Fan Yu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Fan Yu has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Jie Lu
Department of Radiology and Nuclear Medicine, Xuanwu Hospital Capital Medical University, Beijing, China
3 shared publications - Yue Zhang
Department of Radiology and Nuclear Medicine, Xuanwu Hospital Capital Medical University, Changchun Street No. 45, Beijing 100053, China
2 shared publications - Bin Yang
Department of Neurosurgery, Xuanwu Hospital, Capital Medical University, Beijing, China
1 shared publication - Bixiao Cui
Department of Radiology and Nuclear Medicine, Xuan Wu Hospital of the Capital Medical University, Beijing, China
1 shared publication - Bo Cao
Fuwai Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College
1 shared publication - Chong Zheng
Department of Radiology and Nuclear Medicine, Xuanwu Hospital, Capital Medical University, Beijing, China
1 shared publication
Latest Publications
- Open Access
- 28 August 2026
- Published by MDPI AG in Bioengineering
- Vol. 13 (9) , 1005
Annotation-Efficient Vulnerable Carotid Plaque Identification in 3D MRI: A Multimodal Knowledge-Transferred Framework
Multimodal learning has gained attention in recent years due to its ability to effectively utilize data features from various modalities. Diagnosing the vulnerability of atherosclerotic plaques directly from carotid 3D MRI images is challenging for both radiologists and conventional 3D vision networks. In clinical practice, radiologists assess patients using a multimodal approach that incorporates various imaging modalities and domain-specific expertise, paving the way for the creation of multimodal diagnostic networks. In this study, we proposed an effective framework to leverage radiologists’ domain knowledge to improve the automated diagnosis of carotid plaque vulnerability through variational inference and multimodal knowledge distillation (VMD). This framework excels in harnessing cross-modality prior knowledge from limited image annotations and radiology reports within training data, thereby enhancing the diagnostic network’s accuracy for unannotated 3D MRI images. We validated the proposed VMD framework on our in-house dataset, demonstrating its effectiveness.
- 28 August 2026
- Published by MDPI AG in Bioengineering
- Vol. 13 (9) , 1005
Open AccessAnnotation-Efficient Vulnerable Carotid Plaque Identification in 3D MRI: A Multimodal Knowledge-Transferred Framework
Multimodal learning has gained attention in recent years due to its ability to effectively utilize data features from various modalities. Diagnosing the vulnerability of atherosclerotic plaques directly from carotid 3D MRI images is challenging for both radiologists and conventional 3D vision networks. In clinical practice, radiologists assess patients using a multimodal approach that incorporates various imaging modalities and domain-specific expertise, paving the way for the creation of multimodal diagnostic networks. In this study, we proposed an effective framework to leverage radiologists’ domain knowledge to improve the automated diagnosis of carotid plaque vulnerability through variational inference and multimodal knowledge distillation (VMD). This framework excels in harnessing cross-modality prior knowledge from limited image annotations and radiology reports within training data, thereby enhancing the diagnostic network’s accuracy for unannotated 3D MRI images. We validated the proposed VMD framework on our in-house dataset, demonstrating its effectiveness.
- Open Access
- 13 May 2024
- Published by MDPI AG in Diagnostics
- Vol. 14 (10) , 1006
In Vivo Classification and Characterization of Carotid Atherosclerotic Lesions with Integrated 18F-FDG PET/MRI
Background: The aim of this study was to exploit integrated PET/MRI to simultaneously evaluate the morphological, component, and metabolic features of advanced atherosclerotic plaques and explore their incremental value. Methods: In this observational prospective cohort study, patients with advanced plaque in the carotid artery underwent 18F-FDG PET/MRI. Plaque morphological features were measured, and plaque component features were determined via MRI according to AHA lesion-types. Maximum standardized uptake values (SUVmax) and tissue to background ratio (TBR) on PET were calculated. Area under the receiver-operating characteristic curve (AUC) and net reclassification improvement (NRI) were used to compare the incremental contribution of FDG uptake when added to AHA lesion-types for symptomatic plaque classification. Results: A total of 280 patients with advanced plaque in the carotid artery were recruited. A total of 402 plaques were confirmed, and 87 of 402 (21.6%) were symptomatic plaques. 18F-FDG PET/MRI was performed a mean of 38 days (range 1–90) after the symptom. Increased stenosis degree (61.5% vs. 50.0%, p < 0.001) and TBR (2.96 vs. 2.32, p < 0.001) were observed in symptomatic plaques compared with asymptomatic plaques. The performance of the combined model (AHA lesion type VI + stenosis degree + TBR) for predicting symptomatic plaques was the best among all models (AUC = 0.789). The improvement of the combined model (AHA lesion type VII + stenosis degree + TBR) over AHA lesion type VII model for predicting symptomatic plaques was the highest (AUC = 0.757/0.454, combined model/AHA lesion type VII model), and the NRI was 50.7%. Conclusions: Integrated PET/MRI could simultaneously evaluate the morphological component and inflammation features of advanced atherosclerotic plaques and provide supplementary optimization information over AHA lesion-types for identifying vulnerable plaques in atherosclerosis subjects to achieve further stratification of stroke risk.
