Ying Liang

Ying Liang


Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
10 November 2022 in Frontiers in Oncology
Verified
Journal Article
Open access
XWXieraili Wumener
Xieraili Wumener
YZYarong Zhang
Yarong Zhang
ZWZhenguo Wang
Zhenguo Wang
MZMaoqun Zhang
Maoqun Zhang
ZZZihan Zang
Zihan Zang
TSTao Sun
Tao Sun
Objectives: 18F-fluorodeoxyglucose (FDG) PET/CT has been widely used in tumor diagnosis, staging, and response evaluation. To determine an optimal therapeutic strategy for lung cancer patients, accurate staging is essential. Semi-quantitative standardized uptake value (SUV) is known to be affected by multiple factors and may fail to differentiate between benign and malignant lesions. Lymph nodes (LNs) in the mediastinal and pulmonary hilar regions with high FDG uptake due to granulomatous lesions such as tuberculosis, which has a high prevalence in China, pose a diagnostic challenge. This study aims to evaluate the diagnostic value of the quantitative metabolic parameters derived from dynamic 18F-FDG PET/CT in differentiating metastatic and non-metastatic LNs in lung cancer.Methods: One hundred and eight patients with pulmonary nodules were enrolled to perform 18F-FDG PET/CT dynamic + static imaging with informed consent. One hundred and thirty-five LNs in 29 lung cancer patients were confirmed by pathology. Static image analysis parameters including LN-SUVmax, LN-SUVmax/primary tumor SUVmax (LN-SUVmax/PT-SUVmax), mediastinal blood pool SUVmax (MBP-SUVmax), LN-SUVmax/MBP-SUVmax, and LN-SUVmax/short diameter. Quantitative parameters including K1, k2, k3 and Ki and of each LN were obtained by applying the irreversible two-tissue compartment model using in-house Matlab software. Ki/K1 was computed subsequently as a separate marker. We further divided the LNs into mediastinal LNs (N=82) and pulmonary hilar LNs (N=53). Wilcoxon rank-sum test or Independent-samples T-test and receiver-operating characteristic (ROC) analysis was performed on each parameter to compare the diagnostic efficacy in differentiating lymph node metastases from inflammatory uptake. P<0.05 were considered statistically significant.Results: Among the 135 FDG-avid LNs confirmed by pathology, 49 LNs were non-metastatic, and 86 LNs were metastatic. LN-SUVmax, MBP-SUVmax, LN-SUVmax/MBP-SUVmax, and LN-SUVmax/short diameter couldn’t well differentiate metastatic from non-metastatic LNs (P>0.05). However, LN-SUVmax/PT-SUVmax have good performance in the differential diagnosis of non-metastatic and metastatic LNs (P=0.039). Dynamic metabolic parameters in addition to k3, the parameters including K1, k2, Ki, and Ki/K1, on the other hand, have good performance in the differential diagnosis of metastatic and non-metastatic LNs (P=0.045, P=0.001, P=0.001, P=0.001, respectively). For ROC analysis, the metabolic parameters Ki (AUC of 0.672 [0.579-0.765], sensitivity 0.395, specificity 0.918) and Ki/K1 (AUC of 0.673 [0.580-0.767], sensitivity 0.570, specificity 0.776) have good performance in the differential diagnosis of metastatic from non-metastatic LNs than SUVmax (AUC of 0.596 [0.498-0.696], sensitivity 0.826, specificity 0.388), included the mediastinal region and pulmonary hilar region.Conclusion: Compared with SUVmax, quantitative parameters such as K1, k2, Ki and Ki/K1 showed promising results for differentiation of metastatic and non-metastatic LNs with high uptake. The Ki and Ki/K1 had a high differential diagnostic value both in the mediastinal region and pulmonary hilar region.

