Dataset Information

A quantitative SVM approach potentially improves the accuracy of magnetic resonance spectroscopy in the preoperative evaluation of the grades of diffuse gliomas.

ABSTRACT:

Objectives

To investigate the association between proton magnetic resonance spectroscopy (¹H-MRS) metabolic features and the grade of gliomas, and to establish a machine-learning model to predict the glioma grade.

Methods

This study included 112 glioma patients who were divided into the training (n?=?74) and validation (n?=?38) sets based on the time of hospitalization. Twenty-six metabolic features were extracted from the preoperative ¹H-MRS image. The Student's t-test was conducted to screen for differentially expressed features between low- and high-grade gliomas (WHO grades II and III/IV, respectively). Next, the minimum Redundancy Maximum Relevance (mRMR) algorithm was performed to further select features for a support vector machine (SVM) classifier building. Performance of the predictive model was evaluated both in the training and validation sets using ROC curve analysis.

Results

Among the extracted ¹H-MRS metabolic features, thirteen features were differentially expressed. Four features were further selected as grade-predictive imaging signatures using the mRMR algorithm. The predictive performance of the machine-learning model measured by the AUC was 0.825 and 0.820 in the training and validation sets, respectively. This was better than the predictive performances of individual metabolic features, the best of which was 0.812.

Conclusions

¹H-MRS metabolic features could help in predicting the grade of gliomas. The machine-learning model achieved a better prediction performance in grading gliomas than individual features, indicating that it could complement the traditionally used metabolic features.

SUBMITTER: Qi C

PROVIDER: S-EPMC6487359 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A quantitative SVM approach potentially improves the accuracy of magnetic resonance spectroscopy in the preoperative evaluation of the grades of diffuse gliomas.

Qi Chong C Li Yiming Y Fan Xing X Jiang Yin Y Wang Rui R Yang Song S Meng Lanxi L Jiang Tao T Li Shaowu S

NeuroImage. Clinical 20190422

<h4>Objectives</h4>To investigate the association between proton magnetic resonance spectroscopy (<sup>1</sup>H-MRS) metabolic features and the grade of gliomas, and to establish a machine-learning model to predict the glioma grade.<h4>Methods</h4>This study included 112 glioma patients who were divided into the training (n = 74) and validation (n = 38) sets based on the time of hospitalization. Twenty-six metabolic features were extracted from the preoperative <sup>1</sup>H-MRS image. The Stude ...[more]

PMID: 31035232

Similar Datasets

Project description:OBJECTIVES:Vessel size imaging (VSI) could reveal average microvessel diameter. The aim was to investigate microvascular characteristics and the efficacy of VSI in lower-grade glioma (LGG) grading and subtype differentiation based on 2016 classification of central nervous system tumours. METHODS:Fifty-seven LGG (grade II/III, 36/21) patients who received VSI examination before surgery were retrospectively analysed. The average (Rmean) and maximum (Rmax) vessel size indexes were obtained. The long (VDmax) and short (VDmin) vascular diameter, microvascular area (MVA) and density (MVD) were obtained using paraffin specimens. The patients were divided into grades II and III, and histological and molecular subtypes. The differences among microvascular parameters of different subtypes and grades were compared. Two-sample t-test, analysis of variance test, Mann-Whitney test, the Kruskal-Wallis test and Pearson correlation analysis were used for statistics. RESULTS:Rmean, Rmax, VDmin, VDmax, and MVA were higher in grade-III than in grade-II LGGs (p < 0.05) in each type except the isocitrate dehydrogenase (IDH) mutant with 1p/19q-intact type. For grade II, the IDH mutant with 1p/19q co-deleted and IDH wildtype possessed more dominant angiogenesis than IDH mutant with 1p/19q-intact type, revealed by lower Rmean, Rmax and VDmin while higher MVD for the former (p < 0.05), the same as oligodendroglioma versus astrocytoma. Rmean and Rmax correlated with VDmin (r = 0.804, 0.815, p < 0.05), VDmax (r = 0.766, 0.774, p < 0.05) and MVA (r = 0.755, 0.759, p < 0.05), respectively, while they had no correlation with MVD (r = -0.085, -0.080, p > 0.05). CONCLUSIONS:VSI holds great potential for non-invasively revealing microvascular characteristics of LGGs pre-surgery and differentiating their grades and molecular subtypes. KEY POINTS:• VSI can assist in differentiating grade-II and -III gliomas. • The IDH gene and 1p/19q chromosome may influence the angiogenesis in grade-II gliomas. • VSI is valuable for differentiating the molecular subtypes of grade-II gliomas.

Project description:OBJECT:Coinfused surrogate imaging tracers can provide direct insight into the properties of convection-enhanced delivery (CED) in the nervous system. To better understand the distributive properties of CED in a clinical circumstance, the authors analyzed the imaging findings in pediatric diffuse intrinsic pontine glioma (DIPG) patients undergoing coinfusion of Gd-DTPA and interleukin-13-Pseudomonas exotoxin (IL13-PE). METHODS:Consecutive patients undergoing CED (maximal rates of 5 or 10 ?l/minute) of Gd-DTPA (1 or 5 mM) and IL13-PE (0.125 ?g/ml or 0.25 ?g/ml) for DIPG were included. Real-time MRI was performed during infusions, and imaging results were analyzed. RESULTS:Four patients (2 males, 2 females; mean age at initial infusion 13.0 ± 5.3 years; range 5-17 years) underwent 5 infusions into DIPGs. Brainstem infusions were clearly identified on T1-weighted MR images at 1-mM (1 infusion) and 5-mM (4 infusions) coinfused Gd-DTPA concentrations. While the volume of distribution (Vd) increased progressively with volume of infusion (Vi) (mean volume 2.5 ± 0.9 ml; range 1.1-3.7 ml), final Vd:Vi ratios were significantly reduced with lower Gd-DTPA concentration (Vd:Vi for 1 mM of 1.6 compared with a mean Vd:Vi ratio for 5 mM of 3.3 ± 1.0) (p = 0.04). Similarly, anatomical distribution patterns were affected by preferential flow along parallel axial fiber tracts, into prior infusion cannula tracts and intraparenchymal air pockets, and leak back around the infusion cannula at the highest rate of infusion. CONCLUSIONS:Magnetic resonance imaging of a coinfused Gd-DTPA surrogate tracer provided direct insight into the properties of CED in a clinical application. While clinically relevant Vds can be achieved by convective delivery, specific tissue properties can affect distribution volume and pattern, including Gd-DTPA concentration, preferential flow patterns, and infusion rate. Understanding of these properties of CED can enhance its clinical application. Part of clinical trial no. NCT00880061 ( ClinicalTrials.gov ).

Dataset Information

A quantitative SVM approach potentially improves the accuracy of magnetic resonance spectroscopy in the preoperative evaluation of the grades of diffuse gliomas.

Objectives

Methods

Results

Conclusions

Publications

A quantitative SVM approach potentially improves the accuracy of magnetic resonance spectroscopy in the preoperative evaluation of the grades of diffuse gliomas.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets