Genji Bai MM

Genji Bai MM


Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
10 February 2024 in European Radiology
Verified
Journal Article
Closed access
PDF file
QWQian Wang
Qian Wang
YLYingyu Lin
Yingyu Lin
CDCong Ding
Cong Ding
WGWenting Guan
Wenting Guan
XZXiaoling Zhang
Xiaoling Zhang
GMGenji Bai MM
Genji Bai MM

3 May 2023 in Journal of Magnetic Resonance Imaging
Verified
Journal Article
Closed access
MWMingxiang Wei
Mingxiang Wei
YZYu Zhang
Yu Zhang
CDCong Ding
Cong Ding
JJJianye Jia
Jianye Jia
HXHaimin Xu
Haimin Xu
HWHong Wang
Hong Wang

23 May 2019 in Journal of Applied Clinical Medical Physics
Verified
Journal Article
Open access
PDF file
WTWeijing Tao
Weijing Tao
HZHui‐Xin Zhang
Hui‐Xin Zhang
LZLian‐Mei Zhang
Lian‐Mei Zhang
FGFeng Gao
Feng Gao
WHWei Huang
Wei Huang
GMGenji Bai MM
Genji Bai MM
Purpose: Ductal carcinoma in situ (DCIS) is a precursor of invasive ductal breast carcinoma (IDC). This study aimed to use pharamcokinetic dynamic contrast‐enhanced magnetic resonance imaging (DCE‐MRI) and intravoxel incoherent motion diffusion‐weighted imaging (IVIM‐DWI) for the early diagnosis of DCIS.Methods: Forty‐seven patients, including 25 with DCIS (age: 28–70 yr, mean age: 48.7 yr) and 22 with benign disease (age: 25–67 yr, mean age: 43.1 yr) confirmed by pathology, underwent pharamcokinetic DCE‐MRI and IVIM‐DWI in this study. The quantitative parameters Ktrans, Kep, Ve, Vp, and D, f, D* were obtained by processing of DCE‐MRI and IVIM‐DWI images with Omni‐Kinetics and MITK‐Diffusion softwares, respectively. Parameters were analyzed statistically using GraphPad Prism and MedCalc softwares.Results: All low‐grade DCIS lesions demonstrated mass enhancement with clear boundaries, while most middle‐grade and high‐grade DCIS lesions showed non‐mass‐like enhancement (NMLE). DCIS lesions were significantly different from benign lesions in terms of Ktrans, Kep, and D (t = 5.959, P < 0.0001; t = 5.679, P < 0.0001; and t = 5.629, P < 0.0001, respectively). The AUC of Ktrans, Kep, D and the combined indicator of Ktrans, Kep, and D were 0.936, 0.902, 0.860, and 0.976, respectively. There was a significant difference in diagnostic efficacy only between D and the combined indicator (Z = 2.408, P = 0.016).Conclusion: DCE‐MRI and IVIM‐DWI could make for the early diagnosis of DCIS, and reduce the misdiagnosis of DCIS and over‐treatment of benign lesions.

25 March 2024 in Heliyon
Verified
Journal Article
Open access
ZLZiyan Liu
Ziyan Liu
JJJianye Jia
Jianye Jia
FBFan Bai
Fan Bai
YDYuxin Ding
Yuxin Ding
LHLei Han
Lei Han
GMGenji Bai MM
Genji Bai MM
To investigate the effectiveness of a multimodal deep learning model in predicting tumor budding (TB) grading in rectal cancer (RC) patients. A retrospective analysis was conducted on 355 patients with rectal adenocarcinoma from two different hospitals. Among them, 289 patients from our institution were randomly divided into an internal training cohort (n = 202) and an internal validation cohort (n = 87) in a 7:3 ratio, while an additional 66 patients from another hospital constituted an external validation cohort. Various deep learning models were constructed and compared for their performance using T1CE and CT-enhanced images, and the optimal models were selected for the creation of a multimodal fusion model. Based on single and multiple factor logistic regression, clinical N staging and fecal occult blood were identified as independent risk factors and used to construct the clinical model. A decision-level fusion was employed to integrate these two models to create an ensemble model. The predictive performance of each model was evaluated using the area under the curve (AUC), DeLong's test, calibration curve, and decision curve analysis (DCA). Model visualization Gradient-weighted Class Activation Mapping (Grad-CAM) was performed for model interpretation. The multimodal fusion model demonstrated superior performance compared to single-modal models, with AUC values of 0.869 (95% CI: 0.761–0.976) for the internal validation cohort and 0.848 (95% CI: 0.721–0.975) for the external validation cohort. N-stage and fecal occult blood were identified as clinically independent risk factors through single and multivariable logistic regression analysis. The final ensemble model exhibited the best performance, with AUC values of 0.898 (95% CI: 0.820–0.975) for the internal validation cohort and 0.868 (95% CI: 0.768–0.968) for the external validation cohort. Multimodal deep learning models can effectively and non-invasively provide individualized predictions for TB grading in RC patients, offering valuable guidance for treatment selection and prognosis assessment.

9 January 2024 in Academic Radiology
Verified
Journal Article
Closed access
JJJianye Jia
Jianye Jia
ZLZiyan Liu
Ziyan Liu
FWFen Wang
Fen Wang
GMGenji Bai MM
Genji Bai MM

2 August 2020 in Acta Radiologica
Verified
Journal Article
Closed access
PDF file
MWMingxiang Wei
Mingxiang Wei
FBFan Bo
Fan Bo
HCHui Cao
Hui Cao
WZWei Zhou
Wei Zhou
WSWenli Shan
Wenli Shan
GMGenji Bai MM
Genji Bai MM
Background: Accurate preoperative diagnosis of malignant ovarian tumors (MOTs) is particularly important for selecting the optimal treatment strategy and avoiding overtreatment. Purpose: To evaluate the diagnostic efficacy of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for MOTs. Material and Methods: A systematic search was performed in PubMed, Embase, the Cochrane Library, and Web of Science databases to find relevant original articles up to October 2019. The included studies were assessed using the Quality Assessment of Diagnostic Accuracy Studies-2 tool. Studies on the diagnosis of MOTs with quantitative or semi-quantitative DCE-MRI were analyzed separately. The bivariate random-effects model was used to assess the diagnostic authenticity. Meta-regression analyses were performed to analyze the potential heterogeneity. Results: For semi-quantitative DCE-MRI, the pooled sensitivity, specificity, positive likelihood ratio (LR), negative LR, diagnostic odds ratio (DOR), and the area under the summary receiver operating characteristic curves (AUC) were 85% (95% confidence interval [CI] 0.75–0.92), 85% (95% CI 0.77–0.91), 5.8 (95% CI 3.8–8.8), 0.17 (95% CI 0.10–0.30), 33 (95% CI 18–61), and 0.92 (95% CI 0.89–0.94), respectively. For quantitative DCE-MRI, the pooled sensitivity, specificity, positive LR, negative LR, DOR, and AUC were 88% (95% CI 0.65–0.96), 93% (95% CI 0.78–0.98), 12.3 (95% CI 3.4–43.9), 0.13 (95% CI 0.04–0.45), 91 (95% CI 10–857), and 0.96 (95% CI 0.94–0.98), respectively. Conclusion: DCE-MRI has great diagnostic value for MOTs. Semi-quantitative DCE-MRI may be a relatively mature approach; however, quantitative DCE-MRI appears to be more promising than semi-quantitative DCE-MRI.