Related Experiment Video
Updated: Jan 17, 2026
Digital Spatial Profiling for Characterization of the Microenvironment in Adult-Type Diffusely Infiltrating Glioma
Published on: September 13, 2022
Histogram analysis of quantitative susceptibility mapping and diffusion kurtosis imaging for the grading prediction
Pengxin Shen1, Yangyang Li1, Wenji Xu1
1Department of Radiology, First Hospital of Shanxi Medical University, Taiyuan, Shanxi Province 030001, PR China; College of Medical Imaging, Shanxi Medical University, Taiyuan, Shanxi Province 030001, PR China.
Objectives:
To evaluate the role of quantitative susceptibility mapping (QSM) and diffusion kurtosis imaging (DKI) histogram features in improving the 2021 World Health Organization Classification of Central Nervous System Tumors (WHO CNS 5) grading accuracy for adult-type diffuse gliomas when combined with conventional imaging sequences.
Methods:
A total of 62 patients were retrospectively collected. Histogram features of QSM, DKI, CE-T1WI, T2FLAIR were extracted from tumor parenchyma. Independent t-tests and Mann-Whitney U tests were used to compare differences between grade 2/3 and grade 4 gliomas. The evaluation of the model included receiver operating characteristic (ROC) curves, 5-fold cross-validation, nomogram construction, and calibration curve analysis. Prognostic differences between two groups were assessed using Kaplan-Meier survival analysis.
Results:
The functional imaging model (AUC = 0.892) was constructed using QSM mean absolute deviation (MAD) and relative mean kurtosis 90th percentile (rMK P90), while the conventional imaging model (AUC = 0.776) was built using CE-T1WI robust mean absolute deviation (RMAD) and T2FLAIR maximum. The imaging combined model, incorporating CE-T1WI RMAD, QSM MAD, and rMK P90, achieved an AUC of 0.936. Among the clinical factors, age showed a statistically significant difference between the two groups, with an AUC of 0.769. The integrated model combining the imaging model and age achieved the highest AUC of 0.949. The 5-fold internal cross validation showed that the average AUC was 0.944. Survival analysis revealed a significant difference between grade 2/3 and grade 4 gliomas.
Conclusion:
Histogram features of QSM and DKI can complement conventional sequences, enhancing the diagnostic performance for adult-type diffuse gliomas grading.
