oversampling (duplicating MYCN cases), and logistic regression classifier, achieving a mean AUC of 0.78 (SD 0.12). SHAP analysis highlighted lower sphericity, higher flatness, and greater gray-level heterogeneity as predictive for MYCN amp RB1 +/+ status, yielding an AUC of 0.81 (SD 0.11). This study shows the potential of MRI-based radiomics to distinguish MYCN amp RB1 +/+ and RB1 -/- retinoblastoma subtypes. |
| نوع الوثيقة: | article in journal/newspaper |
| اللغة: | English |
| Relation: | info:eu-repo/semantics/altIdentifier/pmid/39443629; info:eu-repo/semantics/altIdentifier/eissn/2045-2322; https://serval.unil.ch/notice/serval:BIB_392CB96E6104 |
| DOI: | 10.1038/s41598-024-76933-6 |
| الاتاحة: | https://serval.unil.ch/notice/serval:BIB_392CB96E6104 https://doi.org/10.1038/s41598-024-76933-6 |
| رقم الانضمام: | edsbas.2060D09C |
| قاعدة البيانات: | BASE |
| DOI: | 10.1038/s41598-024-76933-6 |
|---|