Fan Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Fan Liu has a total of 48 co-authors. Here are the top co-authors by number of shared publications:
- Diwei Shi
Research Institute of Tsinghua University in Shenzhen Shenzhen China
4 shared publications - Hua Guo
Center for Biomedical Imaging Research, School of Biomedical Engineering, Tsinghua University, Beijing, China
3 shared publications - Hao Sun
Department of Radiology, State Key Laboratory of Complex Severe and Rare Disease, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing, China
2 shared publications Huadan Xue
Department of Radiology, State Key Laboratory of Complex Severe and Rare Disease, Peking Union Medical College Hospital Chinese Academy of Medical Sciences and Peking Union Medical College Beijing People's Republic of China
2 shared publications- Junzhong Xu
Vanderbilt University Institute of Imaging Science, Vanderbilt University Medical Center Nashville Tennessee USA
2 shared publications - Thorsten Feiweier
Siemens Healthineers AG, Erlangen, Germany
2 shared publications
Latest Publications
- 11 January 2026
- Published by Wiley in Journal of Magnetic Resonance Imaging
- Vol. 64 (1) , 77-87
The Performance of MR Cytometry Imaging in Differentiating High‐ and Low‐Grade Bladder Cancer
Background: Accurate preoperative grading of bladder cancer is important for determining treatment and prognosis. Purpose: To investigate the diagnostic efficacy of MR cytometry imaging in differentiating high‐ and low‐grade bladder cancer. Study Type: Prospective. Population: Sixty‐participants (male: 27, mean age: 65 years) with pathologically confirmed bladder cancer (37 high‐grade, 23 low‐grade). Field Strength/Sequence: 3.0 T, pulsed gradient spin‐echo ( PGSE ) and oscillating gradient spin‐echo ( OGSE , 20 and 40 Hz) diffusion‐weighted imaging. Assessment: All tumors were manually delineated independently by two radiologists, and inter‐observer agreement was assessed using intraclass correlation coefficient (ICC). Time‐dependent apparent diffusion coefficients (ADCs), including OGSE at 20 HZ (ADC 20HZ ), OGSE at 40 HZ (ADC 40HZ ), and PGSE (ADC PGSE ), and MR cytometry‐derived microstructural parameters (cell diameter [], intracellular volume fraction [], extracellular diffusivity [], and cellularity []) were calculated. Histopathological examination of surgical specimens served as the reference standard for tumor grading. Statistical Tests: Mann–Whitney U test was used for group comparisons. Diagnostic performance was evaluated by logistic regression and receiver operating characteristic (ROC) analysis; area under the ROC curve (AUCs) was compared with the DeLong test. Statistical significance was set at p < 0.05. Results: High‐grade tumors showed significantly higher (median: 0.31 vs. 0.20), (1.97 vs. 1.33 × 10 −2 μm −1 ), and lower ADCs than low‐grade tumors while ( p = 0.85, 95% confidence interval [CI] of mean difference: −0.822 to −0.820) and ( p = 0.053, 95% CI of mean difference: 0.025 to 0.352) were not different. demonstrated the highest AUC (0.89; 95% CI: 0.80–0.97) among single parameters, and the combined model of , , and ADC PGSE achieved the highest diagnostic accuracy (AUC = 0.92; 95% CI: 0.86–0.99). Data Conclusion: MR cytometry noninvasively differentiates high‐ from low‐grade bladder cancer. showed good discriminatory performance, and combining , , and ADC PGSE further improves preoperative assessment. Evidence Level: 1. Technical Efficacy: Stage 3: Diagnostic Thinking.
