Xia Liu

Xia Liu

Auto-generated profile, based on publicly available information

Xia Liu

Peking Union Medical College Hospital

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
1 January 2020 in Physica Medica
Verified
Journal Article
Open access
ZLzhikai liu
zhikai liu
XLXia Liu
Xia Liu
BXBin Xiao
Bin Xiao
SWShaobin Wang
Shaobin Wang
ZMZheng Miao
Zheng Miao
FZFuquan Zhang
Fuquan Zhang
We introduced and evaluated an end-to-end organs-at-risk (OARs) segmentation model that can provide accurate and consistent OARs segmentation results in much less time. We collected 105 patients' Computed Tomography (CT) scans that diagnosed locally advanced cervical cancer and treated with radiotherapy in one hospital. Seven organs, including the bladder, bone marrow, left femoral head, right femoral head, rectum, small intestine and spinal cord were defined as OARs. The annotated contours of the OARs previously delineated manually by the patient's radiotherapy oncologist and confirmed by the professional committee consisted of eight experienced oncologists before the radiotherapy were used as the ground truth masks. A multi-class segmentation model based on U-Net was designed to fulfil the OARs segmentation task. The Dice Similarity Coefficient (DSC) and 95th Hausdorff Distance (HD) are used as quantitative evaluation metrics to evaluate the proposed method. The mean DSC values of the proposed method are 0.924, 0.854, 0.906, 0.900, 0.791, 0.833 and 0.827 for the bladder, bone marrow, femoral head left, femoral head right, rectum, small intestine, and spinal cord, respectively. The mean HD values are 5.098, 1.993, 1.390, 1.435, 5.949, 5.281 and 3.269 for the above OARs respectively. Our proposed method can help reduce the inter-observer and intra-observer variability of manual OARs delineation and lessen oncologists' efforts. The experimental results demonstrate that our model outperforms the benchmark U-Net model and the oncologists' evaluations show that the segmentation results are highly acceptable to be used in radiation therapy planning.

1 June 2025 in Technology in Cancer Research & Treatment
Verified
Journal Article
Open access
JWJiaxin Wang
Jiaxin Wang
BYBo Yang
Bo Yang
TPTingtian Pang
Tingtian Pang
ZWZhiqun Wang
Zhiqun Wang
YZYue Zhang
Yue Zhang
JQJie Qiu
Jie Qiu
Purpose: In our institute, stereotactic radiosurgery (SRS) and fractionated stereotactic radiotherapy (SRT) were performed by Cyberknife-S7 (CK-S7) which provided the selection of different collimators. This study aimed to compare critical plan qualities including conformality, high-dose area, dose fallout, and treatment efficiency between InCise™ multileaf collimator (MLC) based and Iris™ variable aperture collimators (Iris) based treatment plans. Methods: Twenty-five patients with intracranial tumors treated with CK-S7 were retrospectively analyzed. The Precision 3.3.0.0 with VOLO™ optimizer and GPU-accelerated Monte Carlo dose calculation algorithm was used for treatment planning. The new conformality index (nCI), homogeneity index (HI), high-dose ratio (HDR) and distribution inside plan target volume (PTV), dose gradient distance (DGD) outside PTV, organs at risk (OARs) sparing, and treatment efficiency were compared between MLC based and Iris based plans. Results: MLC plan achieved higher nCI, higher HDRs from 135% prescription dose (PD) to 110%PD and trended to form more centralized and gathered high-dose distribution inside PTV, while no statistical difference was found in HI. Iris possessed better dose-engraving ability around the target boundary especially when it close to OARs with strict dose constraints. MLC plan showed shorter DGD from 90%PD to 20%PD. MLC plan achieved less MUs (−67.14%), less nodes (−41.5%), less beams (−74.06%) and shorter treatment time (−51.64%). There were positive correlations between the effective radius of PTV (rPTV) and DGD from 90%PD to 10%PD both in MLC and Iris plans. Conclusions: MLC plan achieved comparable conformality, higher HDRs, more gathered high-dose distribution, faster dose fallout and more efficient treatment which proved it an excellent SRS/SRT choice for intracranial tumors treated with CK-S7. MLC might take an important advantage for the uncompliant and painful patients. However, Iris showed a better dose-engraving ability, it might be taken into consideration especially when the tumor was close to OARs with strict dose constraints.