20 October 2023 in ACS Pharmacology & Translational Science
Verified
Journal Article
Closed access
ZYZiyue Yu
Ziyue Yu
YHYong Huang
Yong Huang
HCHualong Chen
Hualong Chen
ZJZeng Jiang
Zeng Jiang
CLChengze Li
Chengze Li
ZWZehui Wu
Zehui Wu

24 October 2023 in EJNMMI Physics
Verified
Journal Article
Open access
PDF file
GLGanglin Liang
Ganglin Liang
JZJinpeng Zhou
Jinpeng Zhou
ZCZixiang Chen
Zixiang Chen
LWLiwen Wan
Liwen Wan
XWXieraili Wumener
Xieraili Wumener
ZHZhanli Hu
Zhanli Hu
Background: Dynamic positron emission tomography (PET) images are useful in clinical practice because they can be used to calculate the metabolic parameters (Ki) of tissues using graphical methods (such as Patlak plots). Ki is more stable than the standard uptake value and has a good reference value for clinical diagnosis. However, the long scanning time required for obtaining dynamic PET images, usually an hour, makes this method less useful in some ways. There is a tradeoff between the scan durations and the signal-to-noise ratios (SNRs) of Ki images. The purpose of our study is to obtain approximately the same image as that produced by scanning for one hour in just half an hour, improving the SNRs of images obtained by scanning for 30 min and reducing the necessary 1-h scanning time for acquiring dynamic PET images. Methods: In this paper, we use U-Net as a feature extractor to obtain feature vectors with a priori knowledge about the image structure of interest and then utilize a parameter generator to obtain five parameters for a two-tissue, three-compartment model and generate a time activity curve (TAC), which will become close to the original 1-h TAC through training. The above-generated dynamic PET image finally obtains the Ki parameter image. Results: A quantitative analysis showed that the network-generated Ki parameter maps improved the structural similarity index measure and peak SNR by averages of 2.27% and 7.04%, respectively, and decreased the root mean square error (RMSE) by 16.3% compared to those generated with a scan time of 30 min. Conclusions: The proposed method is feasible, and satisfactory PET quantification accuracy can be achieved using the proposed deep learning method. Further clinical validation is needed before implementing this approach in routine clinical applications.

23 November 2022 in Bioinformatics
Verified
Journal Article
Open access
PDF file
KHKaibin Huang
Kaibin Huang
SHShengyun Huang
Shengyun Huang
GCGuojing Chen
Guojing Chen
XLXue Li
Xue Li
SLShawn Li
Shawn Li
YGYi Gao
Yi Gao
Limited by spatial resolution and visual contrast, bone scintigraphy interpretation is susceptible to subjective factors, which considerably affects the accuracy and repeatability of lesion detection and anatomical localization. In this work, we design and implement an end-to-end multi-task deep learning model to perform automatic lesion detection and anatomical localization in whole-body bone scintigraphy. A total of 617 whole-body bone scintigraphy cases including anterior and posterior views were retrospectively analyzed. The proposed semi-supervised model consists of two task flows. The first one, the lesion segmentation flow, received image patches and was trained in a supervised way. The other one, skeleton segmentation flow, was trained on as few as five labeled images in conjunction with the multi-atlas approach, in a semi-supervised way. The two flows joint in their encoder layers so each flow can capture more generalized distribution of the sample space and extract more abstract deep features. The experimental results show that the architecture achieved the highest precision in the finest bone segmentation task in both anterior and posterior images of whole-body scintigraphy. Such an end-to-end approach with very few manual annotation requirement would be suitable for algorithm deployment. Moreover, the proposed approach reliably balances unsupervised labels construction and supervised learning, providing useful insight for weakly labeled image analysis. Supplementary data are available at Bioinformatics online.

19 October 2023 in Pharmaceuticals
Verified
Journal Article
Open access
PDF file
HYHuang Yong
Huang Yong
CLChengze Li
Chengze Li
ZLZhongjing Li
Zhongjing Li
QWQiong Wang
Qiong Wang
SHSize Huang
Size Huang
YLYing Liang
Ying Liang
Immunotherapy targeting the programmed death-ligand 1 (PD-L1)/programmed cell death protein 1 (PD-1) pathway has shown remarkable efficacy against various cancers, but the overall response rate (ORR) is still low. PD-L1 expression in tumors may predict treatment response to immunotherapy. Indeed, ongoing clinical studies utilize a few PD-L1 radiotracers to assess PD-L1 expression as a predictive biomarker for immunotherapy. Here, we present a novel positron emission tomography (PET) radiotracer called [68Ga]BMSH, which is derived from a small molecule inhibitor specifically targeting the binding site of PD-L1. The inhibitor was modified to optimize its in vivo pharmacokinetic properties and enable chelation of 68Ga. In vitro evaluation revealed [68Ga]BMSH possessed a strong binding affinity, high specificity, and rapid internalization in PD-L1 overexpressing cells. Biodistribution studies showed that PD-L1 overexpressing tumors had an uptake of [68Ga]BMSH at 4.22 ± 0.65%ID/g in mice, while the number was 2.23 ± 0.41%ID/g in PD-L1 low-expressing tumors. Micro-PET/CT imaging of tumor-bearing mice further confirmed that, compared to [18F]FDG, [68Ga]BMSH can specifically identify tumors with varying levels of PD-L1 expression. Our findings suggest that the [68Ga]BMSH is a PD-L1 radioligand with ideal imaging properties, and its further application in the clinical screening of PD-L1 overexpressing tumors may improve ORR for immunotherapy.

15 March 2015 in World Journal of Surgical Oncology
Verified
Journal Article
Open access
PDF file
LZLi-Na Zhou
Li-Na Zhou
NWNing Wu
Ning Wu
YLYing Liang
Ying Liang
KGKai Gao
Kai Gao
XLXiao-Ying Li
Xiao-Ying Li
LZLianfeng Zhang
Lianfeng Zhang
The purpose of this study is to investigate whether 18 F-fluorodeoxyglucose (FDG) micro-positron emission tomography-computed tomography (microPET-CT) can be used to monitor a metabolic response to gefitinib in nude mouse tumor xenografts. Sixteen nude mice were implanted with human A431 epidermoid carcinoma cells and ten with human A549 lung adenocarcinoma cells, and the tumors were allowed to grow to an approximate size of 150 mm3. Ten and five of these mice, respectively, received intragastric gefitinib (100 mg/kg) once daily for 14 days, whereas six and five, respectively, received sterile water. Tumor metabolic activity was assessed by 18 F-FDG microPET imaging before treatment (day 0) and on days 2, 7, and 14. Tumor uptake of 18 F-FDG was determined from a region-of-interest drawn around the tumor, and the maximum percentage injected dose per gram (%ID/gmax) was calculated. Tumor volume measured on day 14 by microCT was used to categorize tumors as sensitive, stable, or resistant to gefitinib, and pathologic changes in these tumors were analyzed. On day 2, the average changes in 18 F-FDG uptake by A431 tumors sensitive, stable, and resistant to gefitinib were −30.92% ± 6.66%, −5.68% ± 6.95%, and 7.72% ± 3.85%, respectively (P < 0.05 each), with no significant differences in the sizes of tumors sensitive and stable to gefitinib (P = 0.169). On day 7, sensitive tumors were significantly smaller than stable tumors (P = 0.034). On day 14, areas of necrosis were observed in gefitinib-sensitive tumors, with tumor necrosis ratios differing significantly among the sensitive, stable, and control groups (P < 0.05 each). In mice implanted with A549 cells, however, tumor 18 F-FDG uptake, volume, and percent necrosis did not differ significantly between gefitinib-treated and untreated mice on days 0, 2, 7, and 14 (P > 0.05 each). F-FDG uptake is a sensitive method of detecting metabolic changes in tumors associated with therapy in vivo.