Junhao Zha

Junhao Zha

Auto-generated profile, based on publicly available information

Junhao Zha

Cultivation and Construction Site of the State Key Laboratory of Intelligent Imaging and Interventional Medicine, Department of Radiology, Zhongda Hospital, Medical School of Southeast University, Nanjing, Jiangsu, China (Y.Y., T.X., K.Q., L.G., B.Z., J. Zha, J. Zhou, Y.C., C.L., T.T., Z.X., S.J.).

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 May 2023 in Radiology
Verified
Journal Article
Closed access
TXTianyi Xia
Tianyi Xia
ZZZheng-Hao Zhou
Zheng-Hao Zhou
XMXiang-Pan Meng
Xiang-Pan Meng
JZJun-Hao Zha
Jun-Hao Zha
QYQian Yu
Qian Yu
SJShenghong Ju
Shenghong Ju
Background Prediction of microvascular invasion (MVI) may help determine treatment strategies for hepatocellular carcinoma (HCC). Purpose To develop a radiomics approach for predicting MVI status based on preoperative multiphase CT images and to identify MVI-associated differentially expressed genes. Materials and Methods Patients with pathologically proven HCC from May 2012 to September 2020 were retrospectively included from four medical centers. Radiomics features were extracted from tumors and peritumor regions on preoperative registration or subtraction CT images. In the training set, these features were used to build five radiomics models via logistic regression after feature reduction. The models were tested using internal and external test sets against a pathologic reference standard to calculate area under the receiver operating characteristic curve (AUC). The optimal AUC radiomics model and clinical-radiologic characteristics were combined to build the hybrid model. The log-rank test was used in the outcome cohort (Kunming center) to analyze early recurrence-free survival and overall survival based on high versus low model-derived score. RNA sequencing data from The Cancer Image Archive were used for gene expression analysis. Results A total of 773 patients (median age, 59 years; IQR, 49-64 years; 633 men) were divided into the training set (n = 334), internal test set (n = 142), external test set (n = 141), outcome cohort (n = 121), and RNA sequencing analysis set (n = 35). The AUCs from the radiomics and hybrid models, respectively, were 0.76 and 0.86 for the internal test set and 0.72 and 0.84 for the external test set. Early recurrence-free survival (P < .01) and overall survival (P < .007) can be categorized using the hybrid model. Differentially expressed genes in patients with findings positive for MVI were involved in glucose metabolism. Conclusion The hybrid model showed the best performance in prediction of MVI. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Summers in this issue.

1 October 2023 in Radiology
Verified
Journal Article
Closed access
TXTianyi Xia
Tianyi Xia
MDMulong Du
Mulong Du
HLHuiqin Li
Huiqin Li
YWYuan‑Cheng Wang
Yuan‑Cheng Wang
JZJunhao Zha
Junhao Zha
SJShenghong Ju
Shenghong Ju
Background A better understanding of the association between liver MRI proton density fat fraction (PDFF) and liver diseases might support the clinical implementation of MRI PDFF. Purpose To quantify the genetically predicted causal effect of liver MRI PDFF on liver disease risk. Materials and Methods This population-based prospective observational study used summary-level data mainly from the UK Biobank and FinnGen. Mendelian randomization analysis was conducted using the inverse variance-weighted method to explore the causal association between genetically predicted liver MRI PDFF and liver disease risk with Bonferroni correction. The individual-level data were downloaded between August and December 2020 from the UK Biobank. Logistic regression analysis was performed to validate the association between liver MRI PDFF polygenic risk score and liver disease risk. Mediation analyses were performed using multivariable mendelian randomization. Results Summary-level and individual-level data were obtained from 32 858 participants and 378 436 participants (mean age, 57 years ± 8 [SD]; 203 108 female participants), respectively. Genetically predicted high liver MRI PDFF was associated with increased risks of malignant liver neoplasm (odds ratio [OR], 4.5; P < .001), alcoholic liver disease (OR, 1.9; P < .001), fibrosis and cirrhosis of the liver (OR, 3.0; P < .004), fibrosis of the liver (OR, 3.6; P = .002), cirrhosis of the liver (OR, 3.8; P < .001), nonalcoholic steatohepatitis (OR, 7.7; P < .001), and nonalcoholic fatty liver disease (NAFLD) (OR, 4.4; P < .001). Individual-level evidence supported these associations after grouping participants based on liver MRI PDFF polygenic risk score (all P < .004). The mediation analysis indicated that genetically predicted high-density lipoprotein cholesterol, type 2 diabetes mellitus, and waist-to-hip ratio (mediation effects, 25.1%-46.3%) were related to the occurrence of fibrosis and cirrhosis of the liver, cirrhosis of the liver, and NAFLD at liver MRI PDFF (all P < .05). Conclusion This study provided evidence of the association between genetically predicted liver MRI PDFF and liver health. © RSNA, 2023 Supplemental material is available for this article. See also the editorials by Reeder and Starekova and Monsell in this issue.

