1
|
Yang J, Dong X, Wang F, Jin S, Zhang B, Zhang H, Pan W, Gan M, Duan S, Zhang L, Hu H, Ji W. A deep learning model based on MRI for prediction of vessels encapsulating tumour clusters and prognosis in hepatocellular carcinoma. Abdom Radiol (NY) 2024; 49:1074-1083. [PMID: 38175256 DOI: 10.1007/s00261-023-04141-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/19/2023] [Revised: 11/14/2023] [Accepted: 11/22/2023] [Indexed: 01/05/2024]
Abstract
PURPOSE This study aimed to build and evaluate a deep learning (DL) model to predict vessels encapsulating tumor clusters (VETC) and prognosis preoperatively in patients with hepatocellular carcinoma (HCC). METHODS 320 pathologically confirmed HCC patients (58 women and 262 men) from two hospitals were included in this retrospective study. Institution 1 (n = 219) and Institution 2 (n = 101) served as the training and external test cohorts, respectively. Tumors were evaluated three-dimensionally and regions of interest were segmented manually in the arterial, portal venous, and delayed phases (AP, PP, and DP). Three ResNet-34 DL models were developed, consisting of three models based on a single sequence. The fusion model was developed by inputting the prediction probability of the output from the three single-sequence models into logistic regression. The area under the receiver operating characteristic curve (AUC) was used to compare performance, and the Delong test was used to compare AUCs. Early recurrence (ER) was defined as recurrence within two years of surgery and early recurrence-free survival (ERFS) rate was evaluated by Kaplan-Meier survival analysis. RESULTS Among the 320 HCC patients, 227 were VETC- and 93 were VETC+ . In the external test cohort, the fusion model showed an AUC of 0.772, a sensitivity of 0.80, and a specificity of 0.61. The fusion model-based prediction of VETC high-risk and low-risk categories exhibits a significant difference in ERFS rates, akin to the outcomes observed in VETC + and VETC- confirmed through pathological analyses (p < 0.05). CONCLUSIONS A DL framework based on ResNet-34 has demonstrated potential in facilitating non-invasive prediction of VETC as well as patient prognosis.
Collapse
Affiliation(s)
- Jiawen Yang
- Department of Radiology, Taizhou Hospital of Zhejiang Province Affiliated to Wenzhou Medical University, 150 Ximen St, Linhai, Taizhou, 317000, Zhejiang, China
| | - Xue Dong
- Department of Radiology, Beijing Friendship Hospital, Capital Medical University, Beijing, China
| | - Fang Wang
- Department of Radiology, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, 310016, People's Republic of China
| | - Shengze Jin
- Department of Radiology, Taizhou Hospital of Zhejiang Province, Shaoxing University, Taizhou, 318000, Zhejiang, China
| | - Binhao Zhang
- Department of Radiology, Taizhou Hospital of Zhejiang Province Affiliated to Wenzhou Medical University, 150 Ximen St, Linhai, Taizhou, 317000, Zhejiang, China
| | - Huangqi Zhang
- Department of Radiology, Taizhou Hospital of Zhejiang Province Affiliated to Wenzhou Medical University, 150 Ximen St, Linhai, Taizhou, 317000, Zhejiang, China
| | - Wenting Pan
- Department of Radiology, Taizhou Hospital of Zhejiang Province Affiliated to Wenzhou Medical University, 150 Ximen St, Linhai, Taizhou, 317000, Zhejiang, China
| | - Meifu Gan
- Department of Pathology, Taizhou Hospital of Zhejiang Province Affiliated to Wenzhou Medical University, Taizhou, 317000, Zhejiang, China
| | - Shaofeng Duan
- GE Healthcare, Precision Health Institution, Pudong New Town, No.1, Huatuo Road, Shanghai, 210000, China
| | - Limin Zhang
- Department of Radiology, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China
| | - Hongjie Hu
- Department of Radiology, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, 310016, People's Republic of China
| | - Wenbin Ji
- Department of Radiology, Taizhou Hospital of Zhejiang Province Affiliated to Wenzhou Medical University, 150 Ximen St, Linhai, Taizhou, 317000, Zhejiang, China.
