Deep learning-driven modality imputation and subregion segmentation to enhance high-grade glioma grading.
Authors
Affiliations (20)
Affiliations (20)
- Taizhou Campus, Zhejiang Cancer Hospital (Taizhou Cancer Hospital), Taizhou, Zhejiang, 317502, China.
- College of Information Engineering, China Jiliang University, Hangzhou, Zhejiang, 310018, China.
- State Key Laboratory of Cardiology and Medical Innovation Center, Frontier Science Center for Stem Cell Research, Bioinformatics Department, School of Life Sciences and Technology, Shanghai East Hospital, Tongji University, Shanghai, 200092, China.
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Department of Radiology Imaging, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), No. 50 Zhenxin Road, Xinhe Town, Wenling, Zhejiang, 317502, China.
- Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Hangzhou Institute of Medicine (HIM), Zhejiang Cancer Hospital, Chinese Academy of Sciences, No.1 East Banshan Road, Gongshu District, Hangzhou, Zhejiang, 310022, China.
- Wenling Institute of Big Data and Artificial Intelligence in Medicine, Level 24, Building 1, XinShang Building, Xinhe Town, Wenling, Zhejiang, 317502, China.
- Center of Intelligent Diagnosis and Therapy (Taizhou), Hangzhou Institute of Medicine(HIM), Chinese Academy of Sciences, Taizhou, Zhejiang, 317502, China.
- Department of Radiation Oncology, The Second Affiliated Hospital, National Ministry of Education Key Laboratory of Cancer Prevention and Intervention, Zhejiang University School of Medicine, Hangzhou, Zhejiang, 310009, China.
- Department of Ultrasound, Zhangzhou Affiliated Hospital of Fujian Medical University, Fujian, 363000, China.
- Radiation Oncology Center, Shanghai Medical College, Huashan Hospital, Fudan University, No.12 Wulumuqi Middle Road, Shanghai, 201107, China.
- Department of Computer Science and Technology (Sino-American Joint Program), School of Information Science and Technology, Northeast Normal University, Changchun, Jilin, 130117, China.
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Department of Radiology Imaging, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), No. 50 Zhenxin Road, Xinhe Town, Wenling, Zhejiang, 317502, China. [email protected].
- Wenling Institute of Big Data and Artificial Intelligence in Medicine, Level 24, Building 1, XinShang Building, Xinhe Town, Wenling, Zhejiang, 317502, China. [email protected].
- Center of Intelligent Diagnosis and Therapy (Taizhou), Hangzhou Institute of Medicine(HIM), Chinese Academy of Sciences, Taizhou, Zhejiang, 317502, China. [email protected].
- Taizhou Campus, Zhejiang Cancer Hospital (Taizhou Cancer Hospital), Taizhou, Zhejiang, 317502, China. [email protected].
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Department of Radiology Imaging, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), No. 50 Zhenxin Road, Xinhe Town, Wenling, Zhejiang, 317502, China. [email protected].
- Taizhou Campus, Zhejiang Cancer Hospital (Taizhou Cancer Hospital), Taizhou, Zhejiang, 317502, China. [email protected].
- Taizhou Key Laboratory of Minimally Invasive Interventional Therapy & Artificial Intelligence, Department of Radiology Imaging, Taizhou Campus of Zhejiang Cancer Hospital (Taizhou Cancer Hospital), No. 50 Zhenxin Road, Xinhe Town, Wenling, Zhejiang, 317502, China. [email protected].
- Department of Diagnostic Ultrasound Imaging & Interventional Therapy, Hangzhou Institute of Medicine (HIM), Zhejiang Cancer Hospital, Chinese Academy of Sciences, No.1 East Banshan Road, Gongshu District, Hangzhou, Zhejiang, 310022, China. [email protected].
- Wenling Institute of Big Data and Artificial Intelligence in Medicine, Level 24, Building 1, XinShang Building, Xinhe Town, Wenling, Zhejiang, 317502, China. [email protected].
Abstract
This study aims to develop a deep learning framework that leverages modality imputation and subregion segmentation to improve grading accuracy in high-grade gliomas. A retrospective analysis was conducted using data from 1,251 patients in the BraTS2021 dataset as the main cohort and 181 clinical cases collected from a medical center between April 2013 and June 2018 (51 years ± 17; 104 males) as the external test set. We propose a PatchGAN-based modality imputation network with an Aggregated Residual Transformer (ART) module combining Transformer self-attention and CNN feature extraction via residual links, paired with a U-Net variant for segmentation. Generative accuracy used PSNR and SSIM for modality conversions, while segmentation performance was measured with DSC and HD95 across necrotic core (NCR), edema (ED), and enhancing tumor (ET) regions. Senior radiologists conducted a comprehensive Likert-based assessment, with diagnostic accuracy evaluated by AUC. Statistical analysis was performed using the Wilcoxon signed-rank test and the DeLong test. The best source-target modality pairs for imputation were T1 to T1ce and T1ce to T2 (p < 0.001). In subregion segmentation, the overall DSC was 0.878 and HD95 was 19.491, with the ET region showing the highest segmentation accuracy (DSC: 0.877, HD95: 12.149). Clinical validation revealed an improvement in grading accuracy by the senior radiologist, with the AUC increasing from 0.718 to 0.913 (P < 0.001) when using the combined imputation and segmentation models. The proposed deep learning framework improves high-grade glioma grading by modality imputation and segmentation, aiding the senior radiologist and offering potential to advance clinical decision-making.