Predicting Isocitrate Dehydrogenase Status in Glioma Using Hierarchical Attention-Based Deep 3D Multiple Instance Learning

Qinqin Xie^1,2Yongheng Sun³Yuxia Liang⁴Yu Shang⁵Haifeng Wang³Fan Wang⁶Rong Wei¹Bin Chen⁷Ming Zhang⁸Chen Niu^1,2*

• ¹Department of Network Information, The First Affiliated Hospital of Xi’ an Jiaotong University, Xi'an, China
• ²PET-CT Center, The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China
• ³School of Mathematics and Statistics, Xi'an Jiaotong University, Xi'an, China
• ⁴Department of Health Medicine, The First Affiliated Hospital of Xi’ an Jiaotong University, Xi'an, China
• ⁵School of Future Technology, Xi'an Jiaotong University, Xi'an, China
• ⁶School of Life Science and Technology, Xi'an Jiaotong University, Xi'an, China
• ⁷Hangzhou First People's Hospital, Hangzhou, China
• ⁸Department of Radiology, The First Affiliated Hospital of Xi’ an Jiaotong University, Xi'an, China

Background: According to the 2021 WHO classification of tumors of the central nervous system, isocitrate dehydrogenase (IDH) status serve an independent prognostic biomarker and is closely associated with tumor diagnosis and treatment response. At present, the determination of IDH status still relies on invasive surgical procedures. Method: A total of 345 patients with pathologically confirmed gliomas diagnosed at the First Affiliated Hospital of Xi'an Jiaotong University between October 2019 and October 2024 were retrospectively included, comprising 148 (42.9%) IDH-wild and 197 (57.1%) IDH-mutant. An additional 495 glioma patients were obtained from the public TCIA dataset. Patients were randomly split into training, validation, and test cohorts 6:2:2. A Hierarchical Attention-Based Multiple Instance Learning (HAB-MIL) framework was developed, integrating auxiliary positional encoding into feature maps to capture spatially specific information and generate refined 3D lesion representations. Model performance was evaluated using five-fold cross-validation, with receiver operating characteristic (ROC) curves, area under the curve (AUC), sensitivity, and specificity as assessment metrics. Result: HAB-MIL achieved competitive performance, with AUCs of 0.917 and 0.892 on the glioma datasets from TCIA and the First Affiliated Hospital of Xi'an Jiaotong University. Additionally, our work achieves results that are comparable to the state-of-the-art methods in TCIA dataset and demonstrates that multiple instance learning has great potential for IDH prediction. Conclusion: The proposed HAB-MIL achieved IDH classification based on conventional preoperative MRI images, eliminating the need for pixel-level annotations and significantly reducing the annotation burden for doctors.