Lihui WANG

Dr. Lihui WANG


Research Focus & Metrics

Academic Career Length

37Years (1990-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 February 2021 in IEEE Transactions on Medical Imaging
Verified
Journal Article
Closed access
YQYulei Qin
Yulei Qin
HZHao Zheng
Hao Zheng
YGYun Gu
Yun Gu
XHXiaolin Huang
Xiaolin Huang
Jie YangJie Yang
Jie Yang
GYGuang-Zhong Yang
Guang-Zhong Yang
Training convolutional neural networks (CNNs) for segmentation of pulmonary airway, artery, and vein is challenging due to sparse supervisory signals caused by the severe class imbalance between tubular targets and background. We present a CNNs-based method for accurate airway and artery-vein segmentation in non-contrast computed tomography. It enjoys superior sensitivity to tenuous peripheral bronchioles, arterioles, and venules. The method first uses a feature recalibration module to make the best use of features learned from the neural networks. Spatial information of features is properly integrated to retain relative priority of activated regions, which benefits the subsequent channel-wise recalibration. Then, attention distillation module is introduced to reinforce representation learning of tubular objects. Fine-grained details in high-resolution attention maps are passing down from one layer to its previous layer recursively to enrich context. Anatomy prior of lung context map and distance transform map is designed and incorporated for better artery-vein differentiation capacity. Extensive experiments demonstrated considerable performance gains brought by these components. Compared with state-of-the-art methods, our method extracted much more branches while maintaining competitive overall segmentation performance. Codes and models are available at http://www.pami.sjtu.edu.cn/News/56.

1 August 2019 in EBioMedicine
Verified
Journal Article
Open access
CSCaixia Sun
Caixia Sun
XTXin Tian
Xin Tian
ZLZhenyu Liu
Zhenyu Liu
WLWeili Li
Weili Li
PLPengfei Li
Pengfei Li
JTJie Tian
Jie Tian
We aimed to investigate whether pre-therapeutic radiomic features based on magnetic resonance imaging (MRI) can predict the clinical response to neoadjuvant chemotherapy (NACT) in patients with locally advanced cervical cancer (LACC). A total of 275 patients with LACC receiving NACT were enrolled in this study from eight hospitals, and allocated to training and testing sets (2:1 ratio). Three radiomic feature sets were extracted from the intratumoural region of T1-weighted images, intratumoural region of T2-weighted images, and peritumoural region of T2-weighted images before NACT for each patient. With a feature selection strategy, three single sequence radiomic models were constructed, and three additional combined models were constructed by combining the features of different regions or sequences. The performance of all models was assessed using receiver operating characteristic curve. The combined model of the intratumoural zone of T1-weighted images, intratumoural zone of T2-weighted images,and peritumoural zone of T2-weighted images achieved an AUC of 0.998 in training set and 0.999 in testing set, which was significantly better (p < .05) than the other radiomic models. Moreover, no significant variation in performance was found if different training sets were used. This study demonstrated that MRI-based radiomic features hold potential in the pretreatment prediction of response to NACT in LACC, which could be used to identify rightful patients for receiving NACT avoiding unnecessary treatment.

24 February 2022 in Frontiers in Oncology
Verified
Journal Article
Open access
SGShunchao Guo
Shunchao Guo
LWLihui WANG
Lihui WANG
QCQijian Chen
Qijian Chen
LWLi Wang
Li Wang
JZJian Zhang
Jian Zhang
YZYuemin Zhu
Yuemin Zhu
Purpose: Glioma is the most common primary brain tumor, with varying degrees of aggressiveness and prognosis. Accurate glioma classification is very important for treatment planning and prognosis prediction. The main purpose of this study is to design a novel effective algorithm for further improving the performance of glioma subtype classification using multimodal MRI images. Method: MRI images of four modalities for 221 glioma patients were collected from Computational Precision Medicine: Radiology-Pathology 2020 challenge, including T1, T2, T1ce, and fluid-attenuated inversion recovery (FLAIR) MRI images, to classify astrocytoma, oligodendroglioma, and glioblastoma. We proposed a multimodal MRI image decision fusion-based network for improving the glioma classification accuracy. First, the MRI images of each modality were input into a pre-trained tumor segmentation model to delineate the regions of tumor lesions. Then, the whole tumor regions were centrally clipped from original MRI images followed by max–min normalization. Subsequently, a deep learning-based network was designed based on a unified DenseNet structure, which extracts features through a series of dense blocks. After that, two fully connected layers were used to map the features into three glioma subtypes. During the training stage, we used the images of each modality after tumor segmentation to train the network to obtain its best accuracy on our testing set. During the inferring stage, a linear weighted module based on a decision fusion strategy was applied to assemble the predicted probabilities of the pre-trained models obtained in the training stage. Finally, the performance of our method was evaluated in terms of accuracy, area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), etc. Results: The proposed method achieved an accuracy of 0.878, an AUC of 0.902, a sensitivity of 0.772, a specificity of 0.930, a PPV of 0.862, an NPV of 0.949, and a Cohen’s Kappa of 0.773, which showed a significantly higher performance than existing state-of-the-art methods. Conclusion: Compared with current studies, this study demonstrated the effectiveness and superiority in the overall performance of our proposed multimodal MRI image decision fusion-based network method for glioma subtype classification, which would be of enormous potential value in clinical practice.