15 April 2020 in Frontiers in Oncology
Verified
Journal Article
Open access
PDF file
HWHao Wang
Hao Wang
JLJian Lu
Jian Lu
XZXiao-Ting Zheng
Xiao-Ting Zheng
JZJun-Hao Zha
Jun-Hao Zha
WJWen-Dong Jing
Wen-Dong Jing
JGJinhe Guo
Jinhe Guo
Purpose: To compare the efficacy and safety of computed tomography (CT)–guided 125I seed implantation with second-line chemotherapy in treatment of oligorecurrence non–small cell lung cancer after failure of first-line chemotherapy. Methods: Data of oligorecurrence non–small cell lung cancer patients after failure of first-line chemotherapy at two institutions were retrospectively reviewed from January 2013 to July 2018. A total of 53 patients who received the treatment of 125I seed implantation or second-line chemotherapy were eligible for this study. In group A, 25 patients, 84 lesions, received CT-guided permanent 125I seed implantation, whereas in group B, 28 patients, 96 lesions, received second-line chemotherapy. The outcomes were measured in terms of disease control rate, overall survival, quality of life, and complications. Results: The median follow-up period was 13 months (range, 5–42 months). Disease control rate in group A was higher than that in group B (70.8 vs. 42.3%, P = 0.042) at 6 months after treatment. The median overall survival was 12.8 months (95% confidence interval, 10.5–15.1 months) in group A and 15.2 months (95% confidence interval, 12.2–18.2 months) in group B, with no significant difference (P = 0.847). Since the fourth month, the number of patients in group A with a non-decreasing Karnofsky Performance Scale score was more than that in group B (P < 0.05). The incidence of grade 3 or higher complications especially hematologic toxicity in group A was significantly lower than that in group B (P < 0.05). Conclusion: Radioactive 125I seed implantation is safe and feasible in selected non–small cell lung cancer patients with oligorecurrence after failure of first-line chemotherapy and seems to provide a better long-term quality of life in these patients compared with second-line chemotherapy.

1 January 2022 in SSRN Electronic Journal
Verified
Preprint
Open access
PDF file
JZJunhao Zha
Junhao Zha
SHShan Huang
Shan Huang
TXTianyi Xia
Tianyi Xia
ZCZhiyuan Chen
Zhiyuan Chen
TZTianying Zheng
Tianying Zheng
SJShenghong Ju
Shenghong Ju
Background: Large-scale validation of a non-invasive combined radiomics-clinic (CoRC) model for assessing fibrosis and necroinflammation remains lacking. We aimed to develop and validate a CoRC model based on conventional MRI for diagnosing clinically significant liver fibrosis (≥ F2) and necroinflammation (≥ G2). Methods: This retrospective cohort study recruited 537 patients with pathologically proven liver fibrosis and necroinflammation at center 1 from May 2015 to Aug 2020 and center 2 between Jan 2011 and Jan 2021. 394 patients were randomly allocated into training (n = 276) and internal test (n = 118) cohorts at center 1. A temporal test cohort at center 1 (n = 96) and an external test cohort from center 2 (n = 47) were used for validation. Automated entire liver segmentation used ResUNet-based Human-in-the-Loop approach. Radiomics features were extracted from the mask on fat-suppressed T2-weighted and delayed enhanced T1-weighted images separately. Radiomics signatures were generated using logistic regression. Radiomics-scores and optimal clinical biomarkers as independent risk factors were integrated into the CoRC models in the training cohort with multivariate logistic regression. Models were tested in independent cohorts. Model performances were mainly evaluated with the area under the receiver operating characteristic curves (AUCs). Findings: In the internal, temporal, and external test cohorts, CoRC model 1 yielded AUCs of 0.79 (95% CI: 0.70, 0.86), 0.82 (95% CI: 0.73, 0.89), and 0.83 (95% CI: 0.69, 0.92) for ≥ F2, meanwhile, CoRC model 2 showed AUCs of 0.86 (95% CI: 0.79, 0.92), 0.79 (95% CI: 0.68, 0.86), and 0.89 (95% CI: 0.76, 0.96) for ≥ G2. In internal/temporal test transient elastography-based liver stiffness measurement (TE-LSM) subgroups, we compared the performance of CoRC models with TE-LSM (AUC 0.78 vs.0.79, P = 0.86 / 0.82 vs. 0.73, P = 0.14 for ≥ F2, whereas 0.88 vs. 0.81, P = 0.16 / 0.79 vs. 0.74, P = 0.49 for ≥ G2). Interpretation: CoRC models exhibited promising diagnostic performances for clinically significant liver fibrosis and necroinflammation, which could be a potential alternative when TE-LSM is unavailable. Funding: This research was supported by the National Key R&D Program of China (2021YFF0501504); the National Natural Science Foundation of China (NSFC, No. 81830053, 92059202, and 61821002); the Key Research and Development Program of Jiangsu Province (BE2020717) Declaration of Interest: Yang Song is from a commercial company, Siemens Healthineers Ltd., as a MR Scientific Specialist doing technical support in this study under Siemens collaboration regulation without any payment and personal concern regarding to this study. Our study did not use any commercial product from Siemens. Neither this manuscript, in its entirety, nor any part thereof, has been submitted for publication elsewhere. All authors have no conflicts of interest to disclose Ethical Approval: This retrospective multi-center study was approved by the institutional ethics committee of all participating centers, which waived the requirement for written informed consent. All patient data were anonymized before analysis.