- 13 May 2024
- Published by MDPI AG in Diagnostics
- Vol. 14 (10) , 1006
Open AccessIn Vivo Classification and Characterization of Carotid Atherosclerotic Lesions with Integrated 18F-FDG PET/MRI
Background: The aim of this study was to exploit integrated PET/MRI to simultaneously evaluate the morphological, component, and metabolic features of advanced atherosclerotic plaques and explore their incremental value. Methods: In this observational prospective cohort study, patients with advanced plaque in the carotid artery underwent 18F-FDG PET/MRI. Plaque morphological features were measured, and plaque component features were determined via MRI according to AHA lesion-types. Maximum standardized uptake values (SUVmax) and tissue to background ratio (TBR) on PET were calculated. Area under the receiver-operating characteristic curve (AUC) and net reclassification improvement (NRI) were used to compare the incremental contribution of FDG uptake when added to AHA lesion-types for symptomatic plaque classification. Results: A total of 280 patients with advanced plaque in the carotid artery were recruited. A total of 402 plaques were confirmed, and 87 of 402 (21.6%) were symptomatic plaques. 18F-FDG PET/MRI was performed a mean of 38 days (range 1–90) after the symptom. Increased stenosis degree (61.5% vs. 50.0%, p < 0.001) and TBR (2.96 vs. 2.32, p < 0.001) were observed in symptomatic plaques compared with asymptomatic plaques. The performance of the combined model (AHA lesion type VI + stenosis degree + TBR) for predicting symptomatic plaques was the best among all models (AUC = 0.789). The improvement of the combined model (AHA lesion type VII + stenosis degree + TBR) over AHA lesion type VII model for predicting symptomatic plaques was the highest (AUC = 0.757/0.454, combined model/AHA lesion type VII model), and the NRI was 50.7%. Conclusions: Integrated PET/MRI could simultaneously evaluate the morphological component and inflammation features of advanced atherosclerotic plaques and provide supplementary optimization information over AHA lesion-types for identifying vulnerable plaques in atherosclerosis subjects to achieve further stratification of stroke risk.
- 1 April 2023
- Published by Wolters Kluwer Health in Stroke
- Vol. 54 (4) , 1037-1045
Multidelay Arterial Spin Labeling Versus Computed Tomography Perfusion in Penumbra Volume of Acute Ischemic Stroke
Background:Multidelay arterial spin labeling (ASL) is a novel perfusion method of ASL, with arterial transit time (ATT) calculated by multiple postlabeling delays to correct cerebral blood flow (CBF). We verify the accuracy of multidelay ASL in evaluating the ischemic penumbra and perfusion levels in patients with acute ischemic stroke, compared with computed tomography perfusion (CTP). Methods:Patients with acute ischemic stroke with anterior circulation large vessel occlusion received baseline CTP, multidelay ASL, and diffusion-weighted imaging (DWI) in succession. Multidelay ASL image was processed to reconstruct ATT, CBF without ATT correction, and CBF corrected by ATT. The consistency of hypoperfusion and ischemic penumbra volume calculated by CTP and multidelay ASL were quantified by intraclass correlation coefficient (ICC) in 2-way mixed effects, absolute agreement, and single measure. Wilcoxon signed-rank test was used to compare the difference in penumbra volume between CTP, corrected ASL, and uncorrected ASL. Results:Thirty patients were included. Hypoperfusion volume based on multidelay ASL with different thresholds were 117.95 (87.77-151.49) mL for corrected relative CBF<40%, 130.29 (85.99-249.37) mL for CBF corrected by ATT0.05) compared with the volume of CTP, and consistency was almost excellent (ICC, 0.91) and substantial consistent (ICC, 0.727). The volumes of ischemic penumbra were 91.00 (42.68-125.27) mL for corrected relative CBF<40%-DWI, 108.94 (62.03-150.86) mL for CBF corrected by ATT0.05). The consistency was excellent (ICC, 0.822) and moderate (ICC, 0.501), respectively. The volume of uncorrected relative CBF <40%-DWI was 209.57 (123.21-292.45) mL, statistically larger than corrected relative CBF <40%-DWI (P<0.001) and CTP (P<0.001). The volume of uncorrected CBF<20 mL center dot 100g(-1)center dot min(-1)-DWI was 186.23 (86.56-298.22) mL, statistically larger than CBF corrected by ATT<20 mL center dot 100g(-1)center dot min(-1)-DWI (P<0.001) and CTP(P<0.001). Conclusions:The volume of ischemic penumbra determined by CBF/DWI mismatch based on multidelay ASL is consistent with CTP. The penumbra volume calculated by CBF adjusted by ATT is more accurate.
- 1 April 2023
- Published by Wolters Kluwer Health in Stroke
- Vol. 54 (4) , 1037-1045
Multidelay Arterial Spin Labeling Versus Computed Tomography Perfusion in Penumbra Volume of Acute Ischemic Stroke
Background:Multidelay arterial spin labeling (ASL) is a novel perfusion method of ASL, with arterial transit time (ATT) calculated by multiple postlabeling delays to correct cerebral blood flow (CBF). We verify the accuracy of multidelay ASL in evaluating the ischemic penumbra and perfusion levels in patients with acute ischemic stroke, compared with computed tomography perfusion (CTP). Methods:Patients with acute ischemic stroke with anterior circulation large vessel occlusion received baseline CTP, multidelay ASL, and diffusion-weighted imaging (DWI) in succession. Multidelay ASL image was processed to reconstruct ATT, CBF without ATT correction, and CBF corrected by ATT. The consistency of hypoperfusion and ischemic penumbra volume calculated by CTP and multidelay ASL were quantified by intraclass correlation coefficient (ICC) in 2-way mixed effects, absolute agreement, and single measure. Wilcoxon signed-rank test was used to compare the difference in penumbra volume between CTP, corrected ASL, and uncorrected ASL. Results:Thirty patients were included. Hypoperfusion volume based on multidelay ASL with different thresholds were 117.95 (87.77-151.49) mL for corrected relative CBF<40%, 130.29 (85.99-249.37) mL for CBF corrected by ATT0.05) compared with the volume of CTP, and consistency was almost excellent (ICC, 0.91) and substantial consistent (ICC, 0.727). The volumes of ischemic penumbra were 91.00 (42.68-125.27) mL for corrected relative CBF<40%-DWI, 108.94 (62.03-150.86) mL for CBF corrected by ATT0.05). The consistency was excellent (ICC, 0.822) and moderate (ICC, 0.501), respectively. The volume of uncorrected relative CBF <40%-DWI was 209.57 (123.21-292.45) mL, statistically larger than corrected relative CBF <40%-DWI (P<0.001) and CTP (P<0.001). The volume of uncorrected CBF<20 mL center dot 100g(-1)center dot min(-1)-DWI was 186.23 (86.56-298.22) mL, statistically larger than CBF corrected by ATT<20 mL center dot 100g(-1)center dot min(-1)-DWI (P<0.001) and CTP(P<0.001). Conclusions:The volume of ischemic penumbra determined by CBF/DWI mismatch based on multidelay ASL is consistent with CTP. The penumbra volume calculated by CBF adjusted by ATT is more accurate.