- 11 January 2026
- Published by Wiley in Journal of Magnetic Resonance Imaging
- Vol. 64 (1) , 77-87
The Performance of MR Cytometry Imaging in Differentiating High‐ and Low‐Grade Bladder Cancer
Background: Accurate preoperative grading of bladder cancer is important for determining treatment and prognosis. Purpose: To investigate the diagnostic efficacy of MR cytometry imaging in differentiating high‐ and low‐grade bladder cancer. Study Type: Prospective. Population: Sixty‐participants (male: 27, mean age: 65 years) with pathologically confirmed bladder cancer (37 high‐grade, 23 low‐grade). Field Strength/Sequence: 3.0 T, pulsed gradient spin‐echo ( PGSE ) and oscillating gradient spin‐echo ( OGSE , 20 and 40 Hz) diffusion‐weighted imaging. Assessment: All tumors were manually delineated independently by two radiologists, and inter‐observer agreement was assessed using intraclass correlation coefficient (ICC). Time‐dependent apparent diffusion coefficients (ADCs), including OGSE at 20 HZ (ADC 20HZ ), OGSE at 40 HZ (ADC 40HZ ), and PGSE (ADC PGSE ), and MR cytometry‐derived microstructural parameters (cell diameter [], intracellular volume fraction [], extracellular diffusivity [], and cellularity []) were calculated. Histopathological examination of surgical specimens served as the reference standard for tumor grading. Statistical Tests: Mann–Whitney U test was used for group comparisons. Diagnostic performance was evaluated by logistic regression and receiver operating characteristic (ROC) analysis; area under the ROC curve (AUCs) was compared with the DeLong test. Statistical significance was set at p < 0.05. Results: High‐grade tumors showed significantly higher (median: 0.31 vs. 0.20), (1.97 vs. 1.33 × 10 −2 μm −1 ), and lower ADCs than low‐grade tumors while ( p = 0.85, 95% confidence interval [CI] of mean difference: −0.822 to −0.820) and ( p = 0.053, 95% CI of mean difference: 0.025 to 0.352) were not different. demonstrated the highest AUC (0.89; 95% CI: 0.80–0.97) among single parameters, and the combined model of , , and ADC PGSE achieved the highest diagnostic accuracy (AUC = 0.92; 95% CI: 0.86–0.99). Data Conclusion: MR cytometry noninvasively differentiates high‐ from low‐grade bladder cancer. showed good discriminatory performance, and combining , , and ADC PGSE further improves preoperative assessment. Evidence Level: 1. Technical Efficacy: Stage 3: Diagnostic Thinking.
- 22 December 2025
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 95 (5) , 2594-2610
Robust Fat Suppression for High‐Resolution DWI at 5 T Using Slice‐Selection Gradient Modulation and Chemical Shift Encoding
Purpose: Chemical shift encoding (Dixon) can simultaneously remove multiple fat peaks in multi‐shot EPI diffusion‐weighted imaging (ms‐ EPI DWI ) at 3 T with low sensitivity to B 0 inhomogeneity. However, this method is not applicable at 5 T , since increased fat off‐resonance frequencies cause slice position mismatches between different fat peaks. In this work, we propose a two‐step strategy combining slice‐selection gradient modulation ( SSGM ) and Dixon to enhance fat suppression for ms‐ EPI DWI at 5 T. Methods: In the first step, SSGM adjusts the amplitudes of excitation and refocusing slice‐selection gradients according to the off‐resonance frequencies of methyl/methylene fat peaks, so that these fat slices are excited but not refocused. In the second step, the olefinic fat peak is chemical shift encoded and separated from the diffusion‐weighted water images through a joint water/fat separation algorithm with structured low‐rank regularization. The two‐step strategy was evaluated in the leg, head‐and‐neck, and prostate. Results: In vivo experiments demonstrated that the Dixon‐only methods cannot simultaneously suppress all fat peaks at 5 T , while this problem was addressed by combining SSGM and Dixon. SSGM showed superior suppression for methyl/methylene fat compared to SPAIR . The following Dixon further removed olefinic fat untouched by SPAIR . Qualitative analysis showed improved overall image quality for all anatomies. Prostate experiments showed that the proposed method is also applicable in reduced FOV acquisitions, high‐resolution (1.6‐mm isotropic) and high b value imaging (2800 s/mm 2 ). Conclusion: The proposed two‐step strategy improved fat suppression in ms‐ EPI DWI at 5 T , which can potentially enhance whole‐body disease screening and diagnosis.
- 22 December 2025
- Published by Wiley in Magnetic Resonance in Medicine
- Vol. 95 (5) , 2594-2610
Robust Fat Suppression for High‐Resolution DWI at 5 T Using Slice‐Selection Gradient Modulation and Chemical Shift Encoding
Purpose: Chemical shift encoding (Dixon) can simultaneously remove multiple fat peaks in multi‐shot EPI diffusion‐weighted imaging (ms‐ EPI DWI ) at 3 T with low sensitivity to B 0 inhomogeneity. However, this method is not applicable at 5 T , since increased fat off‐resonance frequencies cause slice position mismatches between different fat peaks. In this work, we propose a two‐step strategy combining slice‐selection gradient modulation ( SSGM ) and Dixon to enhance fat suppression for ms‐ EPI DWI at 5 T. Methods: In the first step, SSGM adjusts the amplitudes of excitation and refocusing slice‐selection gradients according to the off‐resonance frequencies of methyl/methylene fat peaks, so that these fat slices are excited but not refocused. In the second step, the olefinic fat peak is chemical shift encoded and separated from the diffusion‐weighted water images through a joint water/fat separation algorithm with structured low‐rank regularization. The two‐step strategy was evaluated in the leg, head‐and‐neck, and prostate. Results: In vivo experiments demonstrated that the Dixon‐only methods cannot simultaneously suppress all fat peaks at 5 T , while this problem was addressed by combining SSGM and Dixon. SSGM showed superior suppression for methyl/methylene fat compared to SPAIR . The following Dixon further removed olefinic fat untouched by SPAIR . Qualitative analysis showed improved overall image quality for all anatomies. Prostate experiments showed that the proposed method is also applicable in reduced FOV acquisitions, high‐resolution (1.6‐mm isotropic) and high b value imaging (2800 s/mm 2 ). Conclusion: The proposed two‐step strategy improved fat suppression in ms‐ EPI DWI at 5 T , which can potentially enhance whole‐body disease screening and diagnosis.