13 February 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
TXTianyi Xia
Tianyi Xia
HJHanyu Jiang
Hanyu Jiang
FWFang Wang
Fang Wang
YLYing Lei
Ying Lei
YWYawen Wang
Yawen Wang
SJShenghong Ju
Shenghong Ju
Background: Adjuvant transarterial chemoembolization (TACE) after curative-intent resection for hepatocellular carcinoma (HCC) shows conflicting results. We hypothesized heterogeneity of treatment effect (HTE) and aimed to assess it and develop a tool to identify patients benefiting from adjuvant TACE.Methods: We retrospectively included HCC patients from 10 tertiary centers (real-world cohort, n=2,630) between 2012 and 2023, and an independent phase III randomized controlled trial (validation cohort, n=328). After propensity score matching between patients with adjuvant TACE and with active surveillance, the real-world cohort was defined as the development cohort (n=1,632). HTE on overall survival (OS) was assessed using (1) subgroup (subgroup-based), (2) Cox regression (risk-based), and (3) causal survival forest (effect-based) analyses in the development cohort. Thereafter, a policy tree (HAT) to identify patients benefitting from adjuvant TACE was constructed according to individual treatment effects in the development cohort and validated across cohorts. Survival analyses were performed using Cox regression and log-rank tests.Findings: A total of 2,958 patients were included (median age, 58 years; 83·6% male). After matching, adjuvant TACE showed no significant effect on OS (HR, 0·85 [95% CI, 0·70-1·03]; p=0·106]). HTEs were observed across subgroup-, risk-, and effect-based analyses, particularly in patients with microvascular invasion, younger age, or corona enhancement (all p<0·05). HAT stratified patients into two subgroups: those in the rec-adjuvant TACE subgroup achieved better OS (HR, 0·54 [95% CI, 0·43-0·69]), while the rec-active surveillance subgroup had poorer OS (HR, 1·88 [95% CI, 1·32-2·67]) in the development cohort (all p<0·001). The stratified treatment effect trends were confirmed in both the validation and real-world cohorts (p values, 0·059 to <0·001).Interpretation: Adjuvant TACE showed significant HTE after curative-intent resection of HCC. HAT differentiated patients who benefitted from adjuvant TACE, which may guide individualized postoperative management.

1 February 2026 in Hypertension
Verified
Journal Article
Closed access
YYYaoyao Yu
Yaoyao Yu
TXTianyi Xia
Tianyi Xia
YWYinlu Wang
Yinlu Wang
KQKeli Qin
Keli Qin
LGLingling Gao
Lingling Gao
SJShenghong Ju
Shenghong Ju
BACKGROUND: White matter hyperintensities (WMH), a hallmark imaging feature of small vessel disease, are strongly associated with neurodegenerative and cardiovascular conditions. METHODS: We performed bidirectional and mediation Mendelian randomization analyses using summary statistics from GWAS (genome-wide association studies) of 9 large cohorts of plasma proteins (n=997–35 559), blood pressure (n=1 028 980), and WMH (n=21 381). The inverse-variance-weighted method or Wald ratio was applied as the primary Mendelian randomization approach, with false discovery rate correction and independent replication. We further integrated Mendelian randomization with PheWAS (phenome-wide association studies) to prioritize WMH risk factors and assess mediation via systolic blood pressure and diastolic blood pressure. RESULTS: Eighteen plasma proteins were genetically associated with WMH, 13 of which were replicated. Thirteen antihypertensive target genes were also linked to WMH burden, including ADRB3 , AOC1 (amine oxidase copper containing 1), SHBG (sex hormone binding globulin), and KCNH2 (potassium voltage-gated channel subfamily H member 2), which influenced both systolic blood pressure and diastolic blood pressure. Mendelian randomization-PheWAS highlighted systolic blood pressure and diastolic blood pressure as the top-ranked WMH risk factors among 21 976 traits. Antihypertensive drug targets, including angiotensin II receptor blockers, β-blockers, calcium channel blockers, and diuretics, were significantly associated with WMH burden. Mediation analysis showed that systolic blood pressure partially mediated TFPI’s effect (3.04%), and diastolic blood pressure mediated the effects of ACP1 (acid phosphatase 1) (2.74%) and LAMC1 (laminin subunit gamma 1; 4.94%). CONCLUSIONS: The findings outline protein- and blood pressure-centered mechanisms in small vessel disease, highlighting proteins and antihypertensive targets as biomarkers and therapeutic entry points for precision intervention.