Habitat Analysis of the Peritumoral Edema Region Combined with Visually Accessible Rembrandt Images Features and the Intratumoral Heterogeneity Score for Preoperative Prediction of Isocitrate Dehydrogenase Status in Adult-Type Diffuse Glioma: A Two-C

Purpose: To develop and validate magnetic resonance imaging (MRI)-based machine learning models integrating habitat analysis of peritumoral edema regions, Visually Accessible Rembrandt Images (VASARI) features and intratumoral heterogeneity (ITH) score for preoperative prediction of isocitrate dehydrogenase (IDH) status in adult-type diffuse glioma. Methods: This retrospective study included 606 patients with adult-type diffuse glioma from two centers (376 for development; 230 for external validation). The peritumoral edema region was divided into two habitats (H1 and H2) using K -means clustering. Habitat-derived radiomics features, VASARI features, and the ITH score were used to construct five models: H1, H2, H1+H2, VASARI + ITH, and a comprehensive fusion model. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity. Shapley Additive Explanations (SHAP) method was employed for interpretability analysis. Results The H2 model outperformed H1 model (AUC, 0.884 vs 0.773). The VASARI + ITH model achieved an AUC of 0.841, whereas the H1 + H2 model achieved an AUC of 0.890. The comprehensive fusion model exhibited the best AUC (0.892), accuracy, sensitivity, and specificity. SHAP analysis revealed that habitat-derived radiomic features contributed most to model prediction. Conclusion A fusion model combining habitat-derived radiomics features from the glioma peritumoral edema region, VASARI features, and the ITH score demonstrated favorable predictive performance for preoperative prediction of IDH status in adult-type diffuse glioma.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-20
DOI
https://doi.org/10.21203/rs.3.rs-11111225/v1
Primary Topic
Radiomics and Machine Learning in Medical Imaging
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Habitat Analysis of the Peritumoral Edema Region Combined with Visually Accessible Rembrandt Images Features and the Intratumoral Heterogeneity Score for Preoperative Prediction of Isocitrate Dehydrogenase Status in Adult-Type Diffuse Glioma: A Two-C

Yunling Wang, Rui Xu, Wei Zhao, Sijia Li et al.
5 links
Research Square
Radiomics and Machine Learning in Medical Imaging
preprint

Habitat Analysis of the Peritumoral Edema Region Combined with Visually Accessible Rembrandt Images Features and the Intratumoral Heterogeneity Score for Preoperative Prediction of Isocitrate Dehydrogenase Status in Adult-Type Diffuse Glioma: A Two-C

Yunling Wang, Rui Xu, Wei Zhao, Sijia Li, Jingming Jiang, Kukun Hangaerbieke¹, Subi Yimuran¹
preprint en

Abstract

Purpose: To develop and validate magnetic resonance imaging (MRI)-based machine learning models integrating habitat analysis of peritumoral edema regions, Visually Accessible Rembrandt Images (VASARI) features and intratumoral heterogeneity (ITH) score for preoperative prediction of isocitrate dehydrogenase (IDH) status in adult-type diffuse glioma. Methods: This retrospective study included 606 patients with adult-type diffuse glioma from two centers (376 for development; 230 for external validation). The peritumoral edema region was divided into two habitats (H1 and H2) using K -means clustering. Habitat-derived radiomics features, VASARI features, and the ITH score were used to construct five models: H1, H2, H1+H2, VASARI + ITH, and a comprehensive fusion model. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity. Shapley Additive Explanations (SHAP) method was employed for interpretability analysis. Results The H2 model outperformed H1 model (AUC, 0.884 vs 0.773). The VASARI + ITH model achieved an AUC of 0.841, whereas the H1 + H2 model achieved an AUC of 0.890. The comprehensive fusion model exhibited the best AUC (0.892), accuracy, sensitivity, and specificity. SHAP analysis revealed that habitat-derived radiomic features contributed most to model prediction. Conclusion A fusion model combining habitat-derived radiomics features from the glioma peritumoral edema region, VASARI features, and the ITH score demonstrated favorable predictive performance for preoperative prediction of IDH status in adult-type diffuse glioma.

Research Square
Xinjiang Medical University (CN), First Affiliated Hospital of Xinjiang Medical University (CN)
Good health and well-being
Radiomics and Machine Learning in Medical Imaging
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon