Zixiong Gao

Zixiong Gao

Auto-generated profile, based on publicly available information

Zixiong Gao

Sun Yat-Sen University, Guangzhou, China

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
1 June 2022 in Radiology
Verified
Journal Article
Closed access
YCYufan Chen
Yufan Chen
ZGZixiong Gao
Zixiong Gao
YHYanni He
Yanni He
WMWuping Mai
Wuping Mai
JLJinhua Li
Jinhua Li
HLHongmei Liu
Hongmei Liu
An artificial intelligence model based on the American College of Radiology Thyroid Imaging Reporting and Data System was developed to differentiate between malignant and benign nodules as well as ... Background US-based diagnosis of thyroid nodules is subjective and influenced by radiologists’ experience levels. Purpose To develop an artificial intelligence model based on American College of Ra...

6 July 2026 in JHEP Reports
Verified
Journal Article
Open access
DYDaopeng Yang
Daopeng Yang
KLKe Lin
Ke Lin
YYYue Yang
Yue Yang
NZNan Zhang
Nan Zhang
BQBin Qiao
Bin Qiao
BZBowen Zhuang
Bowen Zhuang
After thermal ablation, patients with hepatocellular carcinoma (HCC) should undergo long-term surveillance for disease relapse. However, the optimal follow-up strategy remains unexplored. We retrospectively analyzed the clinical data of patients who received complete ablation for initial HCC within the Milan criteria from two tertiary hospitals in China. The Ranger model was applied to develop low- or high-risk stratification for recurrence according to the cutoff value of recurrence probability. The follow-up strategies for each stratification were developed based on monthly probability of recurrence and validated by internal bootstraps and external validation. In total, 1008 patients were included. Using a cutoff of 0.63 for the Ranger model-derived 2-year recurrence probability (based on GGT, AFP, age, tumor number, and tumor size), patients were classified as low risk (≤0.63) or high risk (>0.63). Based on the Ranger model, an optimal surveillance schedule was established to maximize the power of detecting recurrence and reduce the number of total visits. Risk-based strategies were validated by internal bootstrap resampling and an external cohort. The proposed strategy entailed 10 visits for low-risk patients (2 per year for years 1-5) and 12 visits for high-risk patients (5,4,1,1,1 in years 1-5), whereas the guideline-recommended fixed schedule called for 14 visits. The risk-based surveillance strategies reduced delayed detection time (DDT) compared to clinical guidelines. Detailed follow-up arrangements were available through an interactive website(http://fahultrasound.site:8622/pafah). Compared with the fixed guideline schedule, our risk-stratified surveillance strategy reduced patient burden through fewer follow-up visits and enhanced early detection of recurrence in both low-risk and high-risk patients. This strategy provides a valuable reference for clinicians and patients to standardize the follow-up strategy following local ablation. The Ranger model allows estimation of the prognosis of patients with early-stage HCC undergoing radical ablation and builds a risk-based follow-up strategy for individualized postoperative management. The risk-based follow-up strategy was more efficient in early detection of recurrence without increasing medical costs compared with guidelines. Our interactive website provides a valuable reference for both clinicians and patients to standardize the follow-up of HCC patients after local ablation.