Publications by authors named "Xuanke Hong"
Background: We aimed to develop machine learning models for prediction of molecular subgroups (low-risk group and intermediate/high-risk group) and molecular marker (KIAA1549-BRAF fusion) of pediatric low-grade gliomas (PLGGs) based on radiomic features extracted from multiparametric MRI.
Methods: 61 patients with PLGGs were included in this retrospective study, which were divided into a training set and an internal validation set at a ratio of 2:1 based on the molecular subgroups or the molecular marker. The patients were classified into low-risk and intermediate/high-risk groups, BRAF fusion positive and negative groups, respectively.
Objectives: To investigate whether radiomic features extracted from dynamic susceptibility contrast perfusion-weighted imaging (DSC-PWI) can improve the prediction of the molecular subtypes of adult diffuse gliomas, and to further develop and validate a multimodal radiomic model by integrating radiomic features from conventional and perfusion MRI.
Methods: We extracted 1197 radiomic features from each sequence of conventional MRI and DSC-PWI, respectively. The Boruta algorithm was used for feature selection and combination, and a three-class random forest method was applied to construct the models.
Article Synopsis
- Researchers developed a deep learning imaging signature (DLIS) using multi-parametric MRI to predict survival risk in patients with glioblastoma multiforme (GBM) and explored its biological foundations.
- The DLIS was validated across multiple sets, showing strong associations with survival rates, and improved predictive capabilities over traditional clinicomolecular models.
- The findings linked the DLIS to important cancer pathways, like P53 and RB, as well as key genetic alterations, emphasizing its potential for enhancing personalized treatment strategies in GBM patients.
Article Synopsis
- Isocitrate dehydrogenase (IDH) mutation and 1p19q codeletion are important genetic markers for assessing therapy options and prognosis in lower-grade glioma (LGG), and this study created a machine learning model to predict different molecular subtypes of LGG using MRI data.
- The model was trained on a sample of 269 LGG patients using 5,929 extracted MRI features, and it improved accuracy by combining these features with qualitative assessments and clinical data.
- The final model demonstrated strong predictive performance, achieving area under the curve (AUC) values over 0.80 for key molecular subtypes when incorporating various factors, which suggests a promising non-invasive approach for preoperative diagnosis of LGG.
Determination of 1p/19q co-deletion status is important for the classification, prognostication, and personalized therapy in diffuse lower-grade gliomas (LGG). We developed and validated a deep learning imaging signature (DLIS) from preoperative magnetic resonance imaging (MRI) for predicting the 1p/19q status in patients with LGG. The DLIS was constructed on a training dataset (n = 330) and validated on both an internal validation dataset (n = 123) and a public TCIA dataset (n = 102).
View Article and Find Full Text PDFArticle Synopsis
- A deep learning signature (DLS) was created using diffusion tensor imaging (DTI) to predict overall survival in patients with infiltrative gliomas and explore related biological pathways.
- The DLS demonstrated a strong association with survival, serving as an independent predictor and outperforming existing risk systems when combined, showing improved accuracy in survival predictions.
- Five significant biological pathways were linked to the DLS, indicating that therapies targeting neuron-to-brain tumor communication could be particularly beneficial for high-risk glioma patients identified by the DTI-derived DLS.
The 2016 WHO classification of central nervous system tumors has included four molecular subgroups under medulloblastoma (MB) as sonic hedgehog (SHH), wingless (WNT), Grade 3, and Group 4. We aimed to develop machine learning models for predicting MB molecular subgroups based on multi-parameter magnetic resonance imaging (MRI) radiomics, tumor locations, and clinical factors. A total of 122 MB patients were enrolled retrospectively.
View Article and Find Full Text PDFArticle Synopsis
- A radiomics signature was created using MRI data from patients with medulloblastoma to predict overall survival (OS) and progression-free survival (PFS), showing promising results for both predictions.
- The study involved a training cohort of 83 patients and testing cohort of 83, confirming the increased predictive power of combining radiomic and clinico-molecular data compared to either alone.
- Key biological pathways linked to the radiomics signature were identified, highlighting their potential role in patient risk stratification and improving survival prognosis.