文献类型：期刊文献

英文题名：Hybrid offline and self-knowledge distillation for acute ischemic stroke lesion segmentation from non-contrast CT scans

作者：Wang, Ziying[1]; Zhu, Hongqing[1]; Liu, Jiahao[1]; Chen, Ning[1]; Huang, Bingcang[2]; Lu, Weiping[2]; Wang, Ying[3]

机构：[1] School of Information Science and Engineering, East China University of Science and Technology, Shanghai, 200237, China; [2] Department of Radiology, Gongli Hospital of Shanghai Pudong New Area, Shanghai, 200135, China; [3] Shanghai Health Commission Key Lab of Artificial Intelligence [AI]-Based Management of Inflammation and Chronic Diseases, Sino-French Cooperative Central Lab, Gongli Hospital of Shanghai Pudong New Area, Shanghai, 200135, China

年份：2024

卷号：183

外文期刊名：Computers in Biology and Medicine

收录：EI(收录号：20244517307574)

语种：英文

外文关键词：Brain mapping - Image segmentation - Majority logic - Personnel training - Students - Teaching

摘要：Diagnosing and treating Acute Ischemic Stroke (AIS) within 0-24 h of onset is critical for patient recovery. While Diffusion-Weighted Imaging (DWI) and Computed Tomography Perfusion (CTP) are effective for early infarction identification, Non-Contrast CT (NCCT) remains the first-line imaging modality in emergency settings due to its efficiency and cost-effectiveness. In this work, to enhance lesion segmentation in NCCT using multi-modal information, we propose OS-AISeg, which integrates Offline knowledge distillation with Self-knowledge distillation to realize AIS segmentation. Initially, we trained a multi-modality teacher network by introducing uncertainty through Subjective Logic (SL) theory to reduce prediction errors and stabilize the training process. Subsequently, during student network training, we integrate confidence region knowledge guided by uncertainty weights and feature structure information guided by brain asymmetry. The former facilitates the acquisition of effective contextual information from paired predictions, while the latter leverages asymmetric activation maps to extract high-level structural content from multi-modality images. In self-knowledge distillation, we enhance the student network's learning of consistent global feature distributions by introducing mirrored NCCT images, thereby aiding the network in extracting knowledge directly from the modality. OS-AISeg was evaluated through five-fold cross-validation on two publicly available datasets, achieving a Dice value of 0.6196 on AISD and 0.4841 on ISLES2018. Additionally, experiments were also conducted on an external dataset, BraTS2019, as well as on a private stroke dataset named GLis. Strong correlations were observed between segmented Early Infarct (EI) and ground truth in volume analysis, validating the effectiveness of the proposed method in AIS diagnosis. The code for this project is available at https://github.com/Uni-Summer/OS-AISeg. ? 2024 Elsevier Ltd