4 February 2020 in Frontiers in Oncology
Verified
Journal Article
Open access
XTXin Tian
Xin Tian
CSCaixia Sun
Caixia Sun
ZLZhenyu Liu
Zhenyu Liu
WLWeili Li
Weili Li
HDHui Duan
Hui Duan
CCChunlin Chen
Chunlin Chen
Objective: To investigate whether pretreatment CT-derived radiomic features could be applied for prediction of clinical response to neoadjuvant chemotherapy (NACT) in locally advanced cervical cancer (LACC). Patients and Methods: 277 LACC patients treated with NACT followed by surgery/radiotherapy were included in this multi-institution retrospective study. 1294 radiomic features were extracted from venous contrast enhanced and non-enhanced CT imaging for each patient. Five combined methods of feature selection were used to reduce dimension of features. Radiomics signature was constructed by Random Forest (RF) method in a primary cohort of 221 patients. A combined model incorporating radiomics signature with clinical factors was developed using multivariable logistic regression. Prediction performance was then tested in a validation cohort of 56 patients. Results: Radiomics signature containing pre and post contrast imaging features can adequately distinguish chemotherapeutic responders from non-responders in both primary and validation cohorts [AUCs: 0.773 (95% CI, 0.701-0.845) and 0.816 (95% CI, 0.690-0.942), respectively]. The combined model has a better predictive performance with an AUC of 0.803 (95% CI, 0.734-0.872) in the primary set and an AUC of 0.821 (95% CI, 0.697 to 0.946) in the validation set, compared to radiomics signature alone. Both models showed good discrimination, calibration. Conclusion: Newly developed radiomic model provided an easy-to-use predictor of chemotherapeutic response with improved predictive ability, which might facilitate optimal treatment strategies tailored for individual LACC patients

1 June 2024 in Biomedical Signal Processing and Control
Verified
Journal Article
Closed access
XGXianming Gu
Xianming Gu
LWLihui WANG
Lihui WANG
ZDZeyu Deng
Zeyu Deng
YCYing Cao
Ying Cao
XHXingyu Huang
Xingyu Huang
YZYue-Min Zhu
Yue-Min Zhu

13 April 2020 in IEEE Access
Verified
Journal Article
Open access
KTKun Tang
Kun Tang
ZLZhi Li
Zhi Li
LTLili Tian
Lili Tian
LWLihui WANG
Lihui WANG
YZYuemin Zhu
Yuemin Zhu
We proposed an unsupervised end-to-end Affine and Deformable Medical Image Registration (ADMIR) method based on convolutional neural network (ConvNet). ADMIR includes three key components: an affine registration module for learning the affine transformation parameters, a deformable registration module for learning the displacement vector field, and a spatial transformer for getting the final warped image from both affine and deformable transformation parameters. To evaluate its performance, the magnetic resonance images of drug-addicted brains were used to train and test the model, and we compared it with two state-of-art methods in terms of Dice score, Hausdorff distance, and average symmetric surface distance. The experimental results demonstrated that our proposed ADMIR model outperforms existing methods even with the images without pre-alignment, which suggests that the ADMIR model can be used to achieve quick medical image registration with high accuracy.