- Department of Radiology, Taizhou Hospital, Zhejiang University, Taizhou, 318000, Zhejiang, China.
| |
Collapse
|
2
|
Qu Q, Liu Z, Lu M, Xu L, Zhang J, Liu M, Jiang J, Gu C, Ma Q, Huang A, Zhang X, Zhang T. Preoperative Gadoxetic Acid-Enhanced MRI Features for Evaluation of Vessels Encapsulating Tumor Clusters and Microvascular Invasion in Hepatocellular Carcinoma: Creating Nomograms for Risk Assessment. J Magn Reson Imaging 2023. [PMID: 38116997 DOI: 10.1002/jmri.29187] [Citation(s) in RCA: 1] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/14/2023] [Revised: 12/01/2023] [Accepted: 12/02/2023] [Indexed: 12/21/2023] Open
Abstract
BACKGROUND Vessels encapsulating tumor cluster (VETC) and microvascular invasion (MVI) have a synergistic effect on prognosis assessment and treatment selection of hepatocellular carcinoma (HCC). Preoperative noninvasive evaluation of VETC and MVI is important. PURPOSE To explore the diagnosis value of preoperative gadoxetic acid (GA)-enhanced magnetic resonance imaging (MRI) features for MVI, VETC, and recurrence-free survival (RFS) in HCC. STUDY TYPE Retrospective. POPULATION 240 post-surgery patients with 274 pathologically confirmed HCC (allocated to training and validation cohorts with a 7:3 ratio) and available tumor marker data from August 2014 to December 2021. FIELD STRENGTH/SEQUENCE 3-T, T1-, T2-, diffusion-weighted imaging, in/out-phase imaging, and dynamic contrast-enhanced imaging. ASSESSMENT Three radiologists subjectively reviewed preoperative MRI, evaluated clinical and conventional imaging features associated with MVI+, VETC+, and MVI+/VETC+ HCC. Regression-based nomograms were developed for HCC in the training cohort. Based on the nomograms, the RFS prognostic stratification system was further. Follow-up occurred every 3-6 months. STATISTICAL TESTS Chi-squared test or Fisher's exact test, Mann-Whitney U-test or t-test, least absolute shrinkage and selection operator-penalized, multivariable logistic regression analyses, receiver operating characteristic analysis, Harrell's concordance index (C-index), Kaplan-Meier plots. Significance level: P < 0.05. RESULTS In the training group, 44 patients with MVI+ and 74 patients with VETC+ were histologically confirmed. Three nomograms showed good performance in the training (C-indices: MVI+ vs. VETC+ vs. MVI+/VETC+, 0.892 vs. 0.848 vs. 0.910) and validation (C-indices: MVI+ vs. VETC+ vs. MVI+/VETC+, 0.839 vs. 0.810 vs. 0.855) cohorts. The median follow-up duration for the training cohort was 43.6 (95% CI, 35.0-52.2) months and 25.8 (95% CI, 16.1-35.6) months for the validation cohort. Patients with either pathologically confirmed or nomogram-estimated MVI, VETC, and MVI+/VETC+ suffered higher risk of recurrence. DATA CONCLUSION GA-enhanced MRI and clinical variables might assist in preoperative estimation of MVI, VETC, and MVI+/VETC+ in HCC. EVIDENCE LEVEL 4 TECHNICAL EFFICACY: Stage 2.
Collapse
Affiliation(s)
- Qi Qu
- Nantong University, Nantong, Jiangsu, China
- Department of Radiology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| | - Zixin Liu
- Nantong University, Nantong, Jiangsu, China
- Department of Radiology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| | - Mengtian Lu
- Nantong University, Nantong, Jiangsu, China
- Department of Radiology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| | - Lei Xu
- Department of Radiology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| | - Jiyun Zhang
- Department of Radiology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| | - Maotong Liu
- Department of Radiology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| | - Jifeng Jiang
- Department of Radiology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| | - Chunyan Gu
- Department of Pathology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| | - Qinrong Ma
- Department of Pathology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| | - Aina Huang
- Department of Radiology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| | - Xueqin Zhang
- Department of Radiology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| | - Tao Zhang
- Department of Radiology, Affiliated Nantong Hospital 3 of Nantong University, Nantong Third People's Hospital, Nantong, Jiangsu, China
| |
Collapse
|