- 31 October 2025
- Published by Springer Nature in Abdominal Radiology
- Vol. 51 (6) , 2993-3003
Improving the diagnostic efficacy of MR cytometry in prostate cancer imaging
Objectives To evaluate the improved diagnostic efficacy of the emerging non-invasive microstructure imaging technique, MR cytometry, in differentiating benign and malignant prostate lesions, after removing perfusion effect and introducing histogram features. Methods This study included 33 patients with prostate cancer and 39 patients with benign lesions, who underwent time-dependent diffusion MRI examinations, including pulsed gradient spin-echo (PGSE) diffusion-weighted imaging (DWI) and oscillating gradient spin-echo (OGSE) DWI at frequencies of 20 Hz and 40 Hz. A mono-exponential signal-fitting method was used to eliminate perfusion effects. Apparent diffusion coefficients (ADC), MR-cytometry-derived microstructural parameters (cell diameter \(\:\text{d}\), intracellular volume fraction \(\:{\text{v}}_{\text{i}\text{n}}\), extracellular diffusivity \(\:{D}_{ex}\) and cellularity \(\:{\uprho\:}\)) and corresponding histogram features were calculated to differentiate lesions. Mann-Whitney U-tests were used to evaluate the differences between benign and malignant lesions. Uni- and multi-variate logistic regressions were performed and calculated the area under the receiver operating characteristic (AUC) to quantify the diagnostic performance of classifiers. Results Removal of perfusion effect improved the significance level of the difference in the cell diameter \(\:d\) (from \(\:\text{p}>0.05\) to \(\:\text{p}<0.01\)), where \(\:\text{d}\) is larger in the benign group. The diagnostic efficacy of a single histogram feature was comparable to that of the corresponding ADC or microstructural parameter mean value. Both removing perfusion effect and incorporating histogram features can improve the AUC of the combined regression models (from 0.885 to 0.918 and 0.908, respectively). Combining both strategies further enhanced the AUC to 0.934. Conclusion In MR cytometry analysis, removing the perfusion effect and incorporating histogram features of ADCs and microstructural parameters can provide better diagnostic efficacy.- 31 October 2025
- Published by Springer Nature in Abdominal Radiology
- Vol. 51 (6) , 2993-3003
Improving the diagnostic efficacy of MR cytometry in prostate cancer imaging
Objectives To evaluate the improved diagnostic efficacy of the emerging non-invasive microstructure imaging technique, MR cytometry, in differentiating benign and malignant prostate lesions, after removing perfusion effect and introducing histogram features. Methods This study included 33 patients with prostate cancer and 39 patients with benign lesions, who underwent time-dependent diffusion MRI examinations, including pulsed gradient spin-echo (PGSE) diffusion-weighted imaging (DWI) and oscillating gradient spin-echo (OGSE) DWI at frequencies of 20 Hz and 40 Hz. A mono-exponential signal-fitting method was used to eliminate perfusion effects. Apparent diffusion coefficients (ADC), MR-cytometry-derived microstructural parameters (cell diameter \(\:\text{d}\), intracellular volume fraction \(\:{\text{v}}_{\text{i}\text{n}}\), extracellular diffusivity \(\:{D}_{ex}\) and cellularity \(\:{\uprho\:}\)) and corresponding histogram features were calculated to differentiate lesions. Mann-Whitney U-tests were used to evaluate the differences between benign and malignant lesions. Uni- and multi-variate logistic regressions were performed and calculated the area under the receiver operating characteristic (AUC) to quantify the diagnostic performance of classifiers. Results Removal of perfusion effect improved the significance level of the difference in the cell diameter \(\:d\) (from \(\:\text{p}>0.05\) to \(\:\text{p}<0.01\)), where \(\:\text{d}\) is larger in the benign group. The diagnostic efficacy of a single histogram feature was comparable to that of the corresponding ADC or microstructural parameter mean value. Both removing perfusion effect and incorporating histogram features can improve the AUC of the combined regression models (from 0.885 to 0.918 and 0.908, respectively). Combining both strategies further enhanced the AUC to 0.934. Conclusion In MR cytometry analysis, removing the perfusion effect and incorporating histogram features of ADCs and microstructural parameters can provide better diagnostic efficacy.- 11 March 2025
- Published by Wiley in Journal of Magnetic Resonance Imaging
- Vol. 62 (2) , 521-533
Evaluating the Diagnostic Performance of MR Cytometry Imaging in Differentiating Benign and Malignant Breast Tumors
Background: MR cytometry is a class of diffusion‐MRI‐based methods that characterize tumor microstructures at the cellular level. It involves multicompartmental biophysical modeling of multi‐b and multiple diffusion time data to generate microstructural parameters, which may improve differentiation of benign and malignant breast tumors.Purpose: To implement MR cytometry imaging with transcytolemmal water exchange (JOINT and EXCHANGE) to differentiate benign and malignant breast tumors, and to compare the classification efficacy of IMPULSED, JOINT, and EXCHANGE.Study Type: Prospective.Subjects: 115 patients with pathologically confirmed breast tumors (25 benign and 90 malignant).Field Strength/Sequence: 3T; pulsed gradient spin‐echo (PGSE) diffusion‐weighted imaging (DWI) and oscillating gradient spin‐echo (OGSE) DWI at 25 and 50 Hz.Assessment: Tumor regions were delineated by two radiologists on DWI. Time‐dependent ADC and microstructural parameters (cell diameter , intracellular volume fraction , water exchange rate constant , extracellular diffusivity and intracellular intrinsic diffusivity ) were calculated. Classification performance was assessed in the original cohort and in an age‐adjusted cohort (excluding older malignant patients to eliminate significant age differences).Statistical Tests: Mann–Whitney U‐tests compared benign and malignant tumor values. Multivariable logistic regression used a stepwise approach based on the likelihood ratio test. The area under the receiver operating characteristic (AUC) was computed and compared by using the DeLong test.Results: In the full analysis (25 benign, 90 malignant), microstructural parameters from methods incorporating transcytolemmal water exchange (JOINT and EXCHANGE) demonstrated superior performance (AUC: ADC, 0.822; IMPULSED, 0.840; JOINT, 0.902; EXCHANGE, 0.905). Combining different metrics further improved classification (AUC: IMPULSED [, ], 0.942; JOINT [, ], 0.956; EXCHANGE [, ], 0.954; [], 0.927). These improvements were also observed in the age‐adjusted analysis (25 benign, 42 malignant).Data Conclusion: MR cytometry outperformed ADC in distinguishing benign and malignant breast tumors. Incorporating transcytolemmal water exchange into biophysical modeling further improved its diagnostic performance. Evidence Level: 1 Technical Efficacy: Stage 2
- 11 March 2025
- Published by Wiley in Journal of Magnetic Resonance Imaging
- Vol. 62 (2) , 521-533
Evaluating the Diagnostic Performance of MR Cytometry Imaging in Differentiating Benign and Malignant Breast Tumors
Background: MR cytometry is a class of diffusion‐MRI‐based methods that characterize tumor microstructures at the cellular level. It involves multicompartmental biophysical modeling of multi‐b and multiple diffusion time data to generate microstructural parameters, which may improve differentiation of benign and malignant breast tumors.Purpose: To implement MR cytometry imaging with transcytolemmal water exchange (JOINT and EXCHANGE) to differentiate benign and malignant breast tumors, and to compare the classification efficacy of IMPULSED, JOINT, and EXCHANGE.Study Type: Prospective.Subjects: 115 patients with pathologically confirmed breast tumors (25 benign and 90 malignant).Field Strength/Sequence: 3T; pulsed gradient spin‐echo (PGSE) diffusion‐weighted imaging (DWI) and oscillating gradient spin‐echo (OGSE) DWI at 25 and 50 Hz.Assessment: Tumor regions were delineated by two radiologists on DWI. Time‐dependent ADC and microstructural parameters (cell diameter , intracellular volume fraction , water exchange rate constant , extracellular diffusivity and intracellular intrinsic diffusivity ) were calculated. Classification performance was assessed in the original cohort and in an age‐adjusted cohort (excluding older malignant patients to eliminate significant age differences).Statistical Tests: Mann–Whitney U‐tests compared benign and malignant tumor values. Multivariable logistic regression used a stepwise approach based on the likelihood ratio test. The area under the receiver operating characteristic (AUC) was computed and compared by using the DeLong test.Results: In the full analysis (25 benign, 90 malignant), microstructural parameters from methods incorporating transcytolemmal water exchange (JOINT and EXCHANGE) demonstrated superior performance (AUC: ADC, 0.822; IMPULSED, 0.840; JOINT, 0.902; EXCHANGE, 0.905). Combining different metrics further improved classification (AUC: IMPULSED [, ], 0.942; JOINT [, ], 0.956; EXCHANGE [, ], 0.954; [], 0.927). These improvements were also observed in the age‐adjusted analysis (25 benign, 42 malignant).Data Conclusion: MR cytometry outperformed ADC in distinguishing benign and malignant breast tumors. Incorporating transcytolemmal water exchange into biophysical modeling further improved its diagnostic performance. Evidence Level: 1 Technical Efficacy: